SEM

Mathematica supercomputing at your desktop. Simple.

SEM (Supercomputing Engine for Mathematica) creates a standard way for each Mathematica kernel in the cluster to communicate directly with every other kernel. In contrast to typical grid implementations that are solely master-slave or server-client, all kernels communicate collectively in this solution by closely following the supercomputing industry-standard, Message-Passing Interface (MPI), and The Open MPI Project.

As a result of the collaboration between Advanced Cluster Systems, LLC (ACS) and Dauger Research, Inc., SEM enables Mathematica to be combined with Dauger Research's Pooch clustering technology. This fusion applies the parallel computing paradigm of today's supercomputers and Pooch technology to Mathematica.

Key Features

Ease of programming parallel algorithms
Exceptional parallel scalability by using Mathematica and SEM
Less execution time needed for Mathematica algorithms
Maximum utilization of resources in multi-CPU machines
Symbolic manipulations across a cluster
Distributed memory model for using the RAM of the entire cluster to run memory-consuming applications

Bestellen:

As they do for other cluster applications, Pooch and Open MPI provide the support infrastructure to enable this supercomputing-style, parallel start-up, and inter-kernel communication. After locating, launching, and coordinating Mathematica kernels on a cluster, SEM creates and supports an "all-to-all" communication topology, which high-performance computing practitioners find necessary to address large problems in scientific computing, all within the Mathematica computing environment.

SEM also provides an API that closely follows the MPI standard appropriate for use in Mathematica. By enabling Mathematica kernels to be harnessed together in the manner that supercomputers are, Mathematica, Pooch, and Open MPI embark on a new approach to handling supercomputing challenges.

SEM benötigt gridMathematica 2.2 oder Mathematica Personal Grid Edition 6 und ist erhältlich für Mac OS X 10.4.11 oder höher.

zurück zur Übersicht