Technical Information from The Portland Group
November 2010
CUDA-x86

PGI CUDA-x86: CUDA Programming for
Multi-core CPUs

PGI announced in September its intent to develop CUDA compilers for systems based on the industry-standard x86 architecture. This article provides an overview of the technology, the products and the development timeline.

Also in this issue…

Porting Computational Chemistry Algorithms to GPUs Using PGI Accelerator Directives

This article describes the application of GPU technology to computational chemistry algorithms using the PGI Accelerator Fortran compiler.

Tuning Application Performance Using Hardware Event Counters in the PGPROF Profiler

Application tuning can be challenging. This article describes a couple of ways to use the PGPROF profiler to answer the questions "how can you determine whether performance can be improved, and how best to do it"?

Using MAGMA with PGI Fortran

This article describes application and performance of the MAGMA hybrid multicore GPU-accelerated LAPACK Library from The Innovative Computing Lab at the University of Tennesseee, Knoxville.

Upcoming Events

Concurrent with SC10 in New Orleans, Michael Wolfe will be conducting a one day tutorial on "GPU Programming with CUDA Fortran, CUDA C and the PGI Accelerator Programming Model" Monday, 15 November 2010. The cost is $500US. Attendees may also purchase a license to the PGI Accelerator Workstation product of their choice for just $250 additional.

Join PGI at Supercomputing 2010 in New Orleans, Louisiana the week of November 15. Stop by booth 3229, see what's new from PGI in x86 compilers, CUDA and GPUs and Windows HPC and take away a t-shirt.

Resources

PGInsider back issues

PGI Accelerator C99 & Fortran programming model specification (v1.3 updated Nov. 2010)

CUDA Fortran Programming Reference (v1.4 updated Nov. 2010)

PGI CUDA-x86 Compilers Overview

PGI Accelerator User Forum

Recent News

PGI to Develop Compiler Based on NVIDIA CUDA C Architecture for x86 Platforms

US academics qualify for a free license to PGI Visual Fortran

Next Issue Highlights

The PGI CUDA C/C++ Compiler

Scaling CUDA Fortran Applications for Multiple GPUs

Object Oriented Programming with Fortran 2003

AVX Extensions on Intel and AMD x86 CPUs

MAGMA

MAGMA is an open source GPU-accelerated linear algebra library for Linux-64 developed at the The Innovative Computing Laboratory at the University of Tennessee, Knoxville. Version 1.0 is being released concurrent with Supercomputing 2010.

  • The Portland Group, Inc.
  • Suite 320
  • Two Centerpointe Drive
  • Lake Oswego, OR 97035