In profile: ISC
Born in the MIT labs, Interactive Supercomputing (ISC) (www.interactivesupercomputing.com) is tackling the decades-old problem of solving huge computational problems using parallel supercomputers. ISC co-founder and professor at MIT Alan Edelman was intent on finding new ways to accelerate scientific discovery using parallel computing for faster algorithm development. He concluded that by using simple extensions to scientists’ favorite desktop tool – Matlab –virtually any large scientific problem could be solved on the PC and computed on parallel computers or clusters, without having to reprogram the code. The result was a breakthrough software tool called Star-P. By bringing supercomputing to the masses, Star-P is triggering a new scientific renaissance in which the nation’s most complex problems – from environmental changes to new anti-terrorist technologies – can be solved in a fraction of the time, and at a fraction of the cost. ISC launched in 2004 with a license from MIT to commercialize the technology. It has received government grants for software development from the National Science Foundation, NASA and Oak Ridge National Labs; and has customers in the government, financial services, life science and academic industries.
Product portfolio overview
ISC develops Star-P, an interactive parallel computing platform that enables users to code algorithms and models on their desktops using familiar mathematical software packages such as MathWorks’ Matlab and the open source Python programming language, and run them instantly and interactively on parallel high performance computers (HPCs). Star-P eliminates the need to re-program the applications in low-level programming languages like C, C++, and/or Fortran and does not require the user to know the message passing interface (MPI) to run on parallel computers, which typically takes months to years to complete for large, complex and computationally intensive problems.
By lowering programming barriers, expanding access for more users to HPC resources across organisations, and reusing existing HPC codes, Star-P increases return on organisations’ HPC investment.
Future Star-P releases will support other familiar desktop software packages including: the open source ‘R’ statistical language, Wolfram Research’s Mathematica, Microsoft Excel, Java and others. Star-P’s open platform enables continuous additions of commercial and open source libraries, solvers and algorithms into its server application programming interface (API) with the Star-P software development kit (SDK).
As large multicore-based systems and clusters become common place, Star-P will be enhanced to support hundreds and thousands of processors and terabytes of memory to solve researchers’ toughest problems. Star-P will also support an array of server and cluster management tools for efficiency and productivity. Today, Star-P is an interactive application development tool for scientific analysis and discovery. A future ‘enterprise edition’ will allow all execution of algorithms and applications to exist on the HPC resource – no interaction required by the user – resulting in significant performance improvement and solving larger scale problems – approaching the peta-scale.
Silicon Graphics, Inc. – Global OEM –http://www.sgi.com/
James River Technical, Inc. – US VAR – www.jrti.com/star-p
Federal Edge – US VAR – www.federaledge.com/star-p.html
E&M Computing – Israel VAR – http://www.emet.co.il/