1. Generating Proxy Components using PDT
Sameer Shende
Department of Computer and Information Science
NeuroInformatics Center
University of Oregon

2. Outline Overview of the TAU and PDT projects Proxy Component
Auto-generation of proxies
Applications
Concluding remarks
Apr. 15, 2004

3. TAU Performance System Framework
Tuning and Analysis Utilities
Performance system framework for scalable parallel and distributed high-performance computing
Targets a general complex system computation model
nodes / contexts / threads
Multi-level: system / software / parallelism
Measurement and analysis abstraction
Integrated toolkit for performance instrumentation, measurement, analysis, and visualization
Portable, configurable performance profiling/tracing facility
Open software approach
University of Oregon, LANL, FZJ Germany
Apr. 15, 2004

4. TAU Performance System Architecture
Paraver
EPILOG
Apr. 15, 2004

5. TAU’s Paraprof Profile Browser (ESMF Data)

6. Callpath Profiling in TAU
Apr. 15, 2004

7. Program Database Toolkit
Component
source/ Library
C / C++
parser
Fortran 77/90/95
IL analyzer
Program
Database
Files IL
DUCTAPE
tau_pg
SILOON
CHASM
TAU_instr
Proxy
Application
component glue
C++ / F90
interoperability
Automatic source
instrumentation
Apr. 15, 2004

8. Program Database Toolkit (PDT)
Program code analysis framework for developing source-based tools for C99, C++ and F90
High-level interface to source code information
Widely portable:
IBM (AIX, Linux Power4), SGI, Compaq, HP, Sun, Linux clusters,Windows, Apple, Hitachi, Cray X1,T3E, RedStorm...
Integrated toolkit for source code parsing, database creation, and database query
commercial grade front end parsers
EDG for C99/C++
Mutek Solutions for F90
Cleanscape Flint Parser for F77/F90/F95
Intel/KAI C++ headers for std. C++ library distributed with PDT
portable IL analyzer, database format, and access API
open software approach for tool development
Target and integrate multiple source languages
Used in TAU to build automated performance instrumentation tools
Used in CHASM, XMLGEN, Component method signature extraction,…
Apr. 15, 2004

9. CCA Performance Observation Component
Design measurement port and measurement interfaces
Timer
start/stop
set name/type/group
Control
enable/disable groups
Query
get timer names
metrics, counters, dump to disk
Event
user-defined events
Apr. 15, 2004

10. CCA C++ (CCAFFEINE) Performance Interface
namespace performance {
namespace ccaports { class Measurement: public virtual classic::gov::cca::Port { public: virtual ~ Measurement (){} /* Create a Timer interface */ virtual performance::Timer* createTimer(void) = 0; virtual performance::Timer* createTimer(string name) = 0; virtual performance::Timer* createTimer(string name, string type) = 0; virtual performance::Timer* createTimer(string name, string type,
string group) = 0; /* Create a Query interface */ virtual performance::Query* createQuery(void) = 0; /* Create a user-defined Event interface */ virtual performance::Event* createEvent(void) = 0; virtual performance::Event* createEvent(string name) = 0; /* Create a Control interface for selectively enabling and disabling * the instrumentation based on groups */ virtual performance::Control* createControl(void) = 0; }; }
Measurement port
Measurement interfaces
Apr. 15, 2004

11. CCA Timer Interface Declaration
namespace performance {
class Timer { public: virtual ~Timer() {}
/* Implement methods in a derived class to provide functionality */
/* Start and stop the Timer */ virtual void start(void) = 0;
virtual void stop(void) = 0; /* Set name and type for Timer */
virtual void setName(string name) = 0;
virtual string getName(void) = 0; virtual void setType(string name) = 0; virtual string getType(void) = 0;
/* Set the group name and group type associated with the Timer */ virtual void setGroupName(string name) = 0; virtual string getGroupName(void) = 0;
virtual void setGroupId(unsigned long group ) = 0; virtual unsigned long getGroupId(void) = 0; }; }
Timer interface methods
Apr. 15, 2004

12. Use of Observation Component in CCA Example
#include "ports/Measurement_CCA.h" ...
double MonteCarloIntegrator::integrate(double lowBound, double upBound,
int count) { classic::gov::cca::Port * port; double sum = 0.0; // Get Measurement port port = frameworkServices->getPort ("MeasurementPort"); if (port) measurement_m = dynamic_cast < performance::ccaports::Measurement * >(port); if (measurement_m == 0){ cerr << "Connected to something other than a Measurement port"; return -1;
} static performance::Timer* t = measurement_m->createTimer( string("IntegrateTimer")); t->start(); for (int i = 0; i < count; i++) { double x = random_m->getRandomNumber (); sum = sum + function_m->evaluate (x); } t->stop(); }
Apr. 15, 2004

13. Measurement Port Implementation
Use of Measurement port (i.e., instrumentation)
independent of choice of measurement tool
independent of choice of measurement type
TAU performance observability component
Implements the Measurement port
Implements Timer, Control, Query, Control
Port can be registered with the CCAFEINE framework
Components instrument to generic Measurement port
Runtime selection of TAU component during execution
TauMeasurement_CCA port implementation uses a specific TAU library for choice of measurement type
Apr. 15, 2004

14. … What’s Going On Here? application component application performance
Two instrumentation paths using TAU API
runtime TAU
performance data
TAU API
application
component
performance
other API …
TAU API
Alternative implementations of performance component
Two query and control paths using TAU API
Apr. 15, 2004

15. IntegratorProxy Component
Interpose a proxy component for each port
Inside the proxy, track caller/callee invocations, timings
Automate the process of proxy component creation
Using PDT for static analysis of components Go
Driver
IntegratorPort
MidpointIntegrator
IntegratorPort
IntegratorProxy Component
IntegratorPortUses
IntegratorPortProvides
MeasurementPort
Performance
Apr. 15, 2004

16. TAU’s Proxy Generator for Classic C++ Interface
Proxy generator arguments:
-p <port name> -t <type> -c <component> -d <PDB file> -o <output file> -f <selective instrumentation file> -x <Component tag> e.g.,
% tau_pg -c integrators::ccaports::Integrator -t integrators.ccaports.Integrator -p IntegratorPort -d ParallelIntegrator_CCA.pdb -o Proxy.cc -h ports/Integrator_CCA.h -f select.dat –x ParallelInt
Creating PDB file:
% cxxparse <file.cpp> -I<dir> -D<flags> creates file.pdb.
% pdbmerge -o merged.pdb file1.pdb file2.pdb …
Apr. 15, 2004

17. Selective Instrumentation
Exclude or include list of routines
% tau_pg … -f select.dat
% cat select.dat
# Selective instrumentation: Specify an exclude/include list of routines/files.
BEGIN_EXCLUDE_LIST
void quicksort(int *, int, int)
void sort_5elements(int *)
void interchange(int *, int *)
END_EXCLUDE_LIST
# or use BEGIN_INCLUDE_LIST END_INCLUDE_LIST to bracket the event names
# Instruments routines in Main.cpp, Foo?.c and *.C files only
# Use BEGIN_[FILE]_INCLUDE_LIST with END_[FILE]_INCLUDE_LIST
Apr. 15, 2004

18. Flame Reaction-Diffusion Demonstration
CCAFFEINE
Apr. 15, 2004

19. CFRFS Profiles using Proxy Generator
NODE 0;CONTEXT 0;THREAD 0:
%Time Exclusive Inclusive #Call #Subrs Inclusive Name
msec total msec usec/call
, : int main(int, char **)
, : driver_proxy::go()
, , rk2_proxy::Advance()
, , ee_proxy::Regrid()
, , flux_proxy::compute()
, , sc_proxy::compute()
, , efm_proxy::compute()
, , grace_proxy::GC_Synch()
, grace_proxy::GC_regrid_above_threshold()
icc_proxy::restrict()
icc_proxy::prolong()
MPI_Init()
TAU_GET_FUNCTION_VALUES()
MPI_Isend()
c_proxy::compute()
stats_proxy::compute()
MPI_Waitsome()
MPI_Comm_dup()
MPI_Allreduce()
rk2_proxy::GetStableTimestep()
compute void ( VectorFieldVariable *)
bc_proxy::compute()
MPI_Barrier()
Apr. 15, 2004

20. Performance Modeling
Use MasterMind Component [IPDPS’04] with Measurement Component to track each argument invocation
Proxy for Mastermind component currently tracks all methods.
Develop performance models based on measurements
Specify performance models to a model evaluator library (being developed at UO/Sandia [Nick Trebon, J. Ray]) for evaluating performance models of component ensembles
Specifying performance models research
Performance Database ties in historical performance data
Apr. 15, 2004

21. TAU Performance Database Framework
Performance analysis programs
Performance
data description
Raw performance data
Other tools
PerfDML
translators
Performance analysis
and query toolkit
. . .
ORDB
PostgreSQL
PerfDB
profile data only
XML representation
project / experiment / trial
Apr. 15, 2004

22. Proxy Generator for other Applications
PDT based proxy component for:
QoS tracking [Boyana, ANL]
Debugging Port Monitor (tracks arguments)
SCIRun2 Perfume components [Venkat, U. Utah]
Exploring Babel for auto-generation of proxies:
Direct SIDL to proxy code generation
Generating client component interface in C++, using PDT for generating proxies
Apr. 15, 2004

23. Concluding Remarks
Complex component systems pose challenging performance analysis problems that require robust methodologies and tools
Automating Instrumentation of Component Software
Performance Measurement
Performance Prediction
Debugging
Performance-aware (QoS) intelligent components
Performance engineered components
Performance knowledge, observation, query and control
Apr. 15, 2004

24. Support Acknowledgement
TAU and PDT support:
Department of Energy (DOE)
DOE MICS contracts
DOE ASCI Level 3 (LANL, LLNL)
U. of Utah DOE ASCI Level 1 subcontract
NSF National Young Investigator (NYI) award