CONTACT   |   QUICK LINKS   |   SITEMAP     search
 
Computer Sciences
image aboutBSC computational earth life computer applications marenostrum image
Computer Sciences
Performance Tools
Home > Computer Sciences > Performance Tools pdf print
 
 
Related links

HPC Europa
 

Performance Tools

stroke

OVERVIEW

Parallel architectures enable to target more complex and ambitious problems each year. But in many cases, the achieved performance is far away from what the theoretical values promised us. Performance analysis tools allow application developers to identify and characterize the inefficiencies that caused a poor performance. We consider that this analysis must be the first step towards the optimization of an application. Optimizing without a previous analysis could be like driving without directions as it could mean wasting efforts improving parts of the code that were not the real performance bottlenecks.

In 1991 we started to work on the development of performance analysis tools, initially only for internal use and distributed since 2000. Flexibility, simplicity and the ability to interact between qualitative and quantitative information were the keys we considered most important in the design of the tools 15 years ago. These features allow us to keep working with the same tools.



OBJECTIVES

The main objective of the team is to be able to use efficiently our tools to solve any performance problem or question we face (directly as users of the tools or trough external users). This means that our tools should be easily adaptable to new platforms, environments or programs, and should be able to scale, extend and evolve in the same way that applications, platforms and programming models do.

We consider that performance analysis is, in some sense, still an art where the experience and intuition of the analyst drives the analysis and determines the quality of the results. For this reason a second objective is to work in the definition of methodologies and procedures that would simplify and facilitate the process of extracting information from the performance data. Our belief is that if performance analysis do not require a special skill or expertise, more people would be interested in applying it.

PROJECTS/AREAS

The set of performance tools that we develop, named CEPBA-Tools, is currently comprised by:
  • Paraver: A very powerful performance visualization and analysis tool based on traces that can be used to analyse any information that is expressed on its input trace format.
  • Dimemas: Simulation tool for the parametric analysis of the behaviour of message-passing applications on a configurable parallel platform.
  • Instrumentation packages: Set of programs and libraries to generate or translate Paraver and Dimemas traces. We have packages for instrumenting different programming models (MPI, OpenMP, mixed) under different platforms (IBM-AIX, Linux, SGI-IRIX, HP-Alpha) and translators from IBM-AIXtrace and LTT formats.
  • Utilities: Set of small programs to process Paraver traces to cut, summarize, translate or accumulate the performance data. They can be used independently but they are also integrated within Paraver.



PEOPLE
arrowBAILEY, DAVID , VISITOR
arrowCASAS, MARC , JUNIOR RESEARCHER
arrowGIMENEZ LUCAS, JUDIT , TOOLS GROUP MANAGER
arrowGONZALEZ GARCIA, JUAN , JUNIOR RESEARCHER
arrowGONZALEZ NAVARRO, PEDRO ANTONIO , RESEARCHER
arrowHUCK, KEVIN , SENIOR RESEARCHER
arrowJOKANOVIC, ANA , PHD STUDENT
arrowLLORT SANCHEZ, GERMAN , RESEARCHER
arrowMARTINEZ HORTELANO, ELOY , RESEARCHER
arrowPEGENAUTE BRESME, XAVIER , RESEARCHER
arrowRODRIGUEZ, GERMAN , RESEARCHER
arrowSERVAT GELABERT, HARALD , RESEARCHER


PUBLICATIONS AND COMMUNICATIONS


Publications

Journals

Jesus Labarta, Germán Rodriguez, and Rosa M. Badia. An Evaluation of Marenostrum Performance. International Journal of High Performance Computing Applications (IJHPCA), Special Issue on "Performance Characterization of the World's Most Powerful Supercomputers, Vol. 22, No. 1, pages 81-96 , February 2008.

Xavier Martorell, Nils Smeds, Robert Walkup, José R. Brunheroto, George Almási, John A. Gunnels, Luiz DeRose, Jesús Labarta, Francesc Escalé, Judit Giménez, Harald Servat, and José E. Moreira. Blue Gene/L performance tools . IBM Journal of Research and Development, vol 49, num 2-3, pp. 407-424 , July 2005.

International Conferences

German Llort, Juan Gonzalez, Harald Servat, Judit Gimenez, and Jesus Labarta. On-line Detection of Large-scale Parallel Application's Structure. 24th IEEE International Parallel and Distributed Processing Symposium (IPDPS'2010), , May 2010.

J. Gonzalez, J. Gimenez and J. Labarta. Automatic evaluation of the computation structure of parallel applications. International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT’09), , December 2009.

J. Gonzalez, J. Gimenez and J. Labarta. Automatic Detection of Parallel Applications Computation Phases. IEEE International Symposium on Parallel & Distributed Processing (IPDPS’09), , May 2009.

Marc Casas, Rosa M. Badia, Jesus Labarta. Prediction of Behavior of MPI Applications. IEEE Cluster 2008,
, September 2008.

Marc Casas, Rosa M. Badia, Jesus Labarta. Automatic analysis of speedup of MPI applications. International Conference on Supercomputing 2008 ( ICS 2008),
, June 2008.

Marc Casas, Rosa M. Badia, Jes�s Labarta. Automatic Phase Detection of MPI Applications. proceedings of ParCo 2007,
, September 2007.

Marc Casas, Rosa M. Badia, Jesús Labarta. Automatic Structure Extraction of MPI Applications Tracefiles. Europar 2007,
, August 2007.

Jesús Labarta, Judit Gimenez, E. Martínez, P. González, Harald Servat, G. Llort, Xavier Aguilar. Scalability of Tracing and Visualization Tools. ParCo 2005, , September 2005.

G. Rodriguez, R.M. Badia, and J. Labarta. Generation of Simple Analytical Models for Message Passing Applications. 10th International Euro-Par Conference, Pisa, Italy, August 2004.

Gabriele Jost, Jesús Labarta, Judit Gimenez. Paramedir: A Tool for Programmable Performance Analysis. International Conference on Computational Science (ICCS'04), , June 2004.

Allan Snavely, Xiaofeng Gao, C. Lee, Laura Carrington, Nicole Wolter, Jesús Labarta, Judit Gimenez, P. Jones. Performance Modeling of HPC Applications. ParCo 2003, , September 2003.

Rosa M. Badia, Francesc Escale, Edgar Gabriel , Judit Gimenez, Rainer Keller, Jesús Labarta, Matthias S. Müller . Performance Prediction in a Grid Environment. 1st European Across Grids Conference, Santiago de Compostela
 , July 2003.

Gabriele Jost, Haoquian Jin, Jesus Labarta and Judit Gimenez. Interfacing Computer Aided Parallelization and Performance Analysis. International Conference on Computational Science (ICCS'03), Melbourne, Australia, June 2003.

Gabriele Jost, Haoquian Jin, Jesus Labarta, Judit Gimenez and Jordi Caubet. Performance Analysis of Multilevel Parallel Applications on Shared Memory Architectures. International Parallel and Distributed Processing Symposium (IPDPS), Nice, France, April 2003.

A. Snavely, L. Carrington, N. Wolter, J. Labarta, R. M. Badia and A. Purkayastha. A Framework for Performance Modeling and Prediction. SuperComputing ' 02, Denver, USA, November 2002.

Felix Freitag, Jordi Caubet and Jesus Labarta. A Trace-Scaling Agent for Parallel Application Tracing. IEEE International Conference on Tools with Artificial Intelligence (ICTAI), pp. 494-499. Washington D.C., USA, November 2002.

Felix Freitag, Jordi Caubet and Jesus Labarta. On the Scalability of Tracing Mechanisms. 8th International Euro-Par Conference, pp. 97-104. Paderborn, Germany, August 2002.

Jordi Guitart, Jordi Torres, Eduard Ayguadé and Mark Bull. Performance Analysis of Parallel Java Applications on Shared-memory Systems . 30th International Conference on Parallel Processing (ICPP'01), Valencia, Spain, September 2001.

F. Freitag, J. Corbalan, J. Labarta. A Dynamic Periodicity Detector: Application to Speedup Computation. International Parallel and Distributed Processing Symposium (IPDPS), San Francisco , April 2001.

Sergi Girona, Jesús Labarta and Rosa M. Badia. Validation of Dimemas communication model for MPI collective operations. EuroPVM/MPI'2000, Balatonfüred, Lake Balaton, Hungary , September 2000.

Jordi Guitart, Jordi Torres, Eduard Ayguadé, José Oliver and Jesús Labarta. Java Instrumentation Suite: Accurate Analysis of Java Threaded Applications . 2nd Annual Workshop on Java for High Performance Computing (part of the 14th ACM International Conference on Supercomputing ICS'00), pp. 15-25, Santa Fe, New Mexico, USA, May 2000.

Sergi Girona and Jesús Labarta. Sensitivity of Performance Prediction of Message Passing Programs. 1999 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'99), Monte Carlo Resort, Las Vegas, Nevada, USA , July 1999.

Jesús Labarta, Sergi Girona, Vincent Pillet, Toni Cortés, Luis Gregoris. DiP : A Parallel Program Development Environment. 2nd International EuroPar Conference (EuroPar 96), Lyon (France)
 , August 1996.

A. Snavely, L. Carrington, N. Wolter, and J. Labarta. A Framework for Practical Applications Performance Modeling. Parallel Computing 2003 (ParCo'2003), Dresden, Germany, November 1999.

Workshops

Harald Servat, Germán Llort, Judit Giménez, Jesús Labarta. Detailed performance analysis using coarse grain sampling. PROPER 2009, , August 2009.

Gabriele Jost, Robert Chun, Haoqiang Jin, Jesús Labarta, Judit Gimenez. An Expert Assistant for Computer Aided Parallelization. PARA 2004 - Workshop on State-of-the-art in scientific computing, , June 2004.

Gabriele Jost, Jesús Labarta, Judit Gimenez. What Multilevel Parallel Programs Do When You Are Not Watching: A Performance Analysis Case Study Comparing MPI/OpenMP, MLP, and Nested OpenMP. WOMPAT 2004, , May 2004.

Rosa M. Badia, Germán Rodriguez and Jesus Labarta. Deriving analytical models from a limited number of runs. Minisymposium on Performance Analysis ParCo, , December 2003.

R. M. Badia, Jesus Labarta, Judit Giménez and Francesc Escalé. Dimemas: Predicting MPI applications behaviour in Grid environments. Workshop on Grid Applications and Programming Tools (GGF8), , June 2003.

Jordi Caubet, Judit Gimenez, Jesús Labarta, Luiz DeRose and Jeffrey Vetter. A Dynamic Tracing Mechanism for Performance Analysis of OpenMP Applications. Workshop on OpenMP Application and Tools (WOMPAT 01), Purdue (USA) , July 2001.

 
  top
link_top
  Barcelona Supercomputing Center, 2010 - Legal Notice
 
link_top