This course enable the student to create, analyze, and optimize high-performance applications on clusters of Intel® processor-based systems with the help of Intel® Cluster Tools. This course is designed for parallel MPI application software developers that are interested in optimizing MPI message passing and application performance through the use of Intel® Cluster Tools. Participants will learn what the Intel Cluster Tools can do for them and how to best use its advanced features. The course will cover the optimized Intel® MPI Library with its tuning switches and the advanced Intel® Trace Analyzer and Collector. Intel® Trace Analyzer and Collector analyses the parallel application execution behavior with MPI and provides information critical to understanding and optimizing MPI cluster performance by quickly finding performance bottlenecks with MPI communication, load imbalance and other important cluster computing considerations. Version 7.0.1 now includes trace file comparison, counter data displays, and a unique MPI correctness checking functionality

Prerequisites:
Attendees should be familiar with the basics of MPI message passing programming with C/C++ or FORTRAN under the Linux* operating system. A basic understanding of cluster computing would also be beneficial.

Audience:

• End User
• Channel and ISV Application Software Developers

Format:
Instructor-led (lecture/lab)