- Get the best performance on any 64-bit Arm hardware used for scientific computing and HPC
- Rely on accurate results as our libraries are validated with the NAG test suite
- Get continuous performance improvements with frequent releases
- Get support from Arm engineers
- Commercial 64-bit Arm Math Libraries with optimized BLAS, LAPACK and FFT
Arm Performance Libraries are a commercial math library that meets the needs of scientific computing and HPC community on Arm Architecture.
Our algorithms have been written to exploit two major parallelism features available on Arm hardware – vector engine within a single core and large number of cores within the processor. This allows partners to fully exploit the parallelism available in the hardware.
We work with many Arm-based hardware vendors to ensure that Arm Performance Libraries are tuned to a vendor’s micro-architecture, in order to get the best performance out-of-the-box.
Arm Performance Libraries are validated with the NAG suite, thereby ensuring accurate results when you use them in your scientific computing applications.
- BLAS and LAPACK
- FFT with FFTW compatible API
- Available for GCC and Arm Compiler
- 64-bit Arm platforms including Cavium ThunderX2 and Qualcomm Centriq.
- Leading Linux distributions including Red Hat 7.3+, SLES 12+ and Ubuntu 16.04+
Evaluate Arm Tools for Free
Get free access to Arm Forge and Arm Performance Reports and the Arm Allinea Studio. Take advantage of the following:
- One week trial with free support
- Scalable profiling for instructions, memory access, I/O, communication, multi-process and multi-threaded code
- Scalable debugging with uniquely capable C++, C, Fortran and F90 debugging
- Powerful reports for analyzing application performance at scale
- Arm hardware-optimized commercial compiler and performance libraries
Learn how Arm Performance Libraries can provide optimized standard core math libraries for high-performance computing applications on Arm processors
Arm Performance Libraries Resources
Find resources that describe how to develop, deploy and optimize enterprise and scientific HPC (High Performance Computing) applications, including:
- Request HPC tools support
- Supported platforms
- User guides
- Tutorials
- Release history
