Tutorial Details: Code Profiling and Performance Optimization on Itasca
|Date:||Tuesday, November 27, 2012, 01:00 pm - 03:00 pm|
|Instructor(s):||Andrew A Gustafson, MSI, Shuxia Zhang, MSI|
Itasca is an HP Linux cluster with 1,091 HP ProLiant BL280c G6 blade servers and consists of 8,728 compute cores and 24 TiB of main memory. Itasca is a major component of computing resources at MSI. Hence efficient use of Itasca is important both for research groups to better use their service unit allocations and to enhance Itasca's overall performance.
This tutorial will help MSI researchers profile their parallel (OpenMP and MPI) codes. Attendees will learn through hands-on examples how to use the available tools for code profiling. In addition, we will discuss tips on improving the performance of an application.
|Prerequisites:||Knowledge of FORTRAN or C programming language|