June 2, 2003ICCS Performance Instrumentation and Measurement for Terascale Systems Jack Dongarra, Shirley Moore, Philip Mucci University of Tennessee Sameer Shende, and Allen Malony University of Oregon

June 2, 2003ICCS Requirements for Terascale Systems Performance framework must support a wide range of –Performance problems (e.g., single-node performance, synchronization and communication overhead, load balancing) –Performance evaluation methods (e.g., parameter-based modeling, bottleneck detection and diagnosis) –Programming environments (e.g., multiprocess and /or multithreaded, parallel and distributed, large-scale) Need for flexible and extensible performance observation framework

June 2, 2003ICCS Research Problems Appropriate level and location for implementing instrumentation and measurement How to make the framework modular and extensible Appropriate compromise between level of detail/accuracy and instrumentation cost

June 2, 2003ICCS Instrumentation Strategies Source code instrumentation –Manual or using preprocessor Library level instrumentation –e.g., MPI and OpenMP profiling interfaces Binary rewriting –E.g., Pixie, ATOM, EEL, PAT Dynamic instrumentation –DyninstAPI

June 2, 2003ICCS Types of Measurements Profiling Tracing Real-time Analysis

June 2, 2003ICCS Profiling Recording of summary information during execution –inclusive, exclusive time, # calls, hardware statistics, … Reflects performance behavior of program entities –functions, loops, basic blocks –user-defined “semantic” entities Very good for low-cost performance assessment Helps to expose performance bottlenecks and hotspots Implemented through –sampling: periodic OS interrupts or hardware counter traps –instrumentation: direct insertion of measurement code

June 2, 2003ICCS Tracing –Recording of information about significant points (events) during program execution entering/exiting code region (function, loop, block, …) thread/process interactions (e.g., send/receive message) –Save information in event record timestamp CPU identifier, thread identifier Event type and event-specific information – Event trace is a time-sequenced stream of event records – Can be used to reconstruct dynamic program behavior –Typically requires code instrumentation

June 2, 2003ICCS Real-time Analysis Allows evaluation of program performance during execution Examples –Paradyn –Autopilot –Perfometer

June 2, 2003ICCS TAU Performance System Architecture EPILOG Paraver

June 2, 2003ICCS TAU Instrumentation Manually using TAU instrumentation API Automatically using –Program Database Toolkit (PDT) –MPI profiling library –Opari OpenMP rewriting tool Uses PAPI to access hardware counter data

June 2, 2003ICCS Program Database Toolkit (PDT) Program code analysis framework for developing source-based tools High-level interface to source code information Integrated toolkit for source code parsing, database creation, and database query –commercial grade front end parsers –portable IL analyzer, database format, and access API –open software approach for tool development Targets and integrates multiple source languages Used in TAU to build automated performance instrumentation tools

June 2, 2003ICCS PDT Components Language front end –Edison Design Group (EDG): C, C++ –Mutek Solutions Ltd.: F77, F90 –creates an intermediate-language (IL) tree IL Analyzer –processes the intermediate language (IL) tree –creates “program database” (PDB) formatted file DUCTAPE (Bernd Mohr, ZAM, Germany) –C++ program Database Utilities and Conversion Tools APplication Environment –processes and merges PDB files –C++ library to access the PDB for PDT applications

June 2, 2003ICCS TAU Analysis Profile analysis –pprof parallel profiler with text-based display –Racy / jRacy graphical interface to pprof (Tcl/Tk) jRacy is a Java implementation of Racy –ParaProf Next-generation parallel profile analysis and display Trace analysis and visualization –Trace merging and clock adjustment (if necessary) –Trace format conversion (ALOG, SDDF, Vampir) –Vampir (Pallas) trace visualization –Paraver (CEPBA) trace visualization

June 2, 2003ICCS TAU Pprof Display

June 2, 2003ICCS jracy (NAS Parallel Benchmark – LU) n: node c: context t: thread Global profiles Individual profile Routine profile across all nodes

June 2, 2003ICCS ParaProf Scalable Profiler Re-implementation of jRacy tool Target flexibility in profile input source –Profile files, performance database, online Target scalability in profile size and display –Will include three-dimensional display support Provide more robust analysis and extension –Derived performance statistics

June 2, 2003ICCS ParaProf Architecture

June 2, 2003ICCS Processor Profile (SAMRAI)

June 2, 2003ICCS Three-dimensional Profile Displays 500-processor Uintah execution (University of Utah)

June 2, 2003ICCS Overview of PAPI Performance Application Programming Interface The purpose of the PAPI project is to design, standardize and implement a portable and efficient API to access the hardware performance monitor counters found on most modern microprocessors. Parallel Tools Consortium project References implementations for all major HPC platforms Installed and in use at major government labs, academic sites Becoming de facto industry standard Incorporated into many performance analysis tools – e.g., HPCView,SvPablo, TAU, Vampir, Vprof

June 2, 2003ICCS PAPI Counter Interfaces PAPI provides three interfaces to the underlying counter hardware: 1.The low level interface provides functions for setting options, accessing native events, callback on counter overflow, etc. 2.The high level interface simply provides the ability to start, stop and read the counters for a specified list of events. 3.Graphical tools to visualize information.

June 2, 2003ICCS PAPI Implementation Tools PAPI Low Level PAPI High Level Hardware Performance Counter Operating System Kernel Extension PAPI Machine Dependent Substrate Machine Specific Layer Portable Layer

June 2, 2003ICCS PAPI Preset Events Proposed standard set of events deemed most relevant for application performance tuning Defined in papiStdEventDefs.h Mapped to native events on a given platform –Run tests/avail to see list of PAPI preset events available on a platform

June 2, 2003ICCS Scalability of PAPI Instrumentation Overhead of library calls to read counters can be excessive. Statistical sampling can reduce overhead. PAPI substrate for Alpha Tru64 UNIX –Built on top of DADD/DCPI (Dynamic Access to DCPI Data/Digital Continuous Profiling Interface) –Sampling approach supported in hardware –1-2% overhead compared to 30% on other platforms Using sampling and hardware profiling support on Itanium/Itanium2

June 2, 2003ICCS Vampir v3.x: Hardware Counter Data Counter Timeline Display

June 2, 2003ICCS What is DynaProf? A portable tool to instrument a running executable with Probes that monitor application performance. Simple command line interface. Open Source Software A work in progress…

June 2, 2003ICCS DynaProf Methodology Make collection of run-time performance data easy by: –Avoiding instrumentation and recompilation –Using the same tool with different probes –Providing useful and meaningful probe data –Providing different kinds of probes –Allowing custom probes

June 2, 2003ICCS Why the “Dyna”? Instrumentation is selectively inserted directly into the program’s address space. Why is this a better way? –No perturbation of compiler optimizations –Complete language independence –Multiple Insert/Remove instrumentation cycles

June 2, 2003ICCS DynaProf Design GUI, command line & script driven user interface Uses GNU readline for command line editing and command completion. Instrumentation is done using: –Dyninst on Linux, Solaris and IRIX –DPCL on AIX

June 2, 2003ICCS DynaProf Commands load list [module pattern] use [probe args] instr module [probe args] instr function [probe args] stop continue run [args] Info unload

June 2, 2003ICCS DynaProf Probe Design Probes provided with distribution –Wallclock probe –PAPI probe –Perfometer probe Can be written in any compiled language Probes export 3 functions with a standardized interface. Easy to roll your own (<1day) Supports separate probes for MPI/OpenMP/Pthreads

June 2, 2003ICCS Future development GUI development Additional probes –Perfex probe –Vprof probe –TAU probe Better support for parallel applications

June 2, 2003ICCS Perfometer Application is instrumented with PAPI –call perfometer() –call mark_perfometer(int color, char *label) Application is started. At the call to perfometer, signal handler and a timer are set up to collect and send the information to a Java applet containing the graphical view. Sections of code that are of interest can be designated with specific colors Real-time display or trace file

June 2, 2003ICCS Perfometer Display Machine info Process & Real time Flop/s Rate Flop/s Min/Max

June 2, 2003ICCS Perfometer Parallel Interface

June 2, 2003ICCS Conclusions TAU and PAPI projects are addressing important research problems involved in constructing a flexible and extensible performance observation framework. Widespread adoption of PAPI demonstrates the value of a portable interface to low-level architecture-specific performance monitoring hardware. TAU framework provides flexible mechanisms for instrumentation and measurement.

June 2, 2003ICCS Conclusions (cont.) Terascale systems require scalable low-overhead means of collecting performance data. –Statistical sampling support in PAPI –TAU filtering and feedback schemes for focusing instrumentation –Real-time monitoring capabilities (Dynaprof, Perfometer) PAPI and TAU infrastructure is designed for interoperability, flexibility, and extensibility.

June 2, 2003ICCS More Information TAU ( PDT ( PAPI ( OPARI (