Mar 04

Chameleon-1.0.0 is out !

A new release of Chameleon is out.
Please visit our Gitlab page to know more.

Mar 25

Chameleon-0.9.2 is out !

A new release of Chameleon is out.
Please visit our Gitlab page to know more.

Feb 10

Chameleon development is available on Gitlab

Chameleon development is now hosted on the Inria Gitlab server: Git of Chameleon

Chameleon is open to contribution, please read the CONTRIBUTING.md file to get further information.

Jan 15

Chameleon is available in Spack!

Chameleon software requires many dependencies to be installed on your system (BLAS, LAPACK, hwloc, StarPU, etc):
List of dependencies

To help users installing Chameleon we provide some Spack packages to automatically deploy all the software stack by the mean of simple command invocation.
Spack is a python program developed at LLNL and used to help installing complex HPC software stacks.
First learn how to use Spack

Then use the additional Inria Bordeaux Spack packages to install Chameleon

Jun 22

Chameleon-0.9.1 is out !

A new release of Chameleon is out.

Chameleon includes the following features:

  • BLAS 3, LAPACK one-sided and LAPACK norms tile algorithms
  • Support QUARK and StarPU runtime systems
  • Exploitation of homogeneous and heterogeneous platforms through the use of BLAS/LAPACK CPU kernels and cuBLAS/MAGMA CUDA kernels
  • Exploitation of clusters of interconnected nodes with distributed memory (using OpenMPI)

Computation of very large systems with dense matrices on a cluster of nodes is still being experimented and stabilized.

  • POTRF algorithm has been validated on clusters of heterogeneous nodes
  • Validation of other LAPACK one-sided algorithms will come soon

Download Chameleon-0.9.1:

Documentation:

Dec 17

Chameleon-0.9.0 is out !

The first beta release of Chameleon is out.

Chameleon includes the following features:

  • BLAS 3, LAPACK one-sided and LAPACK norms tile algorithms
  • Support QUARK and StarPU runtime systems
  • Exploitation of homogeneous and heterogeneous platforms through the use of BLAS/LAPACK CPU kernels and cuBLAS/MAGMA CUDA kernels
  • Exploitation of clusters of interconnected nodes with distributed memory (using OpenMPI)

Computation of very large systems with dense matrices on a cluster of nodes is still being experimented and stabilized.

Download Chameleon-0.9.0:

Documentation: