|
|
|
MAGMA 1.1 Released
|
2011-11-14
|
MAGMA 1.1 is now available. This release extends MAGMA 1.0 by providing the following new functionalities:
- Multicore and multi-GPU support;
- Non-GPU-resident LU, QR, and Cholesky factorizations;
- Tile LU, QR, and Cholesky factorizations and solvers with StarPU dynamic scheduling;
- Matrix inversion routines;
- LAPACK testing.
Read more...
|
|
|
2011-11-11
|
|
|
|
MAGMA 1.0 Released
|
2011-08-25
|
MAGMA 1.0 is now available. This release includes the
MAGMA sources! MAGMA 1.0 is intended for a single CUDA
enabled NVIDIA GPU. It extends version 0.2 by adding support for the Fermi GPUs (see the sample performances for LU, QR, and Cholesky factorizations and LS solvers in complex arithmetic). For more details see the MAGMA 1.0 Rlease Notes and the MAGMA
1.0 presentation.
Included are routines for the following algorithms:
- LU, QR, and Cholesky factorizations in both real and complex
arithmetic (single and double);
- Hessenberg, bidiagonal, and tridiagonal reductions in both real and
complex arithmetic (single and double);
- Linear solvers based on LU, QR, and Cholesky in both real and
complex arithmetic
(single and double);
- Eigen and singular value problem solvers in both real and complex
arithmetic (single and double);
- Generalized Hermitian-definite eigenproblem solvers;
- Mixed-precision iterative refinement solvers based on LU, QR, and
Cholesky in both real and complex arithmetic;
- MAGMA BLAS in real arithmetic (single and double), including gemm,
gemv, symv, and trsm.
See the Software section for a download link.
|
|
MAGMA 1.0 RC5 Released (updated on April 14th, 2011)
|
2011-04-14
|
MAGMA 1.0 RC5 is now available. This release includes the MAGMA sources! MAGMA 1.0 RC5 is intended for a single CUDA enabled NVIDIA GPU. It extends version 0.2 by adding support for the Fermi GPUs (see the sample performances for LU, QR, and Cholesky factorizations and LS solvers in complex arithmetic). For more details see the RC5 Rlease Notes and the MAGMA 1.0 presentation.
Included are routines for the following algorithms:
- LU, QR, and Cholesky factorizations in both real and complex
arithmetic (single and double);
- Hessenberg, bidiagonal, and tridiagonal reductions in both real and complex arithmetic (single and double);
- Linear solvers based on LU, QR, and Cholesky in both real and complex arithmetic
(single and double);
- Eigen and singular value problem solvers in both real and complex arithmetic (single and double);
- Mixed-precision iterative refinement solvers based on LU, QR, and
Cholesky in both real and complex arithmetic;
- MAGMA BLAS in real arithmetic (single and double), including gemm,
gemv, symv, and trsm.
See the Software section for a download link.
|
|
|
2010-11-15
|
|
|
|
|
2010-10-19
|
|
|
|
MAGMA GEMM Sources for Fermi Released
|
2010-08-04
|
The MAGMA BLAS SGEMM and DGEMM sources for Fermi GPUs are now released.
These improved GEMMs, developed by Rajib Nath and Stan Tomov, will be
part of the up-coming MAGMA 0.3 library release and will be included in
CUBLAS 3.2 as well.
The basic algorithm is described in:
Nath, R., Tomov, S., Dongarra, J. "An Improved MAGMA GEMM for Fermi GPUs,"
University of Tennessee Computer Science Technical Report, UT-CS-10-655
(also LAPACK working note 227), July 29, 2010.
http://icl.cs.utk.edu/projectsfiles/magma/pubs/fermi_gemm.pdf
On a C2050 GPU the new DGEMM gets up to 300 GFlop/s (58% of peak) and
the SGEMM up to 645 (63% of peak). On a GTX480 DGEMM gets up to 166 GFlop/s
and SGEMM up to 844 GFlop/s.
The sources are available for download at the Software section of the web site.
|
|
MAGMA tutorial at SAAHPC
|
2010-07-10
|
Accelerating Linear Algebra on Heterogeneous Architectures of Multicore and GPUs using MAGMA and the DPLASMA and StarPU Scheduler, by:
Stanimire Tomov, George Bosilca, and Cédric Augonnet
Learn how to develop numerical software for heterogeneous architectures of Multicore and GPUs through a hybridization methodology that is built on:
- Representing algorithms as collections of tasks and data dependencies, and
- Properly scheduling the tasks' execution over the available multicore and GPU hardware components.
Examples will be given from the Matrix Algebra on GPU and Multicore
Architectures ( MAGMA) project, which aims to develop a new generation of linear algebra libraries that extends the sequential LAPACK-style algorithms for the highly parallel GPU and multicore heterogeneous architectures. As MAGMA has stand-alone hybrid algorithms, it also provides hybrid kernels to be used as building blocks in tile and "communication-avoiding" algorithms that must be efficiently scheduled. You will learn how to use dynamic schedulers to easily express these new algorithms, while at the same time fully use and extract high-performance from heterogeneous systems of multicore and GPUs. In particular, we will consider the DPLASMA and StarPU schedulers.
DPLASMA is related to the Parallel Linear Algebra for Scalable Multi-core Architectures ( PLASMA) project but extends its operation to the distributed memory regime,
while StarPU is a runtime system that is specialized into scheduling tasks onto accelerator-based platforms.
Tutorial presentations:
|
|
MAGMA Library
|
2010-06-23
|
Major chip manufacturers are developing next-generation
microprocessor designs that are heterogeneous/hybrid in nature,
integrating homogeneous x86-based multicore CPU components and GPU
components. The MAGMA (Matrix Algebra on GPU and Multicore
Architectures) project’s goal is to develop innovative linear algebra
algorithms and to incorporate them into a library that is
• similar to LAPACK in functionality, data storage, and interface
but targeting the
• next-generation of highly parallel, and heterogeneous processors.
Read more...
|
|
MAGMA version 0.2 Released
|
2009-11-20
|
MAGMA version 0.2 for 32 and 64-bit Linux is now available. This release is intended for a single
CUDA enabled NVIDIA GPU and includes:
- LU, QR, and Cholesky factorizations in both real and complex arithmetic (single and double);
- LQ and QL factorizations in real arithmetic (single and double);
- Linear solvers based on LU, QR, and Cholesky in real arithmetic (single and double);
- Mixed-precision iterative refinement solvers based on LU, QR, and Cholesky in real arithmetic;
- Reduction to upper Hessenberg form in real arithmetic (single and double);
- MAGMA BLAS in real arithmetic (single and double), including gemm, gemv, symv, and trsm.
See the Software section for a download link.
|
|
U of Tennessee Named CUDA Center of Excellence
|
2009-11-04
|
NVIDIA Corp. today recognized the University of Tennessee, Knoxville's
(UTK's) Innovative Computing Laboratory (ICL) as a CUDA Center of
Excellence, noting its adoption of the CUDA programming model in its
curriculum, as well as its pioneering research into the development of
linear algebra libraries for the high-performance computing community.
Read more...
|
|
Why You Should Touch MAGMA
|
2009-08-19
|
Hiding the details of the multi-core and GP-GPU hardware is a really cool goal. Read the full article at Linux Magazine.
|
|
MAGMA User Forum up and running
|
2009-08-05
|
The MAGMA team is happy to announce the creation of the MAGMA User
Forum. Established to aid with feedback and discussion regarding the
newly released MAGMA software, the forum is now up and running.
|
|
MAGMA gets its 1st user
|
2009-08-05
|
The MAGMA team is happy to announce it has it's first user! In less than 24 hours of releasing MAGMA version 0.1 we got the first feedback.
|
|
MAGMA version 0.1 Released
|
2009-08-04
|
MAGMA version 0.1 for 32 and 64-bit Linux is now available. This release is intended for a single
CUDA enabled NVIDIA GPU and includes the 3 one-sided factorizations - LU, QR, and Cholesky in
single and double precision arithmetic. See the Software section for a download link.
|
|
