Data redistribution and remote method invocation for coupled components

• Published in: Journal of parallel and distributed computing Vol. 66; no. 7; pp. 931 - 946
• Main Authors: Bertrand, Felipe, Bramley, Randall, Bernholdt, David E., Kohl, James A., Sussman, Alan, Larson, Jay W., Damevski, Kostadin B.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: San Diego, CA Elsevier Inc 01.07.2006; Elsevier
• Subjects: Applied sciences; Common component architecture; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Exact sciences and technology; Model coupling; Parallel data redistribution; Parallel remote method invocation; Software; Common component architecture; Parallel data redistribution; Parallel remote method invocation; Model coupling; Availability; Performance evaluation; High performance; Parallel algorithm; JAVA language; Computer simulation; Remote method invocation; Parallel machines; Distributed system; Distributed computing; Computational complexity; Modeling; Distributed memory; Remote procedure calls; Software architecture; Massive parallelism; Parallel program; Data structure
• ISSN: 0743-7315; 1096-0848
• DOI: 10.1016/j.jpdc.2005.12.009

With the increasing availability of high-performance massively parallel computer systems, the prevalence of sophisticated scientific simulation has grown rapidly. The complexity of the scientific models being simulated has also evolved, leading to a variety of coupled multi-physics simulation codes. Such cooperating parallel programs require fundamentally new interaction capabilities, to efficiently exchange parallel data structures and collectively invoke methods across programs. So-called “ M × N ” research, as part of the common component architecture (CCA) effort, addresses these special and challenging needs, to provide generalized interfaces and tools that support flexible parallel data redistribution and parallel remote method invocation. Using this technology, distinct simulation codes with disparate distributed data decompositions can work together to achieve greater scientific discoveries.