Company

gridMathematica 2.1 Capitalizes on Large-Scale Improvements in Mathematica 5.2

August 2, 2005--Wolfram Research's gridMathematica solution for high-performance computing has been updated to include Mathematica 5.2, the 64-bit and multicore release that vastly increases memory and CPU availability.

Highlights of gridMathematica 2.1 for large-scale computation include:

  • All-platform support for 64-bit addressing
  • Multicore support on major platforms
  • Multithreaded numerical linear algebra
  • 64-bit-enhanced arbitrary-precision numerics
  • Vector-based performance enhancements
  • MathematicaMark 5.2 benchmark now covering grids and clusters

With these added features, gridMathematica is now an even better solution for handling the most demanding large-scale problems and simulations. Universal 64-bit technology means that there's effectively no memory barrier--theoretically addressing up to 16 exabytes--so gridMathematica is limited only by the operating system and hardware. Version 2.1 also supports automatically threaded numerical linear algebra, enabling linear algebra operations to automatically run in parallel on all available processor cores.

For further system optimization, the new MathematicaMark 5.2 benchmark tool provides for scalability of Mathematica benchmark scores, allowing benchmarking of SMP machines, clusters, and grids. Quick benchmark reports show how a grid or cluster compares for overall Mathematica performance as well as specific computations, providing an accurate, real-world estimation of how well a system performs and whether or not investing in different hardware would be advantageous.

gridMathematica 2.1 is available for all common Linux and Unix systems, Windows, and Mac OS X. Computers in a cluster do not have to run the same operating system. Introductory training on gridMathematica and parallel computation is offered by the Wolfram Education Group. More details are available on the gridMathematica website.