The patent-pending SET (Supercomputing Engine Technology) enables quick, efficient and cost-effective supercomputing-style parallelization of any modular code without modifications!
SUPERCOMPUTING MADE EASY!
Contact us:
Advanced Cluster Systems LLC
65 Enterprise
Aliso Viejo, CA 92656
Tel: 949-494-0440
Fax: 877-626-0379
Latest News
Authorized Reseller
Authorized Reseller
May 29, 2008
SEM takes full advantage of MPI (Message Passing Interface) in multi-core chip infrastructures, avoids multithreading pitfalls.
“Multicore Eroding Moore’s Law” claims Dr. Dean Dauger of Dauger Research. “Do you seek ever increasing speed from your computers? Do you think multicore chips are a sign of a healthy chip industry? In major trade journals, most articles on the subject seem to uncritically accept multicore as the processor solution moving forward, without suggesting any viable alternative. Meanwhile, the chipmaker giants, Intel, IBM, and AMD, increasingly emphasize multicore chips. But I believe the hardware problems that were serious enough to prompt such fierce competitors to agree on multicore herald a major transition in the computer industry whose full consequences, especially for software writers, are not yet appreciated.”
June 6, 2008
While industry giants are concerned about diminishing returns running parallel software (multithreading) on 16 and 32 processor (multicore) systems in the near future, today's SEM (Supercomputing Engine for Mathematica) can now take advantage of systems with hundreds of multicore processors, with no diminishing returns, thanks to its embedded MPI.
Recent news of industry giants funding academia to figure out how to make multicore usable, shows that they see a serious problem with the multicore status quo. The news article cites a Berkeley report predicting:
“This evolutionary approach to parallel hardware [multicore] and software [multithreading] may work from 2 or 8 processor systems, but is likely to face diminishing returns as 16 and 32 processor systems are realized...
We argue general-purpose computing is taking an irreversible step toward parallel architectures. What's different this time? This shift toward increasing parallelism is not a triumphant stride forward based on breakthroughs in novel software and architectures for parallelism; instead, this plunge into parallelism is actually a retreat from even greater challenges that thwart efficient silicon implementation of traditional uniprocessor architectures.”
April 21, 2008
SEM Cluster at UCI
The University of California, Irvine (UCI) has partnered with Advanced Cluster Systems (ACS), Dauger Research (DR), and Wolfram Research to deploy a 32-core, 64 GB RAM XServe G5 2 GHz computational cluster. The cluster is made available to campus researchers to evaluate the use of the ACS/DR Supercomputing Engine for Mathematica (SEM). The cluster provides an easy to use, parallel programming environment that decreases the execution time of standard codes, scalable to larger clusters and newly emerging desktop, multi-core systems. Tests have demonstrated Mathematica code executing 8 times faster on a single 8-core Mac Pro.
July 9, 2008
Wolfram Research now offers SEM (Supercomputing Engine for Mathematica) on their online store.
Created in partnership with Dauger Reseach, the Supercomputing Engine for Mathematica applies the programming paradigm of today's supercomputers and the easy-to-use Pooch clustering technology to Wolfram Research's Mathematica. In contrast to typical master-slave "grid"s, this solution instead closely follows MPI, from inside the Mathematica environment, and has every kernel in the cluster communicate with each other directly and collectively. Besides being available from Advanced Cluster Systems and Dauger Research, SEM is now available via Wolfram Research Web Store.
February 12, 2009
SEM and gridMathematica Comparison - Ping Pong Benchmark.
“The Supercomputing Engine for Mathematica (SEM) is a technology inspired by the way that modern supercomputers perform work” writes Dr. Dean Dauger of Dauger Research. “SEM supports an “all-to-all” communications network between Mathematica kernels, much like a complete telephone network.”
Read more about the comparison results...
Supercomputing Engine for Mathematica -
An Independent Report.
Dr. Yuko Matsuda of Tokyo Institute of Technology, a world-renowned expert in high performance computing and Wolfram Research’s Mathematica, introduces highly advantageous features of this new parallel computing software package.
High Performance Symbolic Computing
October 16, 2009
Supercomputing Engine for Mathematica -
on 64-bit Linux!
Irvine, CA - October 16, 2009 - Dauger Research, Inc., and Advanced Cluster Systems, LLC, announce today that the Supercomputing Engine for Mathematica (SEM), which combines Wolfram Research's Mathematica with the easy-to-use, supercomputer-compatible Pooch clustering technology of Dauger Research, will extend support to 64-bit Linux. Not only will the patent-pending SEM run on Linux and Macintosh separately, but SEM will also utilize mixed clusters of Macintosh and 64-bit Linux. Like Pooch, SEM will further enhance the power of clusters for its users.
Did you know that your existing sequential code can run on a multi-core desktop, a cluster or a supercomputer as a full-featured MPI-based parallel version without any modification?
We have the SOLUTION!
SEM - See how we provided supercomputer-style parallelization to Wolfram Research Mathematica. We did not have access to Mathematica’s source code.
SEM - Mathematica kernels communication patterns compared with gridMathematica.
Copyright © 2010 Advanced Cluster Systems, LLC. All rights reserved.