PathScale boosts 64-bit Linux clustering

PathScale boosts 64-bit Linux clustering

Summary: Builder: The company has added support for the AMD 64-bit architecture to its compiler suite for Linux clusters

SHARE:
The latest version of PathScale's compiler suite includes support for AMD's 64-bit architecture, Intel's Extended Memory 64 Technology and performance improvements, the company said on Tuesday.

PathScale EKOPath Compiler Suite 2.0 allows you to build C, C++ and Fortran applications for Linux clusters running on 32-bit or 64-bit chips. It includes support for two common 64-bit architectures -- Intel Xeon EM64T and AMD64. It has been tested on SuSE Linux Professional 9.1 and 9.2, SuSE Linux Enterprise Server 8 and 9, Red Hat Enterprise Linux Workstation 3 and 4 and Red Hat Fedora Cores 2 and 3.

The company claims that code compiled using version 2.0 of the product will run up to 20 percent faster than previous PathScale compiler releases and up to 40 percent faster than alternative compiler products. PathScale claims that the compiler suite is fully compatible with the GNU gcc toolkit, the compiler used by many Linux developers. This should mean that no source code changes are required to compile many applications using EKOPath 2.0.

The compiler suite also includes support for OpenMP 2.0 for Fortran, an Application Programming Interface that allows programmers to parallelise applications on a multi-processor machine.

A 30-day trial version of EKOPath 2.0 is available for download from PathScale's Web site.

Topics: Apps, Software Development

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

1 comment
Log in or register to join the discussion
  • Your article says: The company has added support for the AMD 64-bit architecture to its compiler suite for Linux clusters.

    PathScale has always had AMD support. Your sentence should say... The company has added support for Intel EM64T architecture to its compiler suite for Linux clusters
    anonymous