High Performance Computing
High End Visualization
About CCR Contact Us Events Facilities Faculty Project Highlights History Job Opportunities News Partners Staff
Bioinformatics Consulting Services Grid Computing Visualization
Accounts Application Software Consulting Getting Started Hardware Resources Frequently Asked Questions Machine Status MyStats Overview Request Help Storage Resources Training/Courses Web Portals
Access Grid Training Outreach
Events Logos Media News Photo Album Videos
Contact Information Directions
  Dashboard > CCR Web > Getting Started > Profiling
Log In   View a printable version of the current page.
Profiling

Profiling tools are available that allow users to profile their application/code to determine the performance of the component routines. This can be particularly useful when designing a parallel version of a sequential code.

GNU Profiler: gprof

PGI Profiler: pgprof

  • This is a graphical performance profiler for parallel MPI and OpenMP applications.
  • Load one of the PGI modules to set the PATH for pgprof.
    • Show available PGI modules: module avail PGI
    • Load the default PGI module: module load pgi
  • pgprof guide

Performance Application Programming Interface: PAPI

  • This is an API (Application Programming Interface) for accessing hardware performance counters.
  • These interface is a library.
  • Load the PAPI module to set the paths for papi: module load papi
  • PAPI user guide
  • PAPI programmer's reference

Lightweight, Scalable MPI Profiling: mpip

  • The mpip profiler gathers statistical information about MPI functions.
  • This is a link-time library.
  • Load one of the mpip modules.
    • Show available PAPI modules: module avail mpip
  • mpip documentation
Center for Computational Research - University at Buffalo - State University of New York