Advanced Cluster Systems Announces HPC Tool

By DE Editors

At Supercomputing 2007 this week, Advanced Cluster Systems (ACS;  Aliso Viejo, CA ) announced its Supercomputing Engine for Mathematica (SEM), a high-performance supercomputing tool that reduces the execution time of Wolfram Research‘s ( Champaign, IL ) Mathematica algorithms by up to 700 percent on a single 8-CPU machine.

SEM applies the paradigm of distributed-memory Message-Passing Interface (MPI) and is said to be the industry’s first implementation of an MPI library within Mathematica. Now researchers and scientists can develop new codes and algorithms for parallel computing faster by combining Mathematica’s rich feature set with MPI.

Closely following the supercomputing industry-standard MPI, SEM creates a standard way for every Mathematica kernel in a Multi-CPU workstation or a cluster of Multi-CPU nodes to communicate with each other directly. SEM enables all kernels to communicate with each other directly and collectively the way modern supercomputers do, in contrast to typical grid implementations that are solely master-slave or server-client.

SEM enables Wolfram Research’s Mathematica to be combined with the easy-to-use, supercomputer-compatible Pooch (Pooch is an acronym for Parallel OperatiOn and Control Heuristic application) clustering technology of Dauger Research. This fusion applies the parallel computing paradigm of today’s supercomputers and the ease-of-use of Pooch to Mathematica, enabling possibilities none of these technologies could do alone.

As they do for other cluster applications, Pooch and MacMPI provide the support infrastructure to enable this supercomputing-style parallel startup 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 (HPC) practitioners find necessary to address the largest problems in scientific computing.

SEM is also now an integral part of ACS’s Math Supercomputer-In-A-Box, a low cost, 8-CPU Mathematica supercomputing workstation. The solution is built on an 8-core 3GHz Mac Pro with gridMathematica and is networked for increased processing power. For example, a network of four ACS Math Supercomputers-In-a Box delivers a 32 CPU with 64 GB ECC RAM solution that can reduce execution time of Mathematica algorithms by up to 2700 percent.

For details, contact Advanced Cluster Systems, LLC or Dauger Research, Inc.

Sources: Press materials received from the company and additional information gleaned from the company’s website.