MAGMA

Description

From the MAGMA website: The MAGMA project aims to develop a dense linear algebra library similar to LAPACK but for heterogeneous/hybrid architectures, starting with current “Multicore + multi-GPU” systems. The goal is to design linear algebra algorithms and frameworks for hybrid multicore and multi-GPU systems that can enable applications to fully exploit the power that each of the hybrid components offers. Designed to be similar to LAPACK, the MAGMA library allows scientists to easily port their existing software components from LAPACK to MAGMA, to take advantage of the new hybrid architectures. MAGMA users do not have to know CUDA in order to use the library. There are two types of LAPACK-style interfaces. The first one, referred to as the CPU interface, takes the input and produces the result in the CPU’s memory. The second, referred to as the GPU interface, takes the input and produces the result in the GPU’s memory. In both cases, a hybrid CPU/GPU algorithm is used. Also included is MAGMA BLAS, a complementary to CUBLAS routines.

Version

  • 2.24-1

Authorized Users

  • Restricted Research-Use Only

Platforms

  • CIRCE cluster
  • SC cluster

Modules

MAGMA requires the following module file to run:

  • apps/magma/2.24-1

Running MAGMA on CIRCE/SC

The MAGMA user guide is essential to understanding the application and making the most of it. The guide and this page should help you to get started with your simulations. Please refer to the Documentation section for a link to the guide.

  • Note on CIRCE: Make sure to run your jobs from your $WORK directory!
  • Note: Scripts are provided as examples only. Your SLURM executables, tools, and options may vary from the example below. For help on submitting jobs to the queue, see our SLURM User’s Guide.
  • Once loading the above module file, simply refer to the MAGMA library within your code.

Documentation

Home Page, User Guides, and Manuals

Reporting Bugs

Report bugs with MAGMA to the IT Help Desk: rc-help@usf.edu