Free online tutorial on performance tools now available

01.19.11 -

A free online tutorial, Introduction to Performance Tools, has been released in CI-Tutor. This tutorial provides an introduction to a number of commonly-used performance tools on the TeraGrid's high-performance computing resources.

The highlighted tools include two Linux utilities: strace, which traces system calls and signals during program execution; and gprof, which can be used to track how functions are using system resources. More complex toolsets include NCSA's PerfSuite, a collection of tools and supporting libraries that provides a higher-level view of application performance, and TAU (Tuning and Analysis Utilities), a suite of tools for performance instrumentation, analysis, and visualization of large-scale parallel computer systems and applications. The course is organized as individual lessons that provide a general overview of each tool, demonstrate how the tool might be run on a TeraGrid computing resource, provide examples of subsequent data analysis, and conclude with self-assessment and pointers to more comprehensive resources regarding each tool.

CI-Tutor offers a number of courses covering topics in high-performance computing, such as multi-core performance, performance tuning, large-scale parallel simulation, scientific visualization, and debugging. The courses are available to everyone and can be accessed by registering and creating a login. To see the courses offered, go to the course catalog at http://www.citutor.org/browse.php. To login and take a course, go to the CI-Tutor homepage at http://www.citutor.org/.

CI-Tutor is hosted by the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign. The effort is supported in part by the National Science Foundation Office of Cyberinfrastructure through the TeraGrid and Blue Waters projects.


National Science Foundation

Blue Waters is supported by the National Science Foundation through awards ACI-0725070 and ACI-1238993.

National Science Foundation

XSEDE is supported by National Science Foundation through award ACI-1053575.