Home

Vampir 8 User Manual

1. 56 60 Sd RO Ce S x RR X 453 x34 2353553434554 61 5 1 1 61 tee ee 63 53 VO Filter aa eee ea Poe Ge t 64 Cee eee eee eee eee eee eee ee 65 5 41 5 66 69 6 Comparison of Trace Files 79 6 1 Starting and Saving a Comparison Session 80 ee RN PM RT 82 AE ROS 84 6 4 Usage of Predefined 86 88 1 1 General Preferencesi a a 88 7 2 89 90 92 8 1 Introduction eer 92 8 2 Identified Problems and Solutions 93 8 2 1 Computational Imbalance 93 8 2 2 SerialOptimization 95 8 2 3 High Cache Miss Ratej lll lll 96 8 3 98 maana BHHHHH maaa mannana BHBHHHHH DHUBBGHHHH 000006 DBDBBHH manono DBHHBH 00000 0000 BBHHBH 0000000000000000000 CHAPTER 1 1 Introduction Performance optimization is a key issue for the development of efficient parallel soft wa
2. BBBHEUBE 0000000 DUBBHHHH mBHEBBHHH 000006 DBDBBHH mmEnnu DBHHBH 00000 0000 BHHBH CHAPTER 1 INTRODUCTION 0 events status information and event summaries separately A single global master file holds the necessary information for the process to stream mappings The master file is always named name otf 2 Note Open the master file otij2 to load a trace When copying moving or deleting traces it is important to include all files with the same name prefix If not Vampir will render the whole trace invalid Good practice is to hold all files belonging to one trace in a dedicated directory Detailed information can be found in the Open Trace Format documentation for and OTF 1 3 Vampir and Windows HPC Server 2008 The Vampir performance visualization tool usually consists of a performance moni tor e g Score P see Section 2 2 1 or VampirTrace see Section that records performance data and a performance GUI which is responsible for the graphical rep resentation of the data In Windows HPC Server 2008 the performance monitor is fully integrated into the operating system which simplifies its employment and provides ac cess to a wide range of system metrics A simple execution flag controls the generation of performance data This is very convenient and an important difference to solutions based on explicit source
3. maana BHHHHH maaa mannana BHHHHHUHH EHHEBHHEH BHBHHHHH BHEBBHHH 000006 900008 manono BDBHHBH manom 0000 BHHBH CHAPTER 4 PERFORMANCE DATA VISUALIZATION NAME Symbol Description Message Burst Due to a lack of pixels it is not possible to display a large amount of messages in a very short time interval Therefore outgoing messages are summarized as so called message bursts In this representation you cannot deter mine which processes receive these messages Zooming into this interval reveals the corresponding single messages Markers To indicate particular points of interest during the run F multiple time of an application like errors or warnings markers can single be placed in a trace file They are drawn as triangles which are colored according to their types To indicate that two more markers are located at the same pixel a tricolored triangle is drawn Events Vampir shows detailed information about operations if they are included in the trace file events are depicted as mV triangles at the beginning of an interval In order to see the whole interval of a single I O event its triangle has to be bes selected In that case a second triangle indicating the end of the interval appears Multiple I O events
4. 55 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 SOLVE Process 14 Process 15 Opacity Ef tx 50 100M 150 M 200 M Figure 4 35 Using the opacity slider to investigate individual invocations of SOLVE_EM 59 5 Information Filtering and Reduction Due to the large amount of information that can be stored in trace files it is usually nec essary to reduce the displayed information according to some filter criteria In Vampir there are different ways of filtering It is possible to limit the displayed information to a certain choice of processes or to specific types of communication events e g to cer tain types of messages or collective operations Deselecting an item in a filter means that this item is fully masked In Vampir filters are global Therefore masked items will no longer show up in any chart Filtering not only affects all performance charts but also the Zoom Toolbar All filter can be reached via the Filter entry in the main menu The available filter and their respective filter criteria are summarized in Table 5 1 Filtered Object Filter Criteria Processes Process Groups Communicators Process Hierarchy Representative Processes Messages Message Communicators Message Tags Name Call Level Call Path Duration Number of Invocations Functions Communicators Collective
5. EHEBHHHEG BHBHHHHH maHHEHHH KUSHI DHUBBGHHHH BHEBBHHH 000006 DBDBBHH mmEnnu 00000 0000 BHHBH 0000 Call Path does not contain WRF_INPUTIN This example demonstrates the opposite behavior of the previous example In call paths that contain the function WRF_INPUTIN only functions that lead to WRF_INPUTIN are shown The function WRF_INPUTIN itself and their directly or indirectly called sub functions are filtered Other call paths remain unaffected by the filter and are still shown Filter Functions Show only functions that match any of the following conditions Description Filter T Does not contain WRF_INPUTIN Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help tithes S BH Timeline Ax 6s 7s 8s 9s 10s lls 12s 135 145 Process 0 INPUT WRF Process 1 INPUT_WRF Process 2 INPUT_WRF Process 3 INPUT_WRF Process 4 INPUT_WRF Process 5 INPUT_WRF Process 6 INPUT_WRF Process 7 INPUT_WRF Process 1 MED INITIALDATA INPUT INPUT_MODEL_INPUT o OU PWN m 11 355 Figure 5 14 Call path filter which does not contain WRF_INPUTIN 77 GWT 54 FUNCTION FILTER Showing only Functions until a certain Call Level This example demonstrates th
6. 5 eh eh e e e e n n n 6 1 3 Vampir and Windows HPC Server 2008 7 2 Getting Started 8 2 1 Installation of Vampir 8 2 1 1 Linux UNnIXI 8 8 2 1 3 Windows 9 2 2 Generation of Performance Datal 9 ee ee ee ee ee YN 9 eae 9 3 5 39 WX S EAE eee oe 11 2 2 3 Event Tracing for Windows 11 2 3 Starting Vampir and Loading Performance 13 2 3 1 Loading a lrace 14 2 3 2 Loading a File Subset 15 17 3 1 Chart a 18 20 22 E 24 25 3 6 Properties the Trace File 26 3 7 Command Line Parameters 0 0 00 eee eae 26 27 41 Timeline Charts 2 2 0 00 00 a ee 27 41 1 Master Timeline and Process Timeline 27 31 r E N E 32 35 42 4 2 1 Function Summary 1 llle 42 _ E S 44 E Contents EP E 45 46 4 2 5 47 Segoe 48 4 3 Informational Charts 49 4 3 1 Function Legend 1 49 50 51 13 4 3 9 9 4 53 441 Metric Editori 22 lll 54 442
7. The chart as shown in Figure 4 21 is figured as a table Its rows represent the sending processes whereas the columns represent the receivers The color legend on the right indicates the displayed values lt adapts automatically to the currently shown value range It is possible to change the type of displayed values Different metrics like the aver age duration of messages passed from sender to recipient or minimum and maximum bandwidth are offered To change the type of value that is displayed use the context menu option Set Metric Use the Process Filter to define which processes groups should be displayed see Section 5 1 Like in the Master Timeline the context menu entries Expand All and Collapse All hide and expose subordinated information of processes e g threads or CUDA streams The context menu functionality Group Peers by System aggregates matrix entries according to their position in the system tree Please note that the system tree is only available in otf2 traces Using this functionality communication between nodes or on the machine level can be analyzed Note A high duration is not automatically caused by a slow communication path be tween two processes but can also be due to the fact that the time between starting transmission and successful reception of the message can be increased by a recipient that delays reception for some reason This will cause the duration to increase by this delay and the me
8. 000006 900006 _ BDBHHBH 00000 0000 BHHBH 8 AUSE CASE computational work if MICROPHYSICS calls would have the same duration An other hint at this overhead in synchronization is the fact that the MPI receive routine uses 17 6 of the time of one iteration Function Summary in Figure 8 2 Solution To even out this asymmetry the code which determines the size of the work packages for each process had to be changed To achieve the desired effect an improved ver sion of the domain decomposition has been implemented Figure 8 3 shows that all occurrences of the MICROPHYSICS routine are vertically aligned thus balanced Ad ditionally the MPI receive routine calls are now clearly smaller than before Comparing the Function Summary of Figure 8 2 and Figure 8 3 shows that the relative time spent in MPI receive has been decreased and in turn the time spent inside MICROPHYSICS has been increased greatly This means that we now spend more time computing and less time communicating which is exactly what we want 8 2 2 Serial Optimization Inlining of frequently called functions and elimination of invariant calculations inside loops are two ways to improve the serial performance This section shows how to detect candidate functions for serial optimization and suggests measures to speed them up Problem All performance charts in Vampir show info
9. 15 219 170 01 20 0 24 00 00 00 gt 7 00 O E snor 1 341 wrfbdy_d01 342 lt STDERR gt 17 work home0 ml namelist input 10 work home0 un RRTM_DATA 1 work home0 LANDUSE TBL Figure 4 22 Summary additional bar starts at the minimum and ends at the maximum value of the metric see Figure The operations be grouped by the characteristics Transaction Size File Name and Operation Type The group base can be changed via the context menu entry Group Operations by In order to select the I O operation types that should be considered for the statistic calculation use the Set Operations sub menu of the context menu Available op tions are Read Write Read Write and Apply Global I O Operations Filter The latter includes all selected operation types from the O Events filter dialog see Section 5 3 4 2 6 Call Tree The Call Tree depicted in Figure 4 23 illustrates the invocation hierarchy of all mon itored functions in a tree representation The display reveals information about the number of invocations of a given function the time spent in the different calls and the caller callee relationship The entries of the Call Tree can be sorted in various ways Simply click on one header of the tree representation to use its characteristic to re sort the Call Tree Please note that not all available characteristics are enabled by default To add or remove charact
10. Show time as Seconds Automatically open context view Use color gradient in charts Font Sans Serif Select Restore Default Source code Location of source files Browse Remove prefix from source reference Do not open source files bigger than 100 KiB Appearance Analysis Fix number of analysis threads gt E u Enable support for color blindness Automatically check for newer versions Saving Enable presentation mode Document layout Enable multiple document interface Apply Q cancel Figure 7 1 General preferences mode or enable the presentation mode With the presentation mode active the mouse pointer is shown using a larger mouse icon that also animates mouse button clicks On Linux systems there is also the Document layout option available If this option is enabled all open Trace View windows need to stay in one main window If it is disabled the Trace View windows can be moved freely over the Desktop 7 2 Appearance The Appearance settings of the Preferences dialog allow to change the application s color options Available categories are functions function groups markers counters collectives messages and I O events To modify an entry click on its color icon color picker dialog will then allow to select the new color A change of the line width is also available for messages and collectives In order to quickly find a
11. 55 15 1325 Figure 4 31 MPI Wait invocations with longest duration Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help SHiSOSEKSS SAU Timeline X 28 925 s 29 000 s 29 075 29 150 29 225 Processo 0 CUMULLUS Metric Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 0 0 0 3 0 6 09 2 29 048 5 Figure 4 32 Using the opacity slider to reveal MPI_Wait invocations in the timeline together with the superimposed color coded duration 57 a a GWT rimon 44 CUSTOMIZABLE PERFORMANCE METRICS FLOPS of SOLVE EM Custom Metrics Description FLOPS of SOLVE_EM Unit 1 5 Metric Trace Counter PAPI_FP_OPS Increments per Second Operation Multiply Metric x i Function is Active SOLVE EM cancel Figure 4 33 Custom metric showing FLOPS only for function SOLVE_EM Vampir also allows to search for invocations of individual functions below or above a certain threshold In this example invocations of the function SOLVE_EM with a FLOP rate above 150 M are searched Therefore the first step is to construct a custom metric showing the FLOP rate only for the function SOLVE_EM The process of constructing
12. Processes Accumulated Exclusive Time per Function Group 15 ms 10 ms 5ms 0 ms processo 12 5 1 von 2 Application 777 CALCULATION cue e 1 745 ms TEST 1 2 3 Figure 6 5 Comparison View with open charts Figure 6 5 depicts a Comparison View with open Master Timeline Process Timeline and Function Summary charts 82 maana BHHHHH mannana EHEBHHHEH BHHBHHHHH maHHEHHH BBBHEUSE DUBBGHHHH BHEBBHHH 000006 900008 _ DBHHBH mHEBH 0000 BHHBH All available charts work the same way in the Trace View Due to the fact that the Comparison View couples the zoom of all trace files the charts can be used to directly compare performance characteristics between the traces Comparison View Edit Chart Filter Window Help K h M_M M 2 8 0 0 A calcTest otf 8 6015 C B calcTest otf _ C calcTest otf Db 274 ms Timeline Function Summary 6 5 ms 7 0 ms 7 5 ms 8 0 ms 8 5 ms All Processes Accumulated Exclusive Time per Funct 2 ms lms 0 ms 2 O r Process 1 Process 2 Process 3 Process 0 Process 1 Process 2 mm Summary All Processes Accumulated Exclusive Time per Funct Process 3 5 ms 0 ms Process 0 Process 1 Process 2 Proc
13. 000006 DBDBBHH _ ooo 00000 0000 BHHBH Vampir Trace View Vampir Comparison Marker2 potential vt otf B Figure 4 25 A chosen marker A and its representation in the Marker View B 4 3 3 Context View As implied by its name the Context View provides detailed information of a selected object additional to its graphical representation An object e g a function function group message or message burst can be selected directly in a chart by clicking its graphical representation For different types of objects different context information is provided in the Context View For example the object specific information for functions includes properties like nterval Begin Interval End and Duration shown in Figure 4 26 Objects may provide additional information for some items In that case such items are displayed as links A click double click on OS X systems on the link opens a new tab containing the additional information The Context View may contain several tabs A new empty tab can be added by clicking the on the right hand side Information of new selected objects are always displayed in the currently active tab The Context View offers a mode for the comparison of information between tabs The button on the left hand side allows to choose two objects for comparison It is pos sible to compare different objects from different c
14. 2000 Number of Invocations gt Is less than gt 15000 Vampir Trace View Vampir WRF wrf otf Edit Chart Filter Window Help loje gt 1 39 milite 39 042 Timeline Function Summary Os 10s 20s 305 All Processes Number of invocations per Function all 10k Ok Processo EEEEB HEEE Process 1 gt 12876 T Process2 po 10 850 f Process 3 OO ve Bcast Process 4 write Process 5 je 3 840 MM MODULE BC SPEC_BDYTEND Process 6 3 584 DEBUG 10 VVRF Process 7 i 3 584 Ml USE _INPUT_SERVERS Process 8 3 584 MODULE IO R ST OPERATION Process9 3 472 El CALL PKG AND DIST REAL Process 10 gt 2 880 MODULE_ADVE VECT_SCALAR Process 11 2 400 Bill MODULE SMALL CALC RHO Process 12 gt 2 160 MODULE EM RK SCALAR TEND Process 13 gt 2 160 MODULE EM R PDATE SCALAR Process 14 Process 1 1 gt NENE 2 r Figure 5 11 Show functions inside specified range 74 maana HHHHH mannana n EHEBHHEH BHBHHHHH DHUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo manom 0000 BHHBH 0000 This example demonstrates the opposite behavior of the previous example Here functions whose
15. 61 4 e start end option defines a numeric range from value start to value end The argument option including the separating colon is facultative and described below Please note that the arguments start and end can be left blank Blanks denote oo and respectively e numberil number2 number3 option defines an arbitrary set of numbers Options The option parameter introduced above can be used to define e a parity for the given range or set The characters e or o match even or odd numbers respectively e Stride for the given range or set The sequence sn matches every nth number in the corresponding range or set If no start value is given zero is used as reference point It is possible to use options without a range or set specification e g e for all even numbers Examples The following examples are based on regular expressions Use Hegular Expressions needs to be checked via the magnifier icon e Process 1 5 matches process 1 to 5 e Process 0 matches odd processes e Process s10 matches every tenth process Zero is used as reference point e Process 012 matches every hundredth process e Process matches all process labels containing Process e Process 2 matches exactly Process 2 Labels like Process 2 or cess 20 will not be matched 62 BHHHM maana BHHHHH maaa mannana
16. 5 4 1 Filter Options This chapter explains the various options available to build up filter rules Filtering Functions by Name One way of filtering functions is by their name This filter mode provides two different options Name provides a text field for an input string Depending on the options all functions whose names match the input string are shown The matching is not case sensitive Available options 66 Contains The given input string must occur in the function name Does not contain The given input string must not occur in the function name Is equal to The given input string must be the same as the function name Is not equal to The given input string must not be the same as the function name Begins with The function name must start with the given input string Ends with The function name must end with the given input string mamaa maana HHHHH maaa mannana BHHHHUEH EHEBHHHEH BHBHHHHH KUSHI DUBBGHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH List of Names provides a dialog that allows to directly select the desired set of func tions and function groups Available options e Contains The selected functions are shown e Does not contain The selected functions are filtered Filtering Functions by Duration Functions can als
17. EHEBHHEG BZHBHHHHH BBBHEUHE DUBBGHHHH BHEBBHHH DBDHHEBHH 000006 DBDBBHH manono 00000 0000 BHHBH 0000000000000000000 5 2 Message and Collective Operations Filter Filter Messages Message Communicators Include Exclude All Communicator 0 Message Tags Tags 0 15 Example 1 5 3 10 show 1 2 4 5 and 10 Oca Figure 5 2 Message Filter Figure shows a Message Filter dialog This dialog allows to filter messages from the displayed trace data Available options are to select deselect messages based on their Message Tag or Message Communicator The default is to show all messages The Collectives Filter is designed accordingly It allows to filter collective operations from the displayed trace data The collectives can be filtered by their Communicator or their Collective Operation type 63 53 VO FILTER 5 3 VO Filter Filter Events Groups File Names Operation Types Include Exclude All Include Exclude All E Include Exclude All fileio Y dev sgi_fetchop Async stdio dev zero Close Y work hom NDUSE TBL vi Coll vi work hom RRTM_DATA Y Direct work homeo elist input Y Dup Y lt STDERR gt v Failed Y lt STDIN gt IsReadLock lt STDOUT gt Lock rsl error 0000 Open rsl error 0001 v Other rsl error 0002
18. The Comparison View provides two additional ways of navigating with markers If two markers of one trace are selected in the Marker View the button Zoom Between Marker sets the trace zoom to the according timestamps of the markers If two markers of dif ferent traces are selected the button A ign Traces at Marker adjusts the time offset between the respective traces The selected markers are shown next to each other in the timeline charts and consequently both traces are aligned at the respective mark ers 87 7 Customization The appearance of the trace file and various other application settings can be altered in the preferences accessible the main menu entry File Preferences Settings concerning the trace file itself e g layout or function group colors are saved individually next to the trace file in a file with the ending vsettings This way it is possible to adjust the colors for individual trace files without interfering with others The options mport Preferences and Export Preferences provide the loading and sav ing of preferences of arbitrary trace files 7 1 General Preferences The General preferences allow to change application and trace specific values Show time as decides whether the time format for the trace analysis is based on sec onds or ticks With the Automatically open context view option disabled Vampir does not open the context view after the selection of an item like a message or function Us
19. in order to effectively compare multiple trace files their zoom 15 coupled and synchronized For the comparison of areas of interest the displayed trace regions are freely shiftable in time This allows for arbitrary alignments of the trace files and thus enables comparison of user selected areas in the trace data Comparison View Edit Chart Filter Window Help K h K K dhdis h n E 5 97 V A calcTest otf 1761965827619 551212 C B calcTest otf L C calcTest otf Timeline 16 96585 5 16 96590 5 16 96595 5 16 96600 5 16 96605 5 16 96610 5 55 0 _ Process 1 55 2 Process 3 Process 0 55 1 Process 2 Process 3 Process 0 Process 1 Process 2 5 3 Figure 6 1 Comparison View The Comparison View window depicted in Figure provides all comparison fea tures This chapter introduces its usage with the help of screenshots For this purpose the comparison of three trace files is demonstrated step by step The example trace files show one test application performing ten iterations of simple calculations Each trace respectively represents the run of this application on a different machine 79 Amy 6 1 Starting and Saving a Comparison Session S Open New File Help VAMPIR Comparison Session Local File Remote File Figure 6 2 Vampir start window T
20. 14 Process 15 Figure 4 5 Active overlay showing PAPI FP OPS in the Master Timeline for finer coarser accumulation of values The displayed colors represent corresponding functions or function groups The con text menu entry Set Functions specifies the set of functions that is displayed in the chart The context menu entry Options Group Functions aggregates functions and displays them as function groups Shown functions or function groups can be sorted by name or by value via the context menu option Sort By The Set Metric sub menu of the context menu allows to switch between the available metrics Number of Invocations and Exclusive Time Using the Process Filter see Section allows to restrict this chart to a freely se lectable set of processes As a result only the consumed time of these processes 1 displayed for each function or function group Instead of using the filter which affects all other displays by hiding processes it is possible to select a single process via Set Process in the context menu This does not have any effect on other charts 4 1 3 Counter Data Timeline Counters are values collected over time to count certain events like floating point op erations or cache misses Counter values can be used to store not just hardware performance counters but arbitrary sample values There can be counters for different Statistical information as well for instance counting the number of function calls
21. 2 14 6 5 Figure 3 9 Zooming within Chart Chart mode of the Function Summary accessible via the context menu under Set Chart Mode Pie Chart To zoom into an area click and hold the left mouse button and select the area as shown in Figure 3 9 Itis possible to zoom horizontally and in some charts also vertically In the Master Timeline horizontal zooming defines the time interval to be visualized whereas vertical zooming selects a group of processes to be displayed scroll horizontally move the slider at the bottom or use the mouse wheel To get back to the initial state of zooming select Reset Horizontal Zoom or Reset Vertical Zoom see Section in the context menu of the respective performance chart Additionally the zoom can be accessed with help of the Zoom Toolbar by dragging the borders of the selection rectangle or by scrolling of the mouse wheel as described in Chapter In order to return to the previous zooming state an undo functionality accessible via the Edit menu is provided Alternatively the key combination Ctrl Z also reverts the last zoom Accordingly a reverted zooming action can be redone by selecting Redo in the Edit menu or by pressing Ctrl Shift Z The undo functionality is not bound to single performance charts but works across the entire application The labels of the Undo and Redo menu entries also state which kind of action will be undone redone next 23 ZI 3 4 The Zoom Toolbar Vampir pr
22. Definitions name x def pl Events name x events Statistics name x stats Snapshots name x snaps rommene Local Definitions Events Master Control name otf EEEE E a M MEC E E E E ee Local Definitions Events Global Definitions name 0 def a Statistics Snapshots Figure 1 1 Representation of Streams by Multiple Files The original OTF format uses a special ASCII data representation to encode its data items with numbers and tokens in hexadecimal code without special prefixes This allows for a very powerful format with respect to storage size human readability and search capabilities on timed event records In contrast to that its OTF2 successor relies on a binary representation of the data which simplifies and accelerates parsing In order to support fast and selective access to large amounts of performance trace data OTF is based on a stream model i e single separate units representing seg ments of the overall data OTF streams may contain multiple independent processes whereas a process belongs to a single stream exclusively As shown in Figure each stream is represented by multiple files which store definition records performance 5 BHH maana BHHHHH mana m mannana h EHHEBHHEH BHBHHHHH
23. View window will calculate their statistic information ac 24 BHH mamaa maana maaa mannana EHEBHHEH BHBHHHHH DHUBBGHHHH 000006 900008 _ manom 0000 CHAPTERS BASICS 1 cording to the selected time interval zooming state in the Zoom Toolbar The Zoom Toolbar can be enabled and disabled with the toolbars context menu entry Zoom Tool bar 3 5 The Charts Toolbar o LIE AB 5 B cud 42 LE I LE PN it Master Timeline Process Timeline Summary Timeline Counter Data Timeline Performance Radar Function Summary Message Summary Process Summary Communication Matrix View VO Summary Call Tree Function Legend Marker View Context View Description Section 4 1 1 Section 4 1 1 Section 4 1 2 Section 4 1 3 Section 4 1 4 Section 4 2 1 Section 4 2 3 Section 4 2 2 Section 4 2 4 Section 4 2 5 Section 4 2 6 Section 4 3 1 Section 4 3 2 Section 4 3 3 Table 3 1 Icons of the Charts Toolbar The Charts Toolbar is used to open instances of the available performance charts It is located in the upper left corner of the Trace View window as shown in Figure 3 1 The 25 4 A
24. a custom metric is described in more detail in Section 4 4 The constructed custom metric is depicted in Figure 4 33 Figure 4 34 shows the constructed metric in the overlay The color scale is set to highlight only functions above 150 M FLOPS When zooming into an area of interest the opacity slider can be used to reveal individual function invocations in the timeline Figure 4 35 58 CHAPTER 4 PERFORMANCE DATA VISUALIZATION Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help Timeline Ax Os 255 505 Zu 1005 125 5 1505 1755 2005 Metric FLOPS of SOLVE EM v Opacity 4 x Process 0 Process1 MANI 11 Process2 m Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 TNT THAT Process 14 M EN NINH HEN UN Process 15 OM 50 100M _ 150 200 M 4 n 136 Figure 4 34 SOLVE_EM invocations with highest FLOP rate Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help SEMMB OEM LE Timeline 131 7893 5 131 7894 5 131 7895 5 131 7896 5 131 7897 5 FLOPS of SOLVE EM Process 0 Process 1 Process 2 BSOLVE EM Process 3
25. are tricolored and ple drawn as a triangle with a line to the end of the interval Table 4 1 Additional Information in the Master and Process Timeline metric The counter data is displayed in a color coded fashion like in the Performance Radar Section The color scale can be freely customized by clicking on the wrench icon The control window also provides an opacity control slider This slider al lows to adjust the opacity of the overlay and thus makes the underlying functions easily visible without the need to disable the overlay mode 4 1 2 Summary Timeline The Summary Timeline chart Figure depicts the fractions of the number of pro cesses that are actively involved in given activities at a certain point in time This chart is useful for studying communication overhead and load imbalance issues from a high level perspective The information is shown as a vertical histogram The context menu entry Set Step Size alters the width represented duration of the histogram bars This allows to adjust 31 MA GWT uron 41 TIMELINE CHARTS Vampir Trace View Vampir Large wrf otf W File Edit Chart Filter Window Help Srusexs 05 25s 50s 75s 100s 1255 150 5 1755 2005 Metric PAPI_FP_OPS Opacity S x Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process
26. file will be equal to the application name For other systems the default name is a ot but can be defined manually by setting the environment variable VT_FILE_PREFIX to the desired name Detailed information about the installation and usage of VampirTrace can be found in the VampirTrace user manual 2 2 3 Event Tracing for Windows ETW The Event Tracing for Windows ETW infrastructure of the Windows client and server OS s provides a powerful software monitor Starting with Windows HPC Server 2008 MS MPI has built in support for this monitor It enables application developers to quickly produce traces in production environments by simply adding an extra mpiexec flag trace Trace files will be generated during the execution of your application The recorded trace log files include the following events Any MS MPI application call and low level communication within sockets shared memory and NetworkDirect im plementations Each event includes a high precision CPU clock timer for precise visu alization and analysis http www tu dresden de zih vampirtrace 11 Amy Amy The steps necessary for monitoring the MPI performance of an MS MPI application are depicted in Figure First the application needs to be available throughout all compute nodes in the cluster and has to be started with tracing enabled The Event Tracing for Windows ETW infrastructure writes event logs et files containing the respective MPI events of the appl
27. maaa 1 mannana EHEBHHEH BHBHHHHH BBBHEUSE 0000000 DUBBGHHHH BHEBBHHH 000006 DBDBBHH _ DBHHBH mamom 0000 BHHBH 8 AUSE CASE Vampir Trace View Vampir SuccessStory pmp old otf Sle Edit Chart Filter Window Help gt ____ mei ___ _________ ______ Figure 8 4 Before Tuning Counter Data Timeline revealing a high amount of L2 cache misses inside the CLIPPING routine light blue Vampir Trace View Vampir SuccessStory pmp tuned otf Sle Edit Chart Filter Window Help Figure 8 5 After Tuning Visible improvement of the cache usage 97 a 83 CONCLUSION 8 3 Conclusion By using the Vampir toolkit three problems have been identified As a consequence of addressing each problem the duration of one iteration has been decreased from 3 5 seconds to 2 0 seconds 300 Vampir Trace View Vampir SuccessStory pmp tuned otf YW Edit Chart filter Window Help it om 8 sS vuboSsSi SB m Timeline Function Summa 1 0s 50 5 100 5 150 5 209 s All Processes Accumulated Exclusive Time 40 0 30 0 200 100 0 0 EEE METEO 7 25 MP UTIL iL 07 B Application 0 46 VT API 0 27 COUPLE Process 0 Process 7 Process
28. mand line interface see Section 2 3 2 1 2 Mac OS X Open the dmg installation package and drag the Vampir icon into the applications folder on your computer You might need administrator rights to do so Alternatively you can also drag the Vampir application to another directory that is writable for you After that double click on the Vampir application and follow the instructions for license installation 5 8BHH 5 DDDB maana BHHHHH maaa mannana BHHHHHUHH EHHEBHHEH BHBHHHHH BHEBBHHH 000006 900008 manono BDBHHBH manom 0000 BHHBH 2 GETTING STARTED m 2 1 3 Windows On Windows platforms the provided Vampir installer makes the installation very simple and straightforward Just run the installer and follow the installation wizard Install Vampir in a folder of your choice e g C Program Files In order to run the installer in silent unattended mode use the option It is also possible to specify the output folder of the installation with D dir An example of a silent installation command is as follows Vampir 8 4 0 Standard x86 setup exe S D C Program Files You also have the option to associate Vampir with OTF and OTF2 files otf otf2 during the installation proces
29. microphysics function group MP is done here as well The second half is the iteration part where the actual weather forecasting takes place In a normal weather simulation this part would be much larger But in order to keep the recorded trace data and the overhead introduced by tracing as small as possible only a few iterations have been recorded This is sufficient since they are all doing the same work anyway Therefore the simulation has been configured to only forecast the weather 20 seconds into the future The iteration part consists of two large iterations Figurej8 1 B and C each calculating 10 seconds of forecast Each of these in turn is partitioned into several smaller iterations For our observations we focus on only two of these small inner iterations since this is the part of the program where most of the time is spent The initialization work does not increase with a higher forecast duration and would only take a relatively small amount of time in a real world run The constant part at the beginning of each large iteration takes less than a tenth of the whole iteration time Therefore by far the most time is spent in the small iterations Thus they are the most promising candidates for optimization All screenshots starting with Figure 8 2 are in a before and after fashion to point out what changed by applying the specific improvements 8 2 Identified Problems and Solutions 8 2 1 Computational Imbalance A var
30. object or binary modifications Windows HPC Server 2008 is shipped with a translator which produces trace log files in Vampirs Open Trace Format OTF The resulting files can be visualized with the Vampir performance data browser htto www tu dresden de zih ottf https silc zih tu dresden de otf2 current 2 Getting Started 2 1 Installation of Vampir Vampir is available for all major platforms Its installation process depends on the target operation system The following sections explain the particular installation steps for each system 2 1 1 Linux Unix An installer package is provided for Linux Unix systems To install Vampir run the installer from the command line lvamnpir 9 42 0 5L28B0d3rfd l1i50x 1822 8e6tu p b n Additional instructions are provided during installation For an overview of all available options run the installer package with the option ne1p It is possible to run the installer in silent unattended mode with the s command line option In this case the installer assumes default values for all options By default the installer associates Vampir with OTF and OTF2 files otf otf2 This allows to quickly open a trace file by double clicking its master file Furthermore a desktop icon and a desktop dependent menu items are generated During the first start of Vampir the license installation is completed Finally Vampir can be launched via the respective desktop icon or by using the com
31. or a 32 5 5 BHH mamaa maana maaa BUBBH mannana BHBHHHHH 0000000 DUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH Vampir Trace View Vampir Large wrf otf Edit Chart Filter Window Help SEKS mE Ini Timeline Os 5s 10s 15s 205 255 30 5 35 5 All Processes Exclusive Time per Function Group Figure 4 6 Summary Timeline value in an iterative approximation of the final result Counters are defined during the instrumentation of the application and can be individually assigned to processes An example Counter Data Timeline chart is shown in Figure The chart 1 stricted to one counter at a time It shows the selected counter for one measuring point e g process Using multiple instances of the Counter Data Timeline counters or processes can be compared easily The displayed graph in the chart is constructed from actual measurements data points Since display space is limited it is likely that there are more data points than display pixels available In that case multiple data points need to be displayed on one pixel width Therefore the counter values are displayed in two graphs A maximum line red and an average line yellow When multiple data points need
32. particular a search box is provided at the bottom of the dialog 89 a GWT 78 SAVING POLICY Preferences Function Groups Markers Counters Collectives Messages Events Scheme Default Color General Application DYN yo MEM e NETCDF a NoGroup PHYS 2527 7 VT L 1 WRF a Saving Search Apply Q cancel ok Figure 7 2 Appearance preferences Additionally to color modification the Function Groups dialog also allows regrouping of functions By using drag and drop functions can be freely assigned to any function group To create new function groups use the context menu entry Add Group Custom color and grouping schemes can be stored removed using the plus minus but tons at the top of the dialog 7 3 Saving Policy Vampir detects whenever changes to the various settings are made In the Saving Policy dialog it is possible to adjust the saving behavior of the different components to the own needs In the dialog Saving Behavior you tell Vampir what to do in the case of changed prefer ences The user can choose the categories of settings e g the layout that should be affected by the selected behavior Possible options are that the application automati cally Always or Never saves changes The default option is to have Vampir asking you whether to save or di
33. pplication 7 malloc DYN VT_API JO Application _ MEM free malloc realloc MPI i NoGroup PHYS VT WRF Saving Search o Apply cancel Por Figure 4 24 The Function Legend is shown on the left side The corresponding dialog for changing function colors is shown in the middle 4 3 2 Marker View The Marker View 4 25 lists all marker events included in the trace file The display organizes the marker events based on their respective groups and types in a tree like fashion Additional information like the time of occurrence or descriptions are provided for each marker By clicking on a marker event in the Marker View this event becomes selected in the timeline displays If this marker is located outside the visible area the zoom jumps to this event automatically It is possible to select marker events by their type as well Then all events belonging to that type are selected in the Master Timeline and the Process Timeline By holding the Ctrl or Shift key pressed multiple marker events can be selected exactly two marker events are selected the zoom is set automatically to the occurrence time of the markers 50 5 5 BHH mamaa maana maaa BUBBH mannana BHBHHHHH 0000000 DUBBHHHH BHEBBHHH
34. than the specified level shown 68 mamaa maana BHHHHH maaa 1 1 BUBHBH mannana EHEBHHEG BHBHHHHH 0000000 DUBBGHHHH BHEBBHHH 000006 DBDBBHH mmEnnu ooo 00000 0000 BHHBH 5 4 2 Examples In this chapter a few examples explain the usage of the function filter This enables the user to understand the basic principles of function filtering in Vampir at a glace It also illustrates a part of the set of available filter options provided by Vampir Unfiltered Trace File This section introduces the example trace file in an unfiltered state The timelines show a part of the initialization of the WRF weather forecast code The red color corresponds to communication MPI whereas the purple areas represent some input functions of the weather model Vampir Trace View Vampir WRF wrf otf W File Edit Chart Filter Window Help Timeline 5 95 6 05 6 55 7 0 5 7 55 8 05 lt 5 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 1 Figure 5 6 Master Timeline and Process Timeline without filtering 69 Amy Showing only MPI Functions In this example only functions that c
35. the counter are displayed in a color coded fashion The displayed counter in the chart can be chosen via the context menu entry Set Metric Own created custom metrics are listed under this option as well The option Adjust Bar Height to allows to change the height of the displayed value bars in the chart This useful for traces with a large number of processes Here the option Adjust Bar Height to Fit Chart Height tries to display all processes in the chart This provides an overview of the counter data across the entire application run Set Chart Mode allows to define whether minimum maximum or average values should be shown This setting comes into effect when multiple measured data points need to be displayed on one pixel If Maximum or Minimum is active the data point with the highest or lowest value is displayed respectively In case of Average the aver age of all data points on the respective pixel width is displayed This procedure is also explained in section Counter Data Timeline 4 1 3 35 Ap GWT 4 1 TIMELINE CHARTS Options Graph elements Fill Line Points Maximum v v Minimum Show total average line i Show caption i Show zero line i Show y axis label Apply Figure 4 9 Counter Timeline options dialog The value range of the color scale can be easily adjusted with the left mouse button To adjust the color coded value rage just drag the edges
36. the menu as shown in Figure 6 7 The entry Set Time Offset allows to manually set the time offset for the respective trace file The entry Reset Time Offset clears the offset File Edit Chart Filter Window Help dr sr _ A calcTest otf _ B calcTest otf 77 828 ms 15 643 _ C calcTest otf 16 966 s Timeline 05 35 65 95 125 155 Process 0 Process 1 3 Process 2 Process 3 Process 0 i Process 1 Process2 Process3 i Process 0 Process1 Process 2 Process 3 4 Figure 6 8 Alignment in the Navigation Toolbar The easiest way to achieve a coarse alignment is to drag the trace file in the Navigation Toolbar While holding the on Mac OS X modifier key pressed the trace can 84 0 BHHBHH maana BHHHHH maaa mannana _ EHHEBHHEH BHBHHHHH DUBBGHHHH BHEBBHHH 000006 900006 mmmEnnu BDBHHBH 00000 0000 BHHBH be dragged to the desired position with the left mouse button In Figure 6 8 the compute iterations of all example trace files are coarsely aligned comparison View Edit Chart Filter Window Help Or Sisi S S Bry 7 A calcTest otf 8663 9 8 BARE C B calcTe
37. to be displayed on one pixel width the red line shows the data point with the highest value and the yellow line indicates the average of all data points lying on this pixel width An optional blue line shows the lowest value When zooming into a smaller time range less data points need to be displayed on the available pixel soace Eventually when zooming far enough only one data point needs to be display on one pixel Then also the three graphs will merge together The actual measured data points can be displayed in the chart by enabling them via the context menu under Options The context menu entry Select Metric opens the selection dialog depicted in Fig 4 8 This dialog allows to choose the displayed counter in the chart Each counter is defined by its metric and its measuring point Note depending on the measurement 33 Z2 GWT 41 TIMELINE CHARTS Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help a ot FE 85 Timeline Aix Os 25 5 505 75s 100s 125 5 150 5 1755 2005 Process 0 Values of Counter PAPI L3 over Time pall El 13455 155 Figure 4 7 Counter Data Timeline not all metrics might be available on all measurement points The two left buttons in the dialog decide whether the counter should be selected by metric or by measuring point first In the case of Select by Metric there is also the option to Summarize mult
38. to render information in more detail The Trace View window can host an arbitrary number of charts Charts can be added by clicking on the respective icon in the Charts toolbar or the corresponding Chart 18 CHAPTER 3 BASICS Vampir Trace View Vampir Large wrf otf ulated Exclusive Time per 0 5 RADIATION_DRIVER 9177085 wsm MPI_Bcast 17 986 BI SOLVE_EM 171 423 s B wait 160 215 5 BI cALC 138 95 5 PHI 137 865 5 J ADVANCE_w 17 986 SOLVE EM 112 872 s vsu 171 423 5 B mei wit 109 672 s Ill ADVECT_SCALAR 160 215 s BI 107344 s Bill ADVANCE UV 138 95 s CALC P RHO PHI 106 174 5 ADVANCE MU_T 137 865 s ADVANCE_W 112 872 s YSU 109 572 s ADVECT SCALAR 107 344 s ADVANCE UV 106 174 s ADVANCE MU T MI Processes Accumulated Exclusive Time per Fun 250s Os RADIATION DRIVER WSM3 MPI Bcast Figure 3 3 Moving and Arranging Charts in the Trace View Window 1 Vampir Trace View Vampir Large wrf otf Processes Accumulated Exclu 250s Os RADI IVER 317708 wsm3 MPI Bcast 17 986 SOLVE_EM 171 423 5 BE we wait 160 215 s c 138 95 5 Bill CAL c 137 865 s BI w 112 872 s WI vsu 109 672 s IIl ADVE ALAR 107344 s ADVANCE UV 106174 s Bill ADVA MU_T 85 832 ALLO IELD 64 356 s SMAL PREP 59 9525
39. 0 Process 11 Process 12 Process 13 Process 14 Process 15 SES SSS Figure 4 27 Comparison between Context Information 52 BHHHM maana BHHHHH maaa BUBBH mannana EHHEBHHHEH BHBHHHHH BBBHEUSE 0000000 DHUBBHHHH BHEBBHHH 000006 DBDBBHH mmEnnu manom 0000 BHHBH 4 4 Customizable Performance Metrics Vampir is shipped with a set of predefined customizable metrics that reflect known sources for performance issues and can serve as starting point for application specific customizations Figure 4 28 shows the list of custom metrics that are predefined in Vampir The list is accessible via the context menu entry Customize Metrics in the Performance Radar or the Counter Data Timeline chart x custom Metrics Active Description FLOPS in User Defined Function Edit Bandwidth Duplicate Y Volume in Transit Remove v MPI Latencies v Message Data Rate Import Message Transfer Times Export Message Volume in Transit Simultaneous I O Operations Simultaneous Messages Y Time Spent in MPL Wait Figure 4 28 List of predefined customizable performance metrics The following time dependent metrics are provided e FLOPS in User Defined Function Floating point performance
40. 14 Process 21 Process 28 84 560075 5 Process 35 Process 42 Process 49 Process 56 Process 63 Process 70 Process 77 Process 84 Process 91 118 5 202 324368 84 324368 Figure 8 6 Overview showing a significant overall improvement As is shown by the Ruler see Section 4 1 in Figure 8 6 two large iterations now take 84 seconds to finish Whereas at first Figure 8 1 it took roughly 140 seconds making a total speed gain of 40 This huge improvement has been achieved by using the insight into the program s runtime behavior provided by the Vampir toolkit to optimize the inefficient parts of the code 98
41. 5 BHH BHHHM BHHHHH maaa mannana BHHHUHH EHHEBHHEH BHBHHHHH DHUBBGHHHH BHEBBHHH 000006 DBDBBHH _ DBHHBH 00000 0000 BHHBH 2 GETTING STARTED mpiexec wdir share userHome tracefile SUSERPROFILE trace etl myApp exe e wdir sets the working directory has to be there e SUSERPROFILE translates to the local home directory e g C Users userHome on each compute node the event log file etl is stored locally in this directory 2 the event log files throughout all compute nodes mpiexec cores 1 wdir USERPROFILES mpicsync trace etl e cores 1 run only one instance of mpicsync on each compute node 3 Format the event log files to OTF files mpiexec cores 1 wdir USERPROFILES etl2otf trace etl 4 Copy all OTF files from compute nodes to trace directory on share mpiexec cores 1 wdir USERPROFILE cmd c copy y x otf share userHome Trace 2 3 Starting Vampir and Loading Performance Data x Open Recent File Help VAMPIR B Recent Files INampir 5mall wrf otf Nampir Large wrf otf Open Other Figure 2 2 List of recent trace files Viewing performance data with the Vampir GUI is very easy On
42. 55 150s 175s 200s Metric PAPI_FP_OPS Opacity Vampir Large wrf otf Vampir Window Help 05 255 505 Process 0 Process 1 5 2 Process 3 Process 4 755 100 5 125 5 150 5 175 5 200 5 Metric PAPI OPS Opacity XK tx Figure 4 14 Image series showing different opacity settings for the performance data overlay going from zero opacity in the top image to full opacity in the bottom image 40 CHAPTER 4 PERFORMANCE DATA VISUALIZATION Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help 0 Timeline Ax Os 255 50s 75s 1005 1255 1505 1755 2005 aa Metric PAPI_FP_OPS Opacity m tx ud x 1 A A Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 0G 1G 2G 3G 4G 129 5 5 Figure 4 15 Highlighted areas with a low FLOP rate Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help mrisesimu o b Timeline Ax Os 255 505 755 1005 1255 1505 1755 Metric PAPI FP OPS Opacity t x E x pneu Al Process 0 ELA Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Pr
43. 9 MP UTIL Process 35 0 73 Application Process 42 0 32 Process 49 ES 0 19 COUPLE Process 56 Process 63 Process 70 Process 77 Process 84 Process 91 Figure 8 1 Master Timeline and Function Summary showing an overview of the gram run Getting a grasp of the program s overall behavior is a reasonable first step In Figurej8 11 Vampir has been set up to provide such a high level overvievv of the models code This layout can be achieved through two simple manipulations Set up the Master Timeline to adjust the process bar height to fit the chart height All 100 processes are now arranged into one view Likewise change the event category in the Function 92 M BHHHM maana BHHHHH mana mannana EHHEBHHEH BHBHHHHH BHEBBHHH 000006 900008 _ manom 0000 8 AUSE CASE Summary to show function groups This way the many functions are condensed into fewer function groups One run of the instrumented program took 290 seconds to finish The first half of the trace Figure 8 1 A is the initialization part Processes get started and synced input is read and distributed among these processes The preparation of the cloud
44. E manom 0000 BHHBH 00000000 Another way to navigate to a marker in the timeline charts is to use the Vampir zoom If the user zoomed in the Master Timeline or the Process Timeline into the desired zooming level then a click on a marker in the Marker View will shift the timeline zoom to the marker position Thus the selected marker appears in the center of the timeline chart see Figure 6 1 1 Comparison View Edit Chart Filter Window Help gt t ri go amp B 8 7 C pe special otf 3 otential vt otf 0 1134 s 0 1136 s Process 0 Process 1 Process 2 Process 3 Process 0 Process 1 Process 2 Process 3 Timeline 0 1138 s 0 1140 s 0 1142 s main main main main s Lu Type Process Processgroup r1 datatype special otf Warning 7 MARMOT Warning Error fll MARMOT Error Process 2 Process 1 Process 3 Marker View Time 0 113515 s 0 11319 s 0 114129 5 Duration Description 0s ERROR MPI Type contiguous oldty 0s ERROR MPI TYPE zorr M us Process 0 arker2 potential vt otf Warning 7 MARMOT Warning Error MARMOT Error Process 0 0 113857 5 0 1148 5 ERROR contiguous ERROR Send datatype is n Zoom Between Marker Align Traces at Marker Figure 6 11 Jump to a marker in the Master Timeline
45. G AND REDUCTION Show only functions that match al gt of the following conditions Name Contains 2 mpi Duration t is greater than 1250 Milliseconds 2n canca Vampir Trace View Vampir WRF wrf otf W File Edit Chart Filter Window Help m xite BI lt 5 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 1 Timeline 5 5 5 6 0 5 6 5 5 Figure 5 10 Combining rules using all mmuumum nummum DHUHHBHHH nu nu 900008 nnuu The second example illustrates the usage of the all relation Here all shown functions have to satisfy both rules Therefore the filter shows only MPI functions that have a duration time of more than 250 ms 73 P 84 FUNCTION FILTER Building Ranges with Number of Invocation Rules The combination of rules also allows for the filtering of functions in a specified criteria range The following example filter setup shows all functions whose number of invoca tions lie inside the range between 2000 and 15000 Y Show only functions that match al of the following conditions Number of Invocations T is greater than
46. It is still avail able as Open Source software but no longer under active development see Score P section 2 2 1 During a program run of an application VampirTrace generates an OTF trace file which can be analyzed and visualized by Vampir The VampirTrace library allows MPI communication events of a parallel program to be recorded in a trace file Additionally certain program specific events can be included To record MPI communication events simply re link the program with the VampirTrace library A new compilation of the program source code is only necessary if program specific events should be added To perform measurements with VampirTrace the application program needs to be in strumented which is done automatically All the necessary instrumentation steps are handled by the compiler wrappers of VampirTrace vtcc vtcxx 77 vtf90 and the ad ditional wrappers mpicc vt mpicxx vt mpif77 vt and mpif90 vt in Open MPI 1 3 All compile and link commands in the used makefile should be replaced by the Vampir Trace compiler wrapper which performs the necessary instrumentation of the program and links the suitable VampirTrace library Simply use the compiler wrappers without any parameters vtf90 hello 90 o hello Running VampirTrace instrumented application results in an OTF trace file stored the current working directory where the application was executed On Linux Mac OS and Sun Solaris the default name of the trace
47. Operations I O Events I O Groups File Names Operation Types Collective Operations Table 5 1 Object filtering options Note The available selection methods are the same across all filter dialogs except the Function Filter The check box nclude Exclude All either selects or deselects every item Specific items can be selected deselected by clicking into the check box next to it Furthermore it is possible to select deselect multiple items at once For this mark the 60 5 BHH 1 BHHHM maana HHHHH maaa mannana _ manomano BHBHHHHH BBBHEUSE 0000000 DHUBBHHHH BHEBBHHH 000006 DBDBBHH mmmEnnu DBHHBH 00000 0000 BHHBH desired entries by clicking their names while holding either the Shift or the Ctrl key holding the Shift key every item between the two clicked items will be marked Holding the key on the other hand enables you to add or remove specific items from to the marked ones Clicking into the check box of one of the marked entries will cause selection deselection for all of them 5 1 Process Filter Edit process selection using Process Hierarchy v IncdudejeExclude All Process 0 Process 1 vi Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Y Process 8 Process 9 Pr
48. Read rsl error 0003 Rename rsl error 0004 Seek lt rsl error 0005 Y Sync Y rsl error 0006 Sync rsl error 0007 v Unlink rsl error 0008 Y Unlock rsl error 0009 Write rsl error 0010 rsl error 0011 rsl error 0012 rsl error 0013 rsl error 0014 v rsl error 0015 Y rsl out 0000 cil rsl out 0001 b Figure 5 3 Filter Figure 5 3 depicts the VO Filter dialog The dialog allows to selectively filter VO events displayed in timelines and statistics Available filter criteria are Groups or the Operation Type t is also possible to filter VO operations based on input and output files 64 5 BHH BHHHM mamana BHHHHH maaa BUBB mannana BHHHUHH EHHEHHHEH BHBHHHHH BBBHEUHE 0000000 DUBBHHHH BHEBBHHH 000006 900006 mmmEnnu manom 0000 BHHBH CHAPTER 5 INFORMATION FILTERING AND REDUCTION S MBR 5 4 Function Filter The filtering of functions in Vampir is controlled via the Function Filter dialog which can be accessed via the main menu under Filter Functions Initially a list of available rule sets is depicted as can be seen in Figure By default the list only shows a None entry It can only be one filter active at a given time To select the active filter use the radio buttons on the left hand side of the list Cl
49. Ruler The Ruler is enabled by default during every zoom operation in a timeline chart In order to use the Ruler for measurement only i e without performing any zoom hold the Shift key pressed while clicking on any point of interest in a timeline chart and moving the mouse while holding the left mouse button pressed A ruler like pattern appears in the timeline chart which provides the exact time between the start point and the current mouse position 4 1 1 Master Timeline and Process Timeline In the Master Timeline and the Process Timeline detailed information about functions communication and synchronization events is shown Timeline charts are available for individual processes Process Timeline as well as for a collection of processes Master Timeline The Master Timeline consists of a collection of rows Each row represents a single process as shown in Figure 4 1 A Process Timeline shows the different levels of function calls in a stacked bar chart for a single process as depicted in Figure 4 2 Every timeline row consists of a process name on the left and a colored sequence of function calls or program phases on the right The color of a function is defined by its group membership e g MPI Send belonging to the function group has the same color presumably red as MPI Recv which also belongs the function 27 4 1 TIMELINE CHARTS Trace View Vampir Large wrf otf Vampir Edit Chart Filter
50. SICS routine is now equal among all processes Through serial optimization the duration has been decreased from about 1 5 to 1 0 second A decrease in duration of about 339 is quite good given the simplicity of the changes done 8 2 3 High Cache Miss Rate The latency gap between cache and main memory is about a factor of 8 Therefore optimizing for cache usage is crucial for performance you don t access your data in a linear fashion as the cache expects so called cache misses occur and the spe cific instructions have to suspend execution until the requested data arrives from main memory high cache miss rate therefore indicates that performance might be im proved through reordering of the memory access pattern to match the cache layout of the platform Problem As can be seen in the Counter Data Timeline Figure 8 4 the CLIPPING routine light blue causes a high amount of L2 cache misses Also its duration is long enough to make it a candidate for inspection What caused these inefficiencies in cache usage were nested loops which accessed data in a very random non linear fashion Data access can only profit from cache if subsequent read calls access data in the vicinity of the previously accessed data Solution After reordering the nested loops to match the memory order the tuned version of the CLIPPING routine now needs only a fraction of the original time Figure 8 5 96 maana
51. Trace View Vampir WRF wrf otf W File Edit Chart Filter Window Help tre 7 0 Timeline 239 6 0 6 5 7 0 1 9 9 8 0 gt Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 1 MED_INITIALDATA_INPUT MODULE IO DOMAIN INPUT MODEL INPUT WRF_INPUTIN Figure 5 8 Showing only functions with more than 250 ms duration 71 Amy Combining Function Name and Duration Rules This example combines the two previous rules First the any relation is used Thus the filter shows all functions that have at least 250 ms duration time and additionally also all MPI functions Show only functions that match any gt of the following conditions Cwtens Duration Is greaterthan 250 Miliseconds Vampir Trace View Vampir WRF wrf otf W File Edit Chart Filter Window Help m wie ttre sa Br Timeline K 6 05 6 55 7 0 5 7 35 8 05 gt Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 1 Figure 5 9 Combining rules using any 72 5 INFORMATION FILTERIN
52. Window Help Jis BH 0 aye 65 Timeline Ax 84 7 5 84 85 84 95 85 05 85 15 85 25 Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 Sa SE A Figure 4 1 Master Timeline Vampir Trace View Vampir Large wrf otf YW Edit Chart Filter Window Help Timeline 13 255 13 505 13 75 5 14 005 14 255 14 505 14 75 5 15 00 5 15 25 5 0 ml 3 1 181 OW ON OU dt GN Figure 4 2 Process Timeline 28 5 BBHHHM BHHBHH maana maaa 1 BUBB mannana BHHHHUEH mHEHHHEH BHBHHHHH maHHEHHH DUBBHHHH BHEBBHHH 000006 DBDBBHH mmEnnu BDBHHBH manom 0000 BHHBH group MPI Clicking on a function highlights it and causes the Context View display to show detailed information about that particular function e g its corresponding func tion group name time interval and the complete name The Context View display is explained in Chapter 4 3 3 By clicking on a process label additional information about the related process is shown In the Context View Proces
53. Windows the tool can be started by double clicking its desktop icon if installed or by using the Start Menu 13 a au GW T o 2 3 STARTING VAMPIR AND LOADING PERFORMANCE DATA On a Linux based machine run vampir in the directory where Vampir is installed A double click on the application icon opens Vampir on Mac OS X systems At startup Vampir presents a list of recently loaded trace files as depicted in Figure 2 2 Selecting a list entry and clicking the Open button loads the respective trace The recent list is empty when Vampir is started for the first time 2 3 1 Loading a Trace File To open an arbitrary trace file click on Open Other or select Open in the File menu which provides the file open dialog depicted in Figure 2 3 File Help Favorite Links Path Vampir Large E Filesystem All trace files otf otf2 elg esd open Subset Cancel Figure 2 3 Loading trace file in Vampir lt is possible to filter the files in the list The file type input selector determines the visible files The default is All trace files otf otf2 elg esd which only shows trace files that can be processed by the tool All file types can be displayed by using Files Favorite directories can be added to Favorite Links on the left hand side by clicking the plus button below The five most recently visited directories will automatically be listed After selection
54. XT NCD SUPP S LOVVERCASE Process 1 960 ALL SUB 1 960 JMPI Scatterv 2 MODULE INTEGRATE INTEGRATE 960 MPI_Gatherv 3 SOLVE_INTERFACE 960 JMODULE BIG S LC P RHO 4 SOLVE_EM 896 WRF GLOBAL TO PATCH REAL 5 720 MODULE BC EM HYS BC DRY 1 6 720 MODULE BIG S EM CALC PHP 7 720 MODULE_BIG_S EM CALC_ALT 8 720 MODULE BIG S EM CALC CQ 720 MODULE_BIG S CALC WW CP i 720 MODULE BIG LE MOMENTUM i Figure 5 12 Show functions outside a specified range 75 P 84 FUNCTION FILTER Call Path contains WRF_INPUTIN In this example only functions that are called directly or indirectly by WRF_INPUTIN are shown As a consequence all call paths start with WRF_INPUTIN All other functions are filtered Filter Functions Show only functions that match of the following conditions Description Filter call Path Contains WRF_INPUTIN x Cancel or Nc Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help 5 1 nzm 55745 ATO S miumwe jk G pe 0 My Timeline 6s 75 85 95 105 115 125 135 145 Process 0 55 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 1 WN 4 Figure 5 13 Call path filter which contains WRF_INPUTIN 76 BHHHM maana BHHHHH mannana
55. Z r29U SO app will now be built by scorep 1 90 app1 f90 app2 f90 app using the Score P instrumentor When makefiles are employed to build the application it is convenient to define a place holder variable to indicate whether a preparation step like an instrumentation is desired or only the pure compilation and linking For example if this variable is called PREP then the lines defining the C compiler in the makefile can be changed from mpicc to 5 mpicc and analogously for linkers and other compilers One can then use the same makefile to either build an instrumented version with the make PREP scorep command or a fully optimized and not instrumented default build by simply using make in the standard way i e without specifying PREP on the command line Detailed information about the installation and usage of Score P can be found in the Score P user manual http www score p org 10 5 8BHH 5 DDDB maana BHHHHH maaa mannana BHHHHHUHH EHHEBHHEH BHBHHHHH BHEBBHHH 000006 900008 manono BDBHHBH manom 0000 BHHBH 2 GETTINGSTARTED m 2 2 2 VampirTrace Vampir Irace used to be the recommended monitoring facility for Vampir
56. _ _ 54 722 5 _ 46 491 s SUMFLUX 44 281 s 44 016 s CUMU IVER 36 653 s A DRY 33 484 s MOIS P EM Figure 3 4 Moving and Arranging Charts in the Trace View Window 2 19 4 menu entry With few more clicks charts can be combined to a custom chart rangement as depicted in Figure 3 2 Customized layouts can be saved as described in Chapter 7 3 Every chart can be undocked or closed by clicking the dedicated icon in its upper right corner as shown in Figure Undocking a chart means to free the chart from the current arrangement and present it in an own window To dock undock a chart follow Figure 3 6 respectively Figure Function summary il All Processes Accumulated Exclusive 1 000 5 Os M 495 a b 781 5 lt 1 5 lt 0 1 VT Figure 3 5 Closing right and Undocking left of a Chart Considering that labels e g those showing names or values of functions often need more space to show its whole text there is a further option of resizing In order to read labels completely it might be useful to alter the distribution of space shared by the labels and the graphical representation in a chart When hovering the blank space between labels and graphical representation a movable separator appears By drag ging the separator decoration with the left mouse button the chart space provided
57. common charts of Vampir In contrast to the ordinary Trace View the Comparison View opens one chart instance for each trace file i e with three open trace files one click on the Master Timeline icon opens three Master Timeline charts By using the icon menus accessible via the triangles next to the chart icons it is also possible to open only one chart instance for the selected trace Also in order to distinguish the same charts be tween the trace files a dedicated background color is assigned to all charts belonging to one trace The background color can be changed by clicking the respective colored rectangle next to the trace file path in the Navigation Toolbar Comparison View File Edit Chart Filter Window Help dr sr C A calcTest otf _ B calcTest otf _ C calcTest otf 16 966 s Timeline Function Summary Os 5s 10s 155 All Processes Accumulated Exclusive Time per Function Group 40 5 205 05 Process 0 Process 1 6 257 s IBI vT 2 718 456 us TEST Process 3 480 336 us CALCULATION 296 117 us Application Process 0 Process 1 Process 2 Function Summary m All Processes Accumulated Exclusive Time per Function Group 4s 3s 25 15 05 les Process 0 939 444 us VT API Process 1 241 939 us TEST Process 2 228 578 us CALCULATION Process 3 82 214 us Application Process 0 1 Function Summary 3
58. cribed in Chapter 3 The available charts and the information provided by them are explained in Chapter 4 2 3 2 Loading a Trace File Subset To handle large trace files and save time and memory resources it is possible to load only a performance data subset from a trace file For this purpose the open dialog Figure 2 3 provides the button Open Subset Clicking on this button opens a trace data pre selection dialog as depicted in Figure An overview snapshot of the recorded application run is given at the top of the dialog The time range of interest can be set with the edge markers on the left and right of the overview snapshot Likewise the time range to be loaded can be set explicitly in the input fields From and markers are available in the trace file their timing information can be used as reference points as well Two markers need to be selected first use shift mouse click for the second marker Next click on Zoom Between Marker to set the respective time interval in the From and To input fields The event data to be loaded can also be restricted to certain processes or threads of execution by disabling unwanted instances in the selection area entitled Processes see Section 5 1 for further details By using the selection areas Functions Counter and Other Events the loaded trace data can be further restricted to certain events and event types Once the data subset of interest is specified a click on the OK button starts
59. d into levels which represent the different call stack levels of function calls The initial function begins at the first level a sub function called by that function is located a level beneath and so forth If a sub function returns to its caller the graphical representation also returns to the level above In addition to the display of categorized function invocations Vampirs Master and Process Timeline also provide information about communication events Messages exchanged between two different processes are depicted as black lines In timeline charts the progress in time is reproduced from left to right The leftmost starting point of a message line and its underlying process bar therefore identify the sender of the message whereas the rightmost position of the same line represents the receiver of the message The corresponding function calls usually reflect a pair of commu nication directives like MPI Send MPI Recv Collective communication like MPI Gatherv Is also displayed in the Master Timeline as shown in Figure 4 3 Furthermore additional information like message bursts markers and events is available Table 4 1 shows the symbols and descriptions of these objects 29 GWT 41 TIMELINE CHARTS Figure 4 3 Selected MPI Collective Master Timeline since the Process Timeline reveals information of one process only short black arrows are used to indicate outgoing communication Clicking on me
60. d on the timelines Here the red areas indicate high computational activity and therefore mark the compute iterations 38 5 5 BHH mamaa maana maaa BUBBH mannana BHBHHHHH 0000000 DUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH 209 Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help BP Y Timeline Ax Os 255 505 75s 100s 125s 150s 175 5 200 5 gt wm Metric PAPI FP OPS Opacity SP tx Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 e 0 00 0G 1 Values of Metric PAPI FP OPS over Time Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 N UJ gt 4 4 r2 UJ e 4 1325 Figure 4 13 Master Timeline top chart and Performance Radar bottom chart dis playing the same PAPI FP OPS counter High and Low FLOP Rate In order to analyze the FLOP rate the overlay mode of the Master Timeline is config ured to show the performance counter PAPI FP OPS To identify functions with a high or low FLOP rate the value range of the color scale can be limited This is done by dragging the ed
61. d zooming in this view is described in the next section Comparison View File Edit Chart Filter Window Help SO Sris my S S C A calcTest otf C B calcTest otf _ C calcTest otf 16 966 s Timeline Function Summary Os 5s 10s 15s All Processes Accumulated Exclusive Time per Functio 40s 20s Os Process 0 MPI Process 1 6 257 s B Process 2 718 456 us TEST Process 3 480 336 us CALCULATION 296 117 us Application Process 0 Process 1 Process 2 Function Summary Function Summary 2 X z All Processes aie Processes Accumulated 05 10 5 0 5 SSS EM 939 444 us VT VT API 241 939 us TEST 3 68 ms IB Ap on Process 1 228 578 us CA ON 2 176 ms 7 CA ON m 82 214 us Ap on 1 745 ms TEST Process 3 mE KI Figure 6 4 Open Comparison View To save a comparison session use the menu entries File Save or File Save As This will store a vcompare file containing the compared trace files settings and the Comparison View layout To restore a comparison session simply open the respective vcompare file Previous comparison sessions are also available in the recent open files list of Vampir 81 GWT 6 2 USAGE OF CHARTS 6 2 Usage of Charts For the comparison of performance metrics the Comparison View provides all
62. e 4 20 In order to filter out messages click on the associated label or color representation in the chart and then choose Filter from the context menu 4 2 4 Communication Matrix View The Communication Matrix View is another way of analyzing communication imbal ances It shows information about messages sent between processes SOQO Trace View Vampir Large wrf otf Vampir Edit Chart Filter Communication Matrix View Window Help Average Message Data Rate Receiver 46 d lt d lt lt d lt lt lt lt lt lt Process 0 720 MiB s 1 Process 2 3 Process 3 Process C3 lt 5 5 480 MiB s Process 6 g Process 7 CD C 400 MiB s Process 8 LB Process 9 320 MiB s Process 10 11 CD m Process 12 Process 13 nm Process 14 80 MiB s Process 15 EB Figure 4 21 Communication Matrix View 0 MiB s BHHHM maana BHHHHH maaa 1 mBUBB mannana EHHEBHHEG BHBHHHHH BBBHEUSE DUBBHHHH mBHEBBHHH 000006 DBDBBHH mmEnnu 00000 0000 BHHBH
63. e color gradient in charts allows to switch off the color gradient used in the perfor mance charts The next option allows to change the style and size of the font Show source code enables the internal source code viewer This viewer shows the source code corresponding to selected locations in the trace file In order to open a source file first click on the intended function in the Master Timeline and then on the source code path in the Context View For the source code location to work properly you need a trace file with source code location support The path to the source file can be adjusted in the Preferences dialog A limit for the size of the source file to be opened can be set too In the Analysis section the number of analysis threads can be chosen If this option is disabled Vampir determines the number automatically by the number of cores e g two analysis threads on a dual core machine In the Miscellaneous section the user can activate the following functionality Enable an automatic check for newer versions of Vampir activate the color blindness support 88 maana HHHHH maaa mannana BHHHHUuH EHEBHHEH BHBHHHHH manomano 0000000 DUBBHHHH BHEBBHHH 000006 DBDBBHH mmmEnnu BDBHHBH 00000 0000 BHHBH CHAPTER 7 CUSTOMIZATION 1 VMER Preferences Charts
64. e filtering of functions by their call level Here only func tions with an enter event less then call level five are shown All other functions are filtered Show only functions that match of the following conditions Description Filter Call Level gt Is less than 5 5 E Trace View Vampir Large wrf otf Vampir Edit Chart Filter Window Help Timeline 65 ES 8s 95 105 115 125 135 145 155 MES Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 1 1 N 2 MED_INITIALDATA_INPUT INTEGRATE 3 INPUT MODEL INPUT VV START DOMAIN n MED_HISTORY_OUT Figure 5 15 Showing only functions with a call level less than five 78 maana 1 mannana _ EHEBHHEH DHUBBGHHHH 000006 DBDBBHH manono DBHHBH 00000 0000 BBHHBH 0000000000000000000 6 Comparison of Trace Files In Vampir the comparison of trace files seamlessly integrates with the functionality explained in the previous chapters of this document The user can benefit from already gained experiences For the comparison of performance characteristics all common charts are provided Additionally
65. er Function 400s 350s 300 250 200 150 5 100 5 505 05 RADIATION_DRIVER ESS CALC P RHO 2 128775 vsu ADVECT SCALAR ADVANCE UV aDVANCE MU 64 8885 ALLOC SPACE FIELD INPO SMALL STEP_PREP CALC_P_RHO _ 46 491 5 SUMFLUX i 44 281 s IBI Rk TENDENCY 44 016 s CUMULUS DRIVER 36 6535 BI RK_ADDTEND_DRY 33 484 s MOIST_PHYSICS_PREP_EM 33 049 5 IR UPDATE SCALAR 32 674 s IBI U 32 604 5 IB SMALL_STEP_FINISH Figure 3 7 Docking of a Chart 21 a 2 GWT 33 ZOOMING Vampir Trace View Vampir Large wrf otf W File Edit Chart Filter Window Help rus I Function Summary All Processes pe Exclusive Time per Function 400 5 350 5 300 5 250 5 200 5 150 5 100 5 505 Function Summary All Processes Accumulated Exclusive Time per Function 400s 350s 300s 2505 2005 1505 100 5 E Figure 3 8 Resizing Labels A Hover a Separator Decoration B Drag and Drop the Separator e Sort By Rearrange values or bars by a certain characteristic 3 3 Zooming Zooming is a key feature of Vampir In most charts it is possible to zoom in and out to get detailed or abstract views of the visualized data In the timeline charts Zooming produces a more detailed view of a selected time interval and therefore reveals new information that was previously hidden in the larger sectio
66. er functions by their execution time The Process Summary can be shown as Histogram or as Kiviat Chart To switch between these representations use the Set Chart Mode entry of the context menu 4 2 3 Message Summary The Message Summary is a Statistical chart showing an overview of all messages grouped by certain characteristics Figure Vampir Trace View Vampir Large wrf otf AVERAGE MINJ Figure 4 20 Message Summary Chart with metric set to Message Transfer Rate shovv ing the average transfer rate A and the minimal maximal transfer rate B 45 GWT 4 2 STATISTICAL CHARTS All values are represented in a bar chart fashion The number next to each bar is the group base while the number inside a bar depicts the values depending on the chosen metric Therefore the Set Metric sub menu of the context menu can be used to switch between Aggregated Message Volume Message Size Number of Messages and Message Transfer Rate The group base can be selected via the context menu entry Group By Possible options are Message Size Message Tag Communicator and Source Code Location Note There will be one bar for every occurring group However if the metric is set to Message Transfer Rate the minimal and the maximal transfer rate is given in an additional small bar beneath the main bar showing the average transfer rate The additional bar starts at the minimal rate and ends at the maximal rate see Figur
67. eristics use the Set Metric sub menu of the context menu 48 BHHHM mannana EHHEBHHEH BHBHHHHH 0000000 DUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH Vampir Trace View Vampir Large wrf otf File Edit Chart Filter Window Help Call Tree All Processes Function Min Inclusive Time Max Inclusive Time SPEC BDY SCALAR 1 023 ms 3 804 ms sPEC BDYTEND 386 750 us 2 947 ms 5 _ DRY 12 235 ms 0 154 5 SMALL STEP PREP 3 951 5 4 097 5 SMALL STEP FINISH 1 989 5 2 072 5 2 SET TILES2 3 373 ms 4 106 ms VVRF MESSAGE 140 600 us 282 700 us REGION_BOUNDS 1 572 ms 1 803 ms NL GET TILE SZ Y 24 800 us 25 550 ys NL GET TILE sz X 24 950 us 28 450 us NL GET NUMTILES 25 850 us 26 800 us malloc 7 400 us 13 950 us v Callers Callees Bl END TIMING 3 IN LANDUSE INIT 1 IN EXT REALFIELDIO 1 IN EXT NCD INTFIELDIO 1 IN init 1 wRF TERMIO 1 B VVRF MESSAGE 6 Bl INITIAL CONFIG 1 v Find Function write Previous Next Figure 4 23 Call Tree To leaf through the different function calls it is possible to fold and unfold the levels of the tree This can be achieved by double c
68. ess 3 Process 0 Function Summary All Processes Accumulated Exclusive Time per Funct 2ms 1 0 ms 2 662 Application IECIT m9 CALCULATION TESH TEST VT API 418 05 us IBI Process 0 Process 0 Figure 6 6 Zoom to compute iterations of trace C As shown in Figure 6 5 trace A has the biggest duration time The duration of trace C is so short that it is barely visible Zooming into the compute iteration phase of trace C makes them visible but due to the coupled zoom also displays only the MPI_Init phase of trace A and B see Figure In order to compare the compute iterations between the traces they need to be aligned properly This process is described in the next section 83 ZI Amy 6 3 Alignment of Multiple Trace Files The Comparison View functionality to shift individual trace files in time allows to com pare areas between traces that did not occur at the same time For instance in order to compare the compute iterations of the three example trace files these areas need to be aligned to each other For the example traces this is required because the initialization of the application took different times on the three machines Set Time Offset N Reset Zoom CtritR Reset Time Offset Figure 6 7 Context menu controlling the time offset There are several ways to shift the trace files in time One option is to use the context menu of the Navigation Toolbar A right click on the toolbar reveals
69. for the labels can be resized The whole process is illustrated in Figure 3 2 Context Menus All chart displays have their own context menu containing common as well as display specific entries In this section only the most common entries will be discussed A context menu can be accessed by right clicking anywhere in the chart window Common entries are e Reset Zoom Go back to the initial state in horizontal zooming e Reset Vertical Zoom Go back to the initial state in vertical Zooming e Set Metric Set the values which should be represented in the chart e g change from Exclusive Time to Inclusive Time 20 CHAPTER 3 BASICS Vampir W Edit Chart Filter Window Help Srusexvtmss 0 Trace View Vampir Large wrf otf Function Summary Timeline Ax Os 50 100 5 150 5 2005 All Processes Accumulated Exclusive 1 000 5 05 Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 7 781 Process 6 lt 1 Process 7 lt 0 1 VT API Process 8 9 Function Legend Process 10 m DYN Process 11 P yo Process 12 ll mem Process 13 Bl IN Process 14 PHYS Process 15 I m VT API WRF s Figure 3 6 Undocking of a Chart Vampir Function Summary Vampir Large wrf otf File Window Help All Processes Accumulated Exclusive Time p
70. for a given function which can be set by the user see Section 4 4 1 I O Bandwidth Aggregated file I O bandwidth requires that I O events have been recorded VO Volume in Transit Aggregated number of bytes in transit to and from the I O system e MPI Latencies Duration of individual MPI calls Message Data Rate Bytes per second exchanged with message passing direc tives 53 a a GWT cence 444 CUSTOMIZABLE PERFORMANCE METRICS e Message Transfer Times Latencies of individual message passing directives e Message Volume in Transit Aggregated number of bytes in transit via messages e Simultaneous I O Operations Number of interleaved directives e Simultaneous Messages Number of interleaved message passing directives e lime Spent in Wait Times spent in Wait routines 4 4 1 Metric Editor The Custom Metrics Editor allows to define derived metrics based on existing coun ters and functions This is particularly useful as the performance data overlay of the Master Timeline Section 4 1 4 is capable of displaying such custom metrics as well The editor is accessible via the list of customizable performance metrics explained in the previous section by clicking on the Edit button Figure 4 29 shows an example con struction of a custom metric Wait Time This metric is an addition of the time spent in the functions MPI Irecv and Wait Custom metrics are build from input metrics that are linked together usin
71. forschung innovation GWT Vampir 8 User Manual ERR ERED K H E E E EJ ET ET Kena KEE an E E ES E cen E EJ ES ESI L1 EJ ES ES ERI RI I I E ES ES GWT forschung innovation Copyright 2014 GWT TUD GmbH Blasewitzer Str 43 01307 Dresden Germany http gwtonline de Support Feedback Bugreports Please provide us feedback We are very interested to hear what people like dislike or what features they are interested in If you experience problems or have suggestions about this application or manual please contact sezvice amp vampir eu When reporting a bug please include as much detail as possible in order to reproduce it Please send the version number of your copy of Vampir along with the bug report The version is stated in the About Vampir dialog accessible from the main menu under Help About Vampir Please visit http vampir eu for updates http vampir eu Manual Version Vampir 8 4 November 2014 maana BHHHHH maaa mannana BHBHHHHH DHUBBGHHHH 000006 DBDBBHH manono DBHHBH 00000 0000 BBHHBH 0000000000000000000 Contents Contents 00000000 9 1 1 Event based Performance Tracing and Profilingj
72. g a set of available operations In the editor the context menu accessible via the right mouse button allows to add new input metrics and op erations All created custom metrics become available in the Set Metric selections of the Performance Radar and Counter Data Timeline charts They are available as well in the overlay mode of the Master Timeline Custom metrics can be exported and imported in order to use them in multiple trace files 54 CHAPTER 4 PERFORMANCE DATA VISUALIZATION x Custom Metrics Description Wait Time Metric x MPI Irecv m Metric x Function Duration MPI Wait EN Vom Gere Figure 4 29 Custom metrics editor showing the construction of custom Wait Time metric The metric is defined by the addition of the duration of MPI Irecv and MPI Wait functions 55 a a GWT rimon 44 CUSTOMIZABLE PERFORMANCE METRICS 4 4 2 Examples MPI Wait Duration x Custom Metrics Description Wait Duration Unit 1 5 Metric Function Duration MPI Wait Exclusive Figure 4 30 Construction of custom metric showing the MPI_Wait duration In Vampir it is also possible to identify long running functions In this example long running invocations of the function MPI Wait are highlighted First step is to construct a custom metric showing the MPI wait duration time The custom metric editor is described in more detail in Secti
73. ges of the colored area of the scale to the desired minimum maximum values That way only values inside the chosen range appear color coded in the chart Outside values are visualized in gray Figure 4 15 and Figure 4 16 two examples Functions with a low FLOP rate are highlighted in Figure 4 15 The color scale is limited to a range between 100 M and 1 6 G FLOPS The minimum value is raised to 100 M in order to gray out non computing functions like MPI In Figure 4 15 areas with a low FLOP rate are highlighted in red In this example these areas represent functions in the beginning of each iteration Functions with a high FLOP rate are highlighted in Figure 4 16 Here the color scale is set to highlight only areas with the highest FLOP rate These areas are represented by functions in the compute iterations 39 GWT forschung innovation Trace View File Edit Chart Filter zm mie 4 1 TIMELINE CHARTS Vampir Large wrf otf Vampir Window Help li E 05 Process 0 Process 1 Process 2 Process 3 Process 4 Trace View File Edit Chart Filter Timeline 55 100 5 125 5 150 5 Opacity __ 1755 2005 Metric PAPI FP OPS Vampir Large wrf otf Vampir Window Help 5 d 1 Os 255 50 5 Process 0 Process 1 Process 2 Process 3 Process 4 Trace View File Edit Chart Filter mite Timeline 55 100 5 12
74. harts This might be useful in some analysis cases The comparison shows a list of common properties along with the cor responding values Differences are displayed as well The first line always indicates the names of the respective charts see Figure 4 27 51 a GWT 43 INFORMATIONAL CHARTS Vampir Trace View Vampir Large wrf otf Value Master Timeline Function Function MPI Wait Function Group MPI i s ME uS alii Process 11 ess 12 3U Tr THE ess 13 C HH Interval Begin 99 203092 s Interval End 99 28449 5 Duration 81 3979 ms YSU Mi y Figure 4 26 Context View showing context information B of a selected function A Vampir Trace View Vampir Large wrf otf W Edit Chart Filter Window Help gaa m ite Timeline Context View Master Timeline lt Master Timeline Diff X Process 0 Property Value 1 Comparison Value 2 Diff Process 1 Display Master Timeline Master Timeline Process 2 Type Function Function Process 3 Function MPI Wait MPI Wait Function Group MPI MPI Interval Begin 99 346965 s gt 99 203092 5 0 143873 5 Process 5 Interval 99 411091 5 gt 99 28449 5 0 126601 5 Process 6 Duration 64 1259 ms lt 81 3979 ms 17 272 ms Process 7 Process 8 Process 9 Process 1
75. he fist step in order to compare trace files in Vampir is to start a comparison session A comparison session is setup using the Comparison Session Manager This dialog is accessible the main menu entry File New Comparison Session or by clicking the Open Other button in the Vampir start window Figure The Comparison Session Manager depicted in Figure 6 3 holds a list of trace files to be compared in the current session The list is editable at any time using the plus and minus buttons Clicking the OK button will load the respective trace files and open the Comparison View Comparison Session Manager Trace Progress Vampir Comparison A calcTest otf Vampir Comparison B calcTest otf Vampir Comparison C calcTest otf EXE cancel Figure 6 3 Comparison Session Manager listing three trace files for comparison 80 mamaa maana BHHHHH mannana EHEBHHEH BHHBHHHHH DHUBBGHHHH 000006 DBDBBHH mmmEnnu manom 0000 Figure 6 4 shows the resulting Comparison View As indicated by the navigation tool bars at the top of the figure all selected trace files are now included in a single Com parison View instance The files in the view are sharing a coupled zoom The usage of charts an
76. ial vt otf 1 442 5 Timeline 0 25 0 45 0 65 0 85 105 1 25 1 45 0 05 Process 0 5 1 Process 2 Process 3 a Process 0 Process 1 Process 2 Process 3 Eli i nj B Marker View Process Processgroup Time Duration Description Warning M MARMOT Warning Error lll MARMOT Error Process 2 1 191315 5 05 ERROR MPI Type contiguous oldtype i Process 1 1 190995 05 ERROR Type contiguous oldtype i Process 3 1 191929s Os ERROR MPI Type contiguous oldtype i Process 0 1 191657 5 05 ERROR MPI Type contiguous oldtype i arker2 potential vt otf Warnin 7 MARMOT Warning Error ll MARMOT Error Process 0 0 1251845 05 ERROR Send datatype is DAT KI Align Traces at Marker Zoom Between Marker Figure 6 10 Open Marker View First step in order to use markers is to open the Marker View Figure shows a Comparison View with an open Marker View The markers of all open traces are shown combined in one Marker View After a click on one marker in the Marker View the respective marker is highlighted in the Master Timeline and the Process Timeline 86 CHAPTER 6 COMPARISON OF TRACE FILES EHHBHHEH 000006 EEE
77. ication on each compute node In order to achieve consistent event data across all compute nodes clock corrections need to be applied This step is performed after the successful run of the application using the Microsoft tool mpicsync Now the event log files can be converted into OTF files with help of the tool et 12o0t The last necessary step is to copy the generated OTF files from the compute nodes into one shared directory Then this directory includes all files needed by Vampir The application performance can be analyzed now Rank O node myApp exe Run myApp with tracing enabled ___________ gt bg ET Time Sync the ETL logs Convert the ETL logs to OTF mpicsync Copy OTF files to head node zb a cR o etl2otf HEAD NODE share Rank 1 node lt qy o 3 Figure 2 1 MS MPI Tracing Overview The following commands illustrate the procedure described above and show as a prac tical example how to trace an application on the Windows HPC Server 2008 For proper utilization and thus successful tracing the file system of the cluster needs to meet the following prerequisites e share userHome Is the shared user directory throughout the cluster e MS MPI executable myApp exe is available in the shared directory e share userHome Trace is the directory where the OTF files are collected 1 Launch application with tracing enabled use of t 11 option 12
78. icking on the Add button creates a new set of rules and shows the input mask depicted in Figure 5 5 Filter Functions Active Description Add Long MPI Functions Duplicate Short Function Calls Remove Import Export ra Figure 5 4 Function Filter dialog containing a list of rule sets The Function Filter dialog is build on the concept of filter rules The user can define several individual rules The rules are explained in more detail in Chapter 5 4 1 The header of the dialog defines how multiple rules are evaluated One possibility is to build up the filter in a way that combines the filter rules with an and relation To choose this mode must be selected in the combo box in the header of the dialog This means that all rules must evaluate to true in order to produce the filter output The other option is to combine the rules with an or relation To choose this mode any must be selected in the combo box in the header of the dialog In this case any rule must be evaluate to true in order to produce the filter output The examples in Chapter illustrate both modes 65 Filter Functions Show only functions that match any of the following conditions Description Long MPI Functions Name Contains 4 Duration Is greater than 250 Milliseconds yf Apply Figure 5 5 Function Filter dialog showing one rule set
79. ification date Note On loading Vampir always favors settings in the Application Data Folder Default Preferences offers to save preferences of the current trace file as default set tings Then they are used for trace files without settings Another option is to restore the default settings Then the current preferences of the trace file are reverted 91 8 Use Case This chapter explains by example how Vampir can be used to discover performance problems in your code and how to correct them 8 1 Introduction In many cases the Vampir suite has been successfully applied to identify performance bottlenecks and assist their correction To show in which ways the provided toolset can be used to find performance problems in program code one optimization process is illustrated in this chapter The following example is a three part optimization of a weather forecast model including simulation of cloud microphysics Every run of the code has been performed on 100 cores with manual function instrumentation MPI communication instrumentation and recording of the number of L2 cache misses 300 Vampir Trace View Vampir SuccessStory pmp old otf Sle Edit Chart Filter Window Help Timeline Function Summary 3 0s 50 100 150 200 250 All Processes Accumulated Exclusive Time 20 0 same e Process 7 Process 14 Pi Process 21 55 28 58
80. imeline AX 05 25 5 505 75 5 100 5 1255 150 5 1755 2005 Process 0 Metric PAPI_FP_OPS Opacity m 4 7 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 0G 016 026 036 46 1365 Figure 4 12 Master Timeline with active performance data overlay the overlay and thus making underlying functions visible This is particularly useful for first pinpointing performance relevant areas and then directly analyzing the individual identified functions in the Master Timeline The color scale of the performance data overlay is freely customizable Clicking the wrench icon in the overlay control window opens the color scale options dialog The color scale provides three modes Default Highlight and Find Additionally the Cus tom mode allows to manually adapt the color scale to the own preferences Examples This section illustrates the usage of the Performance Radar chart and the Master Time line overlay in a few examples The trace file used for the examples shows a WRF weather forecast code run The timelines show the initialization in the beginning fol lowed by a number of compute iterations Figure J4 14 depicts this trace file The top image shows the pure timelines of the Master Timeline chart the bottom image shows the values of the _ counter superimpose
81. ions is less than the specified number are shown Filtering Functions by Call Path The Call Path filter provides a string input field for a pattern Depending on the options all functions with their related events are shown which satisfy a substring match against the given pattern This filter mode provides two opposing options e Contains The call path must contain a function where the given pattern must occur in the functions name This specifically means that functions that lead to the matched function won t be shown anymore The matched function itself along with its possibly called sub functions is still shown All other call paths that do not contain a matched function are filtered out as well and won t be shown e Does not contain The call path must not contain a function where the given pattern occurs in the function name This specifically means that only functions that lead to the matched function will be shown excluding the matched function itself as well as its possibly called sub functions Call paths that do not contain a matched function are still shown and remain unaffected by the filter Filtering Functions by Call Level Functions can also be filtered by their Call Level This filter mode provides a number input field to select the call level Available options e 15 greater than All functions whose enter event is higher than the specified level are shown e Is less than All functions whose enter event is lower
82. iple measuring points available This option allows to identify outliers by summarizing counters e g PAPILFP_OPS over multiple measuring points e g processes Hence when this option is active multiple measuring points can be selected like in the Process Filter see Section 5 1 for further details The counter for the selected metric is then summarized over all selected measuring points The displayed counter graphs in the chart need then to be read as follows The yellow av erage line in the middle displays the average value e g PAPI FP OPS of all selected measuring points e g processes at a given time The red maximum line shows the highest value that one of the selected measuring points achieved at a given time A click with the left mouse button on any point in the chart reveals its details in the Con text View display Stated are the minimum maximum and average values and the measurement points e g processes that achieved maximum and minimum values at the selected point in time The options dialog is depicted in Figure 4 9 lt is accessible via the context menu under Options allows to enable and disable the display of the graph s line data points and filling It is also possible to enable an average line showing the average value of all data points in the visible area Likewise the charts caption and y axis label can turned on and off The switch Show zero line disables the auto scaling of the y axis for the lower b
83. licking a level by using the fold level buttons next to the function name or by using the provided options in the context menu Functions can be called by many different caller functions what is hardly obvious in the tree representation Therefore a relation view shows all callers and callees of the cur rently selected function in two separated lists shown in the lower area in Figure 4 23 In order to find a certain function by its name Vampir provides a search bar at the bottom of the chart The entered keyword has to be confirmed by pressing the Return key The Previous and Next buttons can be used to flip through the results 4 3 Informational Charts 4 3 1 Function Legend The Function Legend lists all visible function groups of the loaded trace file along with their corresponding color If colors of functions are changed they appear in a tree like fashion under their respec tive function group as well see Figure Clicking on a color box opens a color input dialog which allows to change the color of the respective function or function group 49 a 43 INFORMATIONAL CHARTS Trace View Vampir Small wrf otf Vampir File Edit Chart Filter Window Help gt 1 1 TT Function Legend Function Groups Markers Counters Collectives Messages MO Events PHYS WRF Scheme Default 5 4 IO NETCDF I O Name olor Applicati
84. my toolbar can be dragged and dropped to alternative positions The Charts Toolbar can be disabled with the toolbar s context menu entry Charts Table 3 1 gives an overview of the available performance charts with their correspond ing icons The icons are arranged in three groups divided by small separators The first group represents timeline charts whose zooming states affect all other charts The second group consists of statistical charts providing special information and statistics for a chosen interval Vampir allows multiple instances for charts of these categories The last group comprises of informational charts providing specific textual information or legends Only one instance of an informational chart can be opened at a time 3 6 Properties of the Trace File Vampir provides an info dialog containing important characteristics of the opened trace file This Trace Properties are displayed in the Context View dialog Section 4 3 3 and can be opened the main menu under File Get The information originates from the trace file and includes details such as file name creator or the OTF version 3 Command Line Parameters The Vampir program can be started by clicking on its icon or by calling its program file from the command line as follows vampir parameters file Multiple files can be specified Vampir will open them in separate windows Table gives a brief overview of the parameters that are understo
85. n Short function calls in the Master Timeline may not be visible unless an appropriate zooming level has been reached In other words if the execution time of functions is too short with respect to the available pixel resolution of your computer display Zooming into a shorter time interval is required in order to make them visible Note Other charts are affected by zooming in the timeline displays The interval chosen in a timeline chart such as Master Timeline or Process Timeline also defines the time interval for the calculation of accumulated measurements in the statistical charts Statistical charts like the Function Summary provide zooming of statistic values In these cases zooming does not affect any other chart Zooming is disabled in the Pie 22 BHHHM 5 maana BHHHHH maaa 1 1 1 BUBB mannana BHHHUUH EHHEBHHEH BHBHHHHH BBBHEUHE 0000000 DHUBBHHHH BHEBBHHH 000006 DBDBBHH _ BDBHHBH manom 0000 BHHBH CHAPTER 3 BASICS VAMPIR Vampir Trace View Vampir Large wrf otf W File Edit Chart Filter Window Help ruos 5 2 0 Timeline 705 75 5 80 5 100 5 105 5 1105 Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 SA 83 6 s 98
86. n the elapsed time of that function is added to the MPI function group time The chart gives a condensed view of the execution of the application comparison between the different function groups can be made and dominant function groups can be distinguished easily It is possible to change the information displayed via the context menu entry Set Metric that offers options like Average Exclusive Time Number of Invocations Accumulated Inclusive Time etc 42 5 5 BHH mamaa maana maaa BUBBH mannana BHBHHHHH 0000000 DUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH Vampir Trace View Vampir Large wrf otf File Edit Chart Filter Window Help SEruses mses Br 1 Function Summary Function Summary All Processes Accumulated Exclusive Time per Function Group All Processes Accumulated Exclusive Time per F 1500s 1 0005 500 05 DYN PHYS 495 692 s lm ve 116 338 5 WRF 18 999 5 vo i 7 781 NETCDF i lt 15 i lt 0 15 VT 495 692 DYN 1 713 881 PHYS 980 425 Figure 4 18 Function Summary Note Inclusive means the amount of time spent in a function and all of its subroutines Exclusive means the amount of time spent in just
87. nce Radar for the Master Timeline To fully benefit from this combination the opacity slider of the overlay control window should be used see Figure 4 14 The slider allows to quickly manipulate the opacity of 36 CHAPTER 4 PERFORMANCE DATA VISUALIZATION Vampir Trace View Vampir Large wrf otf Edit Chart Filter Window Help 3 3 5 2 05 25 5 505 75 5 100 5 1255 150 5 1755 2005 Values of Metric PAPI_FP_OPS over Time Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 0G e N e UJ e 4 Figure 4 10 Performance Radar Vampir Trace View Vampir Large wrf otf File Edit Chart Filter Window Help z z z aE E Timeline 05 25s 50s 75s 100s 1255 150 5 1755 2005 Values of Metric PAPI_FP_OPS over Time Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 0G 1G 2G 3G 46 KI gt Figure 4 11 Adjusted value range in color scale 37 GWT 41 TIMELINE CHARTS Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help xime AG 5 T
88. ng by selecting the corresponding value in the spin box Located left of the clustered profile bars is a graphical overview indicating the processes associated to the cluster Moving the cursor over the blue areas in the overview opens a tooltip stating the respective process name 44 5 BHH maana BHHHHH mana n BUBB mannaaa BHHHHUHH mHHEBHHEG BHBHHHHH mmHHEHHH BBBHEUSE DHUBBGHHHH BHEBBHHH 000006 DBDBBHH mmEnnu DBHHBH 00000 0000 BHHBH It is possible to profile only one function or function group or to hide arbitrary functions and function groups from the displayed information To mark the function or function group to be profiled or filtered just click on the associated color representation in the chart The context menu entries Profile of Selected Function Group and Filter Se lected Function Group will then provide the possibility to profile or filter the selected function or function group Using the Process Filter see Section allows you to restrict this view to a set of processes The context menu entry Sort by allows you to order function profiles by Number of Clusters This option is only available if the chart is currently showing clusters Other wise function profiles are sorted automatically by process While profiling one function the menu entry Sort by Value allows to ord
89. number of invocations lie outside the range between 2000 and 15000 are shown i e functions with less than 2000 invocations and functions with more than 15000 invocations Y Show only functions that match any of the following conditions Number of Invocations Is less than 2000 Number of Invocations gt Is greater than 15000 Vampir Trace View Vampir WRF wrf otf W File Edit Chart Filter Window Help lojes Timeline Function Summary Os 10s 20s 30 All Processes Number of Invocations per Function I vP wi occi SER KEE 2576801 send Process2 576809 Process3 malioc Posi ee 1 920 MPI_Gather Process 5 EE 1 920 MODULE PHYSI NDC ADD A2A Process 6 1 680 MODULE BC ZERO GRAD BDY Process7 1 680 MODULE_BC_E DYUPDATE_PH Process 8 HEE F 1 680 MODULE SMAL ADVANCE VV Process 9 EE E eee 1 680 MODULE SMALL EM SUMFLUX Process 10 1 680 MODULE SMAL DVANCE MU T Process11 S E E 1 680 MODULE_SMALL ADVANCE_UV Process 12 S HH gt 1 583 JJEXT NCD SUPP NETCDF ERR Process 13 NR 1 440 MODULE BC RELAX BDYTEND Process 14 1 440 JMODULE BC FLOVV DEP BDY i 1 212 E
90. o be filtered by their duration Duration of a function refers to the time spent in this function from the entry to the exit of the function There are two options available e Is greater than All functions whose duration time is longer than the specified time are shown e Is less than All functions whose duration time is shorter than the specified time are shown Filtering Functions by Number of Invocations The number of invocations of a function can also be used as filter rule This criteria refers to how often a function is executed in an application There are two possible filter rules in this mode Number of Invocations shows functions based on their total number of invocations in the whole application run There are two options available e Is greater than All functions whose number of invocations is greater than the specified number are shown e Is less than All functions whose number of invocations is less than the specified number are shown Number of Invocations per Process shows functions based on their individual num ber of invocations per process Hence if the number of invocations of a function varies over different processes this function might be shown for some processes and filtered for others There are two options available e Is greater than All functions whose number of invocations is greater than the specified number are shown 67 4 Is less than All functions whose number of invocat
91. ocess 10 Process 11 Process 12 Number of processes to be displayed 16 out of 16 Exclude processes with this name part P Set Case Sensitive Use Wildcard Expressions Cancel Use Regular Expressions 1 Figure 5 1 Process Filter Figure 5 1 shows a typical process list in the Filter Processes dialog Processes can be filtered by their Process Hierarchy Communicators Process Group and Hepresen tative Processes ltems to be filtered are arranged in a spreadsheet representation In addition to selecting or deselecting processes by mouse click it is possible to include or exclude processes by name Enter the process name or a part of the name in the input field preceded by Exclude processes with this name part and hit the enter key or click on the set button Inclusion or exclusion of processes can be controlled by tog gling the Exclude and nclude radio button at the beginning of the line The text input field for the process name supports wildcards and regular expressions Corresponding options are shown when clicking on the magnifier icon The regular expressions have been enhanced with a numeric extension to better deal with process numbers which is described in the next paragraph 5 1 1 Numeric Extensions Regular or wildcard expressions have been designed for generic text matching The specification of numeric ranges 15 possible but cumbersome which is why the following extensions have been introduced
92. ocess 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 B 0G 1G 2G 3G 4G 136 Figure 4 16 Highlighted areas with high FLOP rate 41 ME 4 2 STATISTICAL CHARTS Memory Allocation Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help sugi kA U0 50 E Timeline AX 05 255 505 75 5 100 5 125 5 150 5 175 5 200 5 Proceso B Metric MEM APP v Opacity Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 55 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 14 15 25M 50M 75M 100 M 125M 150 175 EN 1305 Figure 4 17 Functions with 160 MB 175 MB allocated memory The performance data overlay can also be used to identify functions with a certain amount of allocated memory Figure 4 17 shows an example Here functions that have between 160 MB and 175 MB memory allocated are highlighted The highlighted range of allocated memory can be easily changed by adjusting the color scale value range 4 2 Statistical Charts 4 2 1 Function Summary The Function Summary chart Figure 4 18 gives an overview of the accumulated time consumption across all function groups and functions For example every time a pro cess calls the MPI_Send functio
93. od by the command line interface help h Show a brief command overview presentation Enable presentation mode i e visualize mouse clicks version V Show program version Table 3 2 Parameters of the Vampir command line interface 26 5 BHH BHHHM BHHHHH maaa mannana BHHHUHH EHHEBHHEH BHBHHHHH DHUBBGHHHH BHEBBHHH 000006 DBDBBHH _ DBHHBH 00000 0000 BHHBH 4 Performance Data Visualization This chapter deals with the different charts that can be used to analyze the behavior of a program and the comparison between different function groups e g MPI and Calculation Communication performance issues are regarded in this chapter as well Various charts address the visualization of data transfers between processes The following sections describe them in detail 4 1 Timeline Charts A very common chart type used in event based performance analysis is the so called timeline chart This chart type graphically presents the chain of events of monitored processes or counters on a horizontal time axis Multiple timeline chart instances can be added to the Trace View window the Chart menu or the Charts toolbar Note To measure the duration between two events in a timeline chart Vampir provides a tool called
94. of the color scale to the desired positions Figure 4 11 depicts the Performance Radar chart shown in Figure 4 10 with a smaller value range of 1 G 3 FLOPS This allows to easily spot areas of high or low performance in the trace file The selected value range can also be dragged to other positions in the color scale A double click with the left mouse button on the color scale resets the selected value range The option Options Color Scale in the context menu of the chart allows to cus tomize the color scale to the own preferences Master Timeline Overlay Mode Figure 4 12 shows an overview of the performance data overlay mode available in the Master Timeline chart The overlay is capable of displaying all metrics available in the Performance Radar chart and the Counter Data Timeline chart It is activated via the chart s context menu under Options Performance Data When the overlay mode is active a control window appears at the top of Master Timeline chart It allows to configure the overlay and to select the displayed performance data metric The selected metric is shown in a color coded fashion like in the Performance Radar chart Figure 4 13 depicts the Master Timeline chart top and the Performance Radar chart bottom both displaying the same performance metric PAPI FP OPS floating point operations per second As can be seen the overlay mode provides the perfor mance data visualization capabilities of the Performa
95. of the trace file the loading process is started by a click on the Open but ton Alternatively a command line invocation is possible The following command line sequence shows an example for a Windows system Other platforms work accordingly C Program Files Vampir Vampir exe trace file To open multiple trace files at once you can give them one after another as command line arguments C Program Files Vampir Vampir exe file 1 file n 14 Me BHHHM BHHHHH mana BUBB mannana BHHHHHUHH EHEBHHEH BHBHHHHH manomano DUBBHHHH BHEBBHHH 000006 DBDBBHH _ ooo 00000 0000 BHHBH 2 GETTING STARTED m If Vampir was associated with otf otf2 files during the installation process it is also possible to start the application by double clicking an otf otf2 file While Vampir is loading the trace file an empty Trace View window with a progress bar at the bottom opens After Vampir loaded the trace data completely a default set of charts will appear The loading process can be interrupted at any time by clicking the Stop amp Show button in the lower right corner of the Trace View The GUI will open and show the information that has been loaded from the trace file so far The basic functionality and navigation elements of the GUI are des
96. on 4 4 The constructed custom metric is depicted in Figure 4 30 Then the performance data overlay is used to show the own metric in the Master Time line Figure The color scale is configured to show only MPI_Wait invocations with a high duration After identification of the areas with the highest duration deep red Zooming into such an area will eventually reveal the respective MPI Nait invoca tions Using the opacity slider Figure 4 32 the individual function occurrences become visible in the Master Timeline 56 BHHHM maana 1 1 HHHHH mannana EHHEBHHEH BHBHHHHH DHUBBHHHH BHEBBHHH 000006 900006 _ manom 0000 BHHBH gt Trace View Vampir Large wrf otf Vampir File Edit Chart Filter Window Help bib ibis o rae Timeline AX 05 255 505 755 100 5 1255 150 5 1755 2005 H X Process 0 Metric Opacity tx Process 1 2 5 3 Process 4 Process 5 5 6 Process 7 Process 8 5 5 9 Process 10 Process 11 i 5 12 Process 13 Process 14 aes
97. ontain the string not case sensitive some where in their name are shown Since only MPI functions start with MPI in their name this filter setting shows all MPI functions and filters the others Show only functions that match any gt of the following conditions Name Contains mpi Vampir Trace View Vampir WRF wrf otf File Edit Chart Filter Window Help loje Timeline 6 5 5 Process 0 Process 1 55 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 1 Figure 5 7 Showing only MPI 70 mamaa mamana maaa mannana EHEBHHHEH BHBHHHHH DHUBBGHHHH BHEBBHHH 000006 DDBBHH mmmnnu DBHHBH 00000 0000 BHHBH Showing only Functions with least 250 ms Duration This example demonstrates the filtering of functions by their duration Here only long function occurrences with a minimum duration time of 250 ms are shown All other functions are filtered Y Show only functions that match any gt of the following conditions Duration 3 S greater 250 Milisecods 2 ence Vampir
98. oolbar B Furthermore a default set of charts is opened automatically after loading has been finished The charts can be divided into three groups timeline statistical and infor mational charts Timeline charts show detailed event based information for arbitrary time intervals while statistical charts reveal accumulated measures which were com puted from the corresponding event data Informational charts provide additional or explanatory information regarding timeline and statistical charts All available charts can be opened with the Charts toolbar which is explained in Chapter 3 5 In the following sections we will explain the basic functions of the Vampir GUI which are generic to all charts If you are already familiar with the fundamentals feel free to skip this chapter The details of the different charts are explained in Chapter 4 17 Ay 3 1 Chart Arrangement Vampir Trace View Vampir Large wrf otf File Edit Chart Filter Window Help Slee i 1 n s i 208 5 miei mS oxi Process Summary Function Summary Similar Processes Accumulated Exclusive Time per Functions Processes Accumulated Exclusive Time per Fun 1 000 s 0s DYN 1980 25 PHYS 493 094 5 IB 93 04 WRF 7 779 s Timeline Function Legend 50 5 100 5 150 5 200 5 H m VO mem m IN NETCDF PHYS 4 Process 0 Process 1 Process 2 Proce
99. or every process independently This is useful for analyzing the balance between processes to reveal bottlenecks For instance finding that one process spends a significantly high time performing the calculations could indicate an unbalanced distribution of work and therefore can slow down the whole application Vampir Trace View Vampir Large wrf otf File Edit Chart Filter Window Help olew 5 7 0 Process Summary Individual Processes Accumulated Exclusive Time per Functions 605 u u u 1005 1205 Process 0 Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 2221235 Figure 4 19 Process Summary The chart calculates statistics based on Number of Invocations Accumulated Inclusive Time or Accumulated Exclusive Time To change between these three modes use the context menu entry Set Metric The displayed colors represent corresponding functions or function groups The con text menu entry Set Functions specifies the set of functions that is displayed in the chart The context menu entry Options Group Functions aggregates functions and displays them as function groups The number of clustered profile bars is based on the chart height by default You can also disable the clustering or set a fixed number of clusters via the context menu entry Clusteri
100. ore P offers the user a maximum of convenience by providing the Opari2 instrumentor as a common infrastructure for a number of analysis tools like Periscope Scalasca Vampir and Tau Amy Amy that obviates the need for multiple repetitions of the instrumentation and thus substan tially reduces the amount of work required It is open for other tools as well Moreover Score P provides the new Open Trace Format Version 2 OTF2 for the tracing data and the new CUBE4 profiling data format which allow a better scaling of the tools with respect to both the run time of the process to be analyzed and the number of cores to be used Score P supports the programming paradigms serial OpenMP MPI and hybrid MPI combined with OpenMP Internally the instrumentation itself will insert special measurement calls into the ap plication code at specific important points events This can be done in an almost automatic way using corresponding features of typical compilers but also semi auto matically or in a fully manual way thus giving the user complete control of the process In general an automatic instrumentation is most convenient for the user This is done by using the scorep command that needs to be prefixed to all the compile and link commands usually employed to build the application Thus an application executable app that is normally generated from the two source files app1 190 and app2 f90 the command gt app
101. ound and enforces a zero line in all situations 34 5 BHH BHHHM i maana HHHHH maaa mannana _ EHHEBHHEG BHBHHHHH maHHEHHH BBBHEUHE DUBBGHHHH BHEBBHHH 000006 900008 mmEnnu manom 0000 BHHBH Select Metric Select by Metric Metrics Measuring Points MEM APP ALLOC v Include Exclude All PAPI PAPI L3 TCM Select by Measuring Point 1 FLOPS in User Defined Function Bandwidth Volume in Transit Latencies Message Data Rate Message Transfer Times Message Volume in Transit Simultaneous O Operations Simultaneous Messages Time Spent in Wait Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 i Summarize multiple measuring points Figure 4 8 Select metric dialog The Counter Data Timeline chart allows to create custom metrics This process is described in Section Created custom metrics become available in the Select Metric dialog 4 1 4 Performance Radar The Performance Radar chart Figure displays counter data and provides the possibility to create custom metrics In contrast to the Counter Data Timeline the Per formance Radar shows one counter for all processes at once The values of
102. ovides a Zoom Toolbar that can be used for zooming and navigation in the trace data It is located in the upper right corner of the View window shown in Figure It is possible to adjust its position via drag and drop The Zoom Toolbar offers an overview and summary of the loaded trace data The currently zoomed area is highlighted as a rectangle within the Zoom Toolbar By dragging of the two boundaries of the highlighted rectangle the horizontal zooming state can be adjusted Note Instead of dragging boundaries it is also possible to use the mouse wheel for zooming Hover the Zoom Toolbar and scroll up and down to zoom in and out respec tively Dragging the zoom area changes the section that is displayed without changing the zoom factor For dragging click into the highlighted zoom area and drag and drop it to the desired position Zooming and dragging within the Zoom Toolbar is illustrated in Figure 3 10 If the user double clicks in the Zoom Toolbar the initial zooming state is reverted Vampir Trace View Vampir Large wrf otf Edit Chart Filter Window Help Figure 3 10 Zooming and Navigation within the Zoom Toolbar A B Zooming in out with the Mouse Wheel C Scrolling by Moving the Highlighted Zoom Area D Zooming by Selecting and Moving a Boundary of the Highlighted Zoom Area The colors represent user defined groups of functions or activities Please note that all charts added to the Trace
103. parallel program runs is to record so called trace log files during runtime The data collection process itself is also re ferred to as tracing a program Unlike profiling the tracing approach records timed application events like function calls and message communication as a combination of timestamp event type and event specific data This creates a stream of events which allows very detailed observations of parallel programs With this technology synchronization and communication patterns of parallel program runs can be traced and analyzed in terms of performance and correctness The analysis is usually carried out in a postmortem step i e after completion of the program It is needless to say am am GWT 1 2 THE OPEN TRACE FORMATS OTF AND 2 that program traces can also be used to calculate the profiles mentioned above Com puting profiles from trace data allows arbitrary time intervals and process groups to be specified This is in contrast to profiles accumulated during runtime 1 2 The Open Trace Formats OTF and OTF2 The Open Trace Formats have been designed as well defined trace formats with open public domain libraries for writing and reading This open specification of the trace information enables analysis and visualization tools like Vampir to operate efficiently at large scale The formats address large applications written in an arbitrary combination of Fortran77 Fortran 90 95 etc C and Local
104. re applications Vampir provides a manageable framework for analysis which en ables developers to quickly display program behavior at any level of detail Detailed performance data obtained from a parallel program execution can be analyzed with a collection of different performance views Intuitive navigation and zooming are the key features of the tool which help to quickly identify inefficient or faulty parts of a pro gram code Vampir implements optimized event analysis algorithms and customizable displays which enable a fast and interactive rendering of very complex performance monitoring data Ultra large data volumes can be analyzed with a parallel version of Vampir which is available on request Vampir has a product history of more than 15 years and is well established on Unix based HPC systems This tool experience is also available for HPC systems that are based on Microsoft Windows HPC Server 2008 1 1 Event based Performance Tracing and Profiling In software analysis the term profiling refers to the creation of tables which summarize the runtime behavior of programs by means of accumulated performance measure ments Its simplest variant lists all program functions in combination with the number of invocations and the time that was consumed This type of profiling is also called inclusive profiling as the time spent in subroutines 15 included in the statistics tation A commonly applied method for analyzing details of
105. rmation of the time span currently selected in the timeline Thus the most time intensive routine of one iteration can be determined by zooming into one or more iterations and having a look at the Function Summary The function with the largest bar takes up the most time In this example Figure 8 2 the MICROPHYSICS routine can be identified as the most costly part of an iteration Therefore it is a good candidate for gaining speedup through serial optimization tech niques Solution In order to get a fine grained view of the MICROPHYSICS routine s inner workings we had to trace the program using full function instrumentation Only then it was possible to inspect and measure subroutines and subsubroutines of MICROPHYSICS This way the most time consuming subroutines have been spotted and could be analyzed for optimization potential 95 a au 8 2 IDENTIFIED PROBLEMS AND SOLUTIONS The review showed that there were a couple of small functions which were called a lot 50 we simply inlined them With Vampir you can determine how often a functions is called by changing the metric of the Function Summary to the number of invocations The second inefficiency we discovered had been invariant calculations being done in side loops So we just moved them in front of the respective loops Figure 8 3 sums up the tuning of the computational imbalance and the serial optimiza tion In the timeline you can see that the duration of the MICROPHY
106. s This allows you to load a trace file quickly by double clicking its master file Subsequently Vampir can be launched by double clicking its icon or by using the command line interface see Chapter 2 3 At the first start Vampir will display instructions for license installation 2 2 Generation of Performance Data The generation of trace log files for the Vampir performance visualization tool requires a working monitoring system to be attached to your parallel program The following software packages provide compatible monitoring systems with built in support for the Vampir performance data file format 2 2 1 Score P Score P is the recommended code instrumentation and run time measurement frame work for Vampir The goal of Score P is to simplify the analysis of the behavior of high performance computing software and to allow the developers of such software to find out where and why performance problems arise where bottlenecks may be expected and where their codes offer room for further improvements with respect to the run time A number of tools have been around to help in this respect but typically each of these tools has only handled a certain subset of the questions of interest A crucial problem in the traditional approach used to be the fact that each analysis tool had its own in strumentation system so the user was commonly forced to repeat the instrumentation procedure if more than one tool was to be employed In this context Sc
107. s labels can also be used for quick process selection in other charts Just use the mouse to drag and drop the respective process label from the Master Timeline to Process Timeline or Function Summary charts Process rows can be re ordered by clicking and dragging the process label at the front of each row If a process has been recorded with subordinated information like threads this information can be hidden and exposed by clicking the black arrow shape in front of the process label or by using the context menu entries Expand All and Collapse All Some function invocations are very short Hence these are not shown in the overall view due to a lack of display pixels A zooming mechanism is provided to inspect a specific time interval in more detail For further information on zooming see Section 3 3 lf zooming has been performed scrolling in horizontal direction is possible with the mouse wheel or the scroll bar at the bottom The context menu entry Group CUDA Streams collapses all streams of a CUDA de vice into one new summarized timeline For the corresponding time interval of each visualized pixel of the new timeline the most prominent activity highest time share across all device streams Is identified That way the new timeline always shows the most important activities and visualizes idle times only when all device streams are unused The Process Timeline resembles the Master Timeline with some differences The charts timeline is divide
108. scard changes 90 CHAPTER 7 CUSTOMIZATION Q General Preferences save preferences Preferences to be saved Displays Process Summary Y Zoom Display Message Profile Counter Display Y ProcessTimeline Display CommunicationMatrix Display Function Summary Call Tree Function Summary v Time Axis Summary Performance Radar Y MasterTimeline Display Appearance Counters Messages Events Function Groups Y Custom Metrics Collectives v Markers Filter Messages Functions O Events Collectives Processes General Layout mamaa maana mannana EHEBHHEH Dodon mmHHEHHH DHUBBGHHHH BHEBBHHH 000006 900006 00000 00000 0000 BHHBH 00000000 0000 Default preferences Preferences to be saved Locally stored preferences Ask to o Apply Q cancel Figure 7 3 Saving policy preferences Usually the settings are stored in the folder of the trace file If the user has no write access to it it is possible to place them alternatively in the Application Data Folder All such stored settings are listed in the tab Locally Stored Preferences with creation and mod
109. ss 3 Process 4 WRF Call Tree All Processes Process 0 Values of Counter PAPI_FP_OPS over Time fa 40G d mE TEN Function Min Inclusi write malloc par free WRF IOINIT WRF IOEXIT M WRF GET DM COMMUNICATOR 13 4 u Ld Figure 3 2 A Custom Chart Arrangement in the Trace View Window The utility of charts can be increased by correlating them and their provided informa tion Vampir supports this mode of operation by allowing to display multiple charts at the same time All timeline charts such as the Master Timeline and the Process line display a sequence of events Those charts are therefore aligned vertically This alignment ensures that the temporal relationship of events is preserved across chart boundaries The user can arrange the placement of the charts according to his preferences by dragging them into the desired position When the left mouse button is pressed while the mouse pointer is located above a placement decoration the layout engine will give visual clues as to where the chart may be moved As soon as the user releases the left mouse button the chart arrangement will be changed according to his intentions The entire procedure is depicted in Figures 3 3 and The layout engine furthermore allows a flexible adjustment of the screen space that is used by a chart Charts of particular interest may get more space in order
110. ssage lines or arrows shows message details like sender process receiver process message length mes sage duration and message tag in the Context View display Vampir Trace View Vampir Large wrf otf Edit Chart Filter Window Help m wie iimu x uu 7 Timeline 05 255 505 75 5 100 5 1255 150 5 175 5 2005 Process 0 Find UNE _____ Process 1 Process 2 Process 3 Process 4 Process 5 Process 6 Process 7 Process 8 Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 Figure 4 4 Search for MPI_Bcast in the Master Timeline Both timeline charts also provides the possibility to search for function and function group occurrences In order to activate the search mode use the context menu and select Find After activation an input field appears at the top of the respective chart A search string can be written in this field and all corresponding function and func tion group occurrences are highlighted in yellow An example search for the function MPI Bcast is depicted in Figure 4 4 Furthermore the Master Timeline also features an overlay mode for performance counter data Figure 4 5 In order to activate the overlay mode use the context menu Options Performance Data When the overlay mode is active a control window ap pears at the top of Master Timeline allows to select the displayed counter data 30 5 8BHH 5 DDDB
111. ssage rate which is the size of the message divided by the duration to decrease accordingly 4 2 5 Summary The O Summary depicted in Figure 4 22 is a statistical chart giving an overview of the input output operations recorded in the trace file All values are represented in a histogram like fashion The text label indicates the group base while the number inside each bar represents the value of the chosen metric The Set Metric sub menu of the context menu is used to switch between the available metrics Number of I O Operations Aggregated I O Transaction Size Aggregated I O Transaction Time and values of Transaction Size Transaction Time or Bandwidth with respect to their selected value type Therefore one has the opportunity to switch between the value types Minimum Average Maximum and Average amp Range via the context menu entry Set Value Note There will be one bar for every occurring metric Furthermore the value type Average amp Range gives a quick and convenient overview and shows minimum max imum and average values at once The minimum and maximum values are shown in an additional smaller bar beneath the main bar indicating the average value The 47 A GWT 42 STATISTICAL CHARTS Vampir Trace View Vampir Large wrf otf W File Edit Chart Filter Window Help Srusexstmzs Summary All Processes Number of I O Operations per File Name 30k 15k 0 k Sum
112. st otf _ C calcTest otf Timeline 16 9650 s 16 9640 s 16 9645 s 16 9655 s 16 9660 s Process 0 Process 1 Process 2 Process 3 Process 0 Process 1 Process 2 Process 3 Be Process 0 Process 1 Process 2 Process 3 16 9578586 5 0 0002094 5 Figure 6 9 Alignment in the Master Timeline After the coarse shifting a finer alignment can be achieved in the Master Timeline or Process Timeline charts Therefore the user needs to zoom into the area to compare Then while keeping the on Mac OS X modifier key pressed the trace can be dragged with the left mouse button in the Master Timeline Figure 6 9 depicts the process of dragging trace C to the compute iterations of trace A and B As shown in the Figure 6 9 although the initialization of trace A took the longest this machine was the fastest in computing the calculations 85 Amy Amy 6 4 Usage of Predefined Markers Markers in traces point to particular places of interest in the trace data These markers can be used to navigate in the trace files For trace file comparison markers are inter esting due to their potential to quickly locate places in large trace data sets With the help of markers it is possible to find the same location in multiple trace files with just a few clicks Comparison View Edit Chart Filter Window Help 9 0 5 L pe special otf otent
113. the loading process 15 GWT File forschung innovation 2 3 STARTING VAMPIR AND LOADING PERFORMANCE DATA Open Subset Help Time To 208 364 5 Processes 16 Seconds el Edit process selection using Representative Processes Include Exclude All Process 0 Y Process 1 Process 2 Y Process Process 4 Y Process 5 Process 6 Process 7 Process 8 x Process 9 Process 10 Process 11 Process 12 Process 13 Process 14 Process 15 Number of processes to be displayed 16 out of 16 processes with this name part Functions Counter Other Events Y Messages IO Events Collective Operations Figure 2 4 Selecting a trace data subset to be loaded ODMR maana BHHHHH mannana n EHHEBHHEG BHBHHHHH DHUBBGHHHH BHEBBHHH DBDHHEBHH 000006 900008 _ BDBHHBH mamom 0000 BHHBH 0000000000000000000 CHAPTERS BASICS 1 3 Basics After loading has been completed the View window title displays the trace file s name as depicted in Figure 3 1 By default the Charts toolbar and the Zoom Toolbar are available Vampir Trace View Vampir Large wrf otf Figure 3 1 Trace View Window with Charts Toolbar A and Zoom T
114. this function The displayed colors represent corresponding functions or function groups The con text menu entry Set Functions specifies the set of functions that is displayed in the chart The context menu entry Options Group Functions aggregates functions and displays them as function groups Shown functions or function groups can be sorted by name or by value via the context menu option Sort By lt is possible to hide functions and function groups from the displayed information with the context menu entry Filter In order to mark the function or function group to be filtered just click on the associated label or color representation in the chart Using the Process Filter see Section 5 1 allows you to restrict this chart to a set of processes As a result only the consumed time of these processes is displayed for each function group or function Instead of using the filter which affects all other displays by hiding processes it is possible to select a single process via Set Process in the context menu of the Function Summary This does not have any effect on other charts The Function Summary can be shown as Histogram a bar chart like in timeline charts or as Pie Chart To switch between these representations use the Set Chart Mode entry of the context menu 43 A 4 2 STATISTICAL CHARTS 4 2 2 Process Summary The Process Summary depicted in Figure 4 19 is similar to the Function Summary but shows the information f
115. ying size of work packages thus varying processing time of this work means waiting time in subsequent synchronization routines This section points out two easy ways to recognize this problem Problem As can be seen in Figurej8 2jeach occurrence of the MICROPHYSICS routine purple color starts at the same time on all processes inside one iteration but takes between 1 7 and 1 3 seconds to finish This imbalance leads to idle time in subsequent syn chronization calls on the processes 1 to 4 because they have to wait for process 0 to finish its work marked parts in Figure 8 2 This is wasted time which could be used for 93 a 8 2 IDENTIFIED PROBLEMS AND SOLUTIONS Vampir Trace View Vampir SuccessStory pmp old otf Edit Chart Filter Window Help MicROPHYSics Figure 8 2 Before Tuning Master Timeline and Function Summary identifying MICRO PHYSICS purple color as predominant and unbalanced Vampir Trace View Vampir SuccessStory pmp tuned otf WwW Sle Edit Chart Filter Window Help 8 x MICROPHYSICS 11 55 Recv Figure 8 3 After Tuning Timeline and Function Summary showing an improvement in communication behavior 94 BHHHM BSBHBHH maana BHHHHH mana mannana BHHHHHUHH EHHEBHHEG BHBHHHHH DHUBBGHHHH

Vampir 8 User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents