ACML Library at the ITP Opteron Cluster
ACML 4.2.0 ACML documentation |
AMD Core Math Library (ACML) provides a free set of thoroughly optimized and threaded math routines for HPC, scientific, engineering and related compute-intensive applications. ACML is ideal for weather modeling, computational fluid dynamics, financial analysis, oil and gas applications and more.
ACML consists of the following main components:
- A full implementation of Level 1, 2 and 3 Basic Linear Algebra Subroutines (BLAS), with key routines optimized for high performance on AMD Opteron™ processors.
- A full suite of Linear Algebra (LAPACK) routines. As well as taking advantage of the highly-tuned BLAS kernels, a key set of LAPACK routines has been further optimized to achieve considerably higher performance than standard LAPACK implementations.
- A comprehensive suite of Fast Fourier Transforms (FFTs) in both single-, double-, single-complex and double-complex data types.
- Fast scalar, vector, and array math transcendental library routines optimized for high performance on AMD Opteron processors.
- Random Number Generators in both single- and double-precision.
ACML is specifically designed to support multi-threading and other key features of AMD’s next-generation processors. ACML currently supports OpenMP, and future releases will expand upon its support of multi-platform, shared memory multiprocessing. Beginning with the 4.0 release, ACML also features hand-tuned "Barcelona" support for SGEMM and DGEMM matrix multiplication routines, and the CFFT complex-complex Fast Fourier Transforms. More Barcelona optimizations will be available in future ACML releases.
ACML 4.2.0
The ACML 4.2.0 is available in 64 Bit (AMD64/emt64) for two different compilers (GNU, Intel) and for single- and multiprocessor systems.
Setting up the ACML environment
To use the ACML please use the below description. You can use the following entry interactively. If you want to use this module permanently, place this entry below the global definitions in your ".bashrc" file.
GNU Compiler
module load acml/4.2.0_gnu-singleproc
module load acml/4.2.0_gnu-multiproc
INTEL Compiler
module load acml/4.2.0_intel-singleproc
module load acml/4.2.0_intel-multiproc
Please be aware that the INTEL modulefiles needs the INTEL Compiler as a prerequisite. Use, e. g., the following command:
module load intel/10.1.018 acml/4.2.0_intel-multiproc
ACML documentation
You can find the ACML documentation on the AMD Developer Central webpages: