MATHEMATICS, SIMULATION AND MODELLING

Accelerate Your MATLAB Codes

8 January 2007

Interactive Supercomputing

Star-P from Interactive Supercomputing is a software platform that allows existing desktop simulation tools to operate interactively and automatically on high-performance computers. Star-P acts as a bridge between popular scientific and engineering desktop computing tools, such as MATLAB from The MathWorks, and the parallel computing power of high-performance computers.

http://www.interactivesupercomputing.com/scn

Take the Star-P Interactive Tour

  • View streaming presentations and demonstrations
  • Benefits of an interactive parallel computing platform
  • Fine- and coarse-grained parallelism with MATLAB
  • Google PageRank algorithm demonstration
  • And MORE...

http://www.interactivesupercomputing.com/scn

Download the "Going Parallel" Information Kit

  • Critical challenges in solving large and complex problems in science and engineering
  • Current trends in custom parallel application development
  • Using Star-P enables custom HPC code development using the familiar MATLAB environment

http://www.interactivesupercomputing.com/scn

Download IDC's Report: Star-P at Air Force Research Lab

This IDC Buyer Case Study examines the usage of Star-P at Air Force Research Lab. IDC explores the conditions of the facility, motivations of the product adoption, the results of the implementation, and the benefits and challenges of the solution.
http://www.interactivesupercomputing.com/scn

Download the 2006 Report on "Development of Custom Parallel Applications"

Based on hundreds of surveys and interviews, the recent study of the current state of custom application development in high performance technical computing examines the software tools currently used among several industries, probes current application development environments, practices, and limitations, and catalogs critical issues and bottlenecks.

  • Debugging, limits of the HPC software, and code writing, programming efficiency, and translation are the most frequently cited bottlenecks across all industries.
  • MATLAB, Python, Mathematica, C and Fortran were the most commonly mentioned prototyping tools. On average, respondents use three different types of application prototyping tools.
  • A significant amount of time is spent connecting to, and integrating with, existing parallel libraries. In addition to time spent prototyping, respondents are also seeking ways to decrease the actual development phase.
  • Data sets are expected to grow rapidly. The expected median data set within three years ranges from 200 GB to 600 GB, with the most dramatic estimate of growth in commercial enterprises.

Click here to find out more