Intel Cluster Toolkit 3.2
Performance
The Intel Cluster Toolkit provides advanced performance solutions to speed parallel computing development and deliver high performance on cluster systems.
Intel MPI Library 2.0 delivers high-performance messaging on a variety of hardware fabrics.
- Speed development and QA, facilitating one code stream for multiple fabrics.
- Take advantage of drivers that are tuned for performance relative to best-in-class reported performance.
Intel Math Kernel Library Cluster Edition 8.1 utilises automatic runtime processor detection to execute code that has been specifically optimised for individual Intel processors.
- Benefit from improved ScaLAPACK performance in this version in the areas of scalability, block size insensitivity, and raw performance.
- Realise better performance for the Intel Itanium 2 processor, relative to previous versions, from several Level-2 BLAS functions and for the Intel Pentium 4 processor and Intel Xeon processor in the areas of BLAS, DFTs, and LAPACK.
Intel Trace Analyzer and Collector 6.0 is a powerful tool to understand MPI application behavior and achieve high execution performance.
- Visualise and understand parallel application behavior, and analyse performance of subroutines or code-blocks.
- Learn about communication patterns, parameters, and performance data, including identifying communication hotspots.
Intel MPI Benchmarks 2.3 is a comprehensive set of MPI benchmarks, formerly known as Pallas MPI Benchmarks (PMB).