Software facilitates interactive supercomputing.July 25, 2006 -
Star-P(TM) v2.3 enables scientists and engineers to code algorithms and models on their desktops using familiar mathematical software packages, and run them instantly and interactively on parallel HPCs. Included Software Development Kit lets users connect to and reuse existing applications and numerical libraries without complexities of parallel programming. Software supports HDF5 file format for storage of large data sets used in scientific and engineering applications.
(Archive News Story - Products mentioned in this Archive News Story may or may not be available from the manufacturer.)
Original Press release
Interactive Supercomputing
135 Beaver St.
Waltham, MA, 02452
USA
ISC Announces Faster, More Powerful Star-P Interactive Supercomputing Software
Enhanced Software Development Kit lets users leverage existing applications and libraries
WALTHAM, Mass., June 20 / -- Interactive Supercomputing Inc. (ISC) today announced a new version of its Star-P(TM) parallel computing software with faster performance and more processing muscle. The new Star-P also features an enhanced Software Development Kit (SDK) that makes it easier to connect to and re-use existing applications and numerical libraries without having to deal with the complexities of parallel programming.
Star-P is an interactive parallel computing platform that lets scientists and engineers code algorithms and models on their desktops using familiar mathematical software packages such as MathWorks' MATLAB(R), and run them instantly and interactively on a variety of parallel high-performance computers (HPC). The new Star-P 2.3 delivers up to 10 times faster performance on a broad set of computational operations, thanks to improved server libraries and a new streamlined client that enables faster communications between the desktop and the HPC. Performance is also improved by its support for larger memory arrays and parallel clusters with more processors.
In addition, Star-P 2.3 now supports enhanced file input/output (I/O) operations, including HDF5, a file format designed for the storage of extremely large data sets used in a broad array of scientific and engineering applications, including magnetic resonance imaging (MRI), radar imaging, environmental science, neutron scattering, and space sciences. Together, these improvements enable researchers to tackle much larger problems on their desktops than ever before possible, while arriving at a solution in a fraction of the time.
Star-P automatically connects desktop applications to HPCs and parallelizes the code on the fly, enabling users to scale their applications across any multi-processor system or parallel cluster in real time. It eliminates the need to re-program MATLAB and other desktop applications in C, Fortran, or MPI to run on parallel computers -- a productivity-killing step that can take months to years to complete for large, complex and computationally intensive problems.
For example, engineers at the Air Force Research Lab (AFRL) in Rome, NY, are taking advantage of Star-P 2.3 performance improvements to accelerate discovery gleaned from microwave image data, according to Kevin Madge, electronics engineer at AFRL. "Star-P 2.3 has dramatically accelerated the development of microwave image reconstruction algorithms by allowing us to stay with the MATLAB programming environment and exploit supercomputing resources for algorithm development and processing throughput," he said. "It's an enabler -- without it we would not be using parallel hardware, and the increased capabilities fundamentally transform what we are able to do in the Advanced Radar Waveforms & Processing Branch."
To accelerate application development, the improved Star-P SDK now enables users to plug in existing serial and parallel numerical libraries and applications and run them on HPC servers, all while working from within the MATLAB desktop environment. They can rapidly create new custom applications by re-using existing codes, but without the complexity of parallel programming. "We were able to easily interface to Star-P the Trilinos software framework and library, which provides broad-ranging high-performance capabilities for solving numerical systems at the heart of many complex multiphysics applications," said Ken Stanley of 500 Software and principal architect of the Amesos direct sparse solver package in the Trilinos library developed by Sandia National Labs. "This dramatically expands the range of solvers and computing power that Star-P and MATLAB users can leverage."
Pricing and availability
Star-P 2.3 is available immediately with an ISC list price starting at $7,995.
About Interactive Supercomputing
Interactive Supercomputing (ISC) launched in 2004 to commercialize Star-P, an interactive parallel computing platform. With automatic parallelization and interactive execution of existing desktop simulation applications, Star-P merges two previously distinct environments -- desktop computers and high performance servers -- into one. Based in Waltham, Mass., the privately held company markets Star-P for a range of security, intelligence, manufacturing, energy, biomedical, financial, and scientific research applications. Additional information is available at http://www.interactivesupercomputing.com/.
CONTACT: Ilya Mirman of Interactive Supercomputing, +1-781-419-4088, imirman@interactivesupercomputing.com
|
|
|
|
|
|
|
| 
