GRIDtoday Logo Crosswalk

DAILY NEWS AND INFORMATION FOR THE GLOBAL GRID COMMUNITY / JANUARY 27, 2003: VOL. 2 NO. 4

( Previous Article )   ( Table of Contents )   ( Next Article )

EI RELEASES COMPILER FOR GRID CLUSTER COMPUTING

EI is launching its new software product solution for parallel programming, the CxC compiler, to generate executable programs using this innovative language.

The parallel programming language CxC is optimized to model and simulate the behavior of particles and their interactions in that historically required parallel supercomputing power, such as life sciences and biotechnology companies, where computational biology and quantum chemistry depends on modeling the interaction of molecules and particles in complex compounds.

Aerospace companies, where fluid dynamics and airflow simulation is a key ingredient in product development. Automotive manufacturers, where car-crash simulation requires the complex interaction of millions of particles with increasing demand for accuracy. Atmospheric modeling and meteorology, where complex mathematical models of particle simulation are used for weather forecasting and environmental analyses.

The scalability of CxC allows scientists and engineers to develop, test and execute parallel algorithms on a PC or laptop and run the same executable on hundreds of grid clustered computers without dead-locks and synchronization problems.

"Parallel computation is key for many simulation applications. Cluster computing combines affordable hardware with high performance, particularly in industries such as biochemistry and computer-aided engineering," said EI's CEO, Matt Oberdorfer.

"With the CxC programming language and EI's compiler, companies can realize massive processing and computation capability across low-cost hardware platforms, enabling them to achieve greater levels of productivity, increased ROI, reduced time to market, and higher quality results by running simulations and analyses faster than previously possible."

Dramatically different from current approaches for parallel programming, EI's pioneering high-performance software is designed to simplify modeling and simulation of a large number of interdependent elements, their parallel dynamics and interactions. CxC is multi-platform and allows for the creation of parallel applications on a laptop or PC. By running the same executable on hundreds of low-cost processors on a compute cluster, supercomputing performance is achieved.

CxC allows easy prototyping and development of parallel algorithms, combining simple deadlock-free syntax and semantics with the most powerful parallel programming paradigm. Its programs are easier to develop and are faster in performance than equivalent message passing or shared memory programs. "CxC unleashes the true capabilities of clusters," said Oberdorfer.

"With EI, customers can truly build their own grid supercomputers from low-cost, high-performance component systems."

Derek C. Walter, Research Associate in the Department of Chemistry and Biochemistry, University of California at Los Angeles, relates the effectiveness of the new CxC Compiler in demanding applications like Molecular Modeling.

"Using the CxC language and accompanying runtime environment, we were able to parallelize portions of our configuration interaction program. The process involved painlessly converting our FORTRAN subroutines to CxC and then seamlessly incorporating them into the pre-existing FORTRAN package."

Professor Emily A. Carter, Department of Chemistry and Biochemistry at UCLA, said, "The results were quite good; we observed nearly ideal scalability and speedups in test calculations on alkanes having up to 6 heavy atoms. We intend to use CxC in subsequent development projects involving our configuration interaction program."

U.S Pricing And Availability

CxC Compiler is currently available from EI and through selected resellers with prices beginning at $2975 for a single-user system.

About Engineered Intelligence

Engineered Intelligence Corporation (EI) is the technological leader in parallel programming languages for high-performance grid cluster computing. As developer of software that enables a cluster of low-cost computing resources to become a virtual single high-performance computing system, EI solutions meet the needs of customers in life sciences, computer-aided engineering, financial modeling and other industries where high performance computing is critical.

EI owns all intellectual property and several patents (pending) related to this technology.

http://www.engineeredintelligence .com

( Top of Page )

( Previous Article )   ( Table of Contents )   ( Next Article )