PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing
PVM: Parallel virtual machine: a users' guide and tutorial for networked parallel computing
Global arrays: a nonuniform memory access programming model for high-performance computers
The Journal of Supercomputing
COM and DCOM: Microsoft's vision for distributed objects
COM and DCOM: Microsoft's vision for distributed objects
MPI-The Complete Reference, Volume 1: The MPI Core
MPI-The Complete Reference, Volume 1: The MPI Core
Applying Enterprise Javabeans: Component-Based Development for the J2ee Platform
Applying Enterprise Javabeans: Component-Based Development for the J2ee Platform
Parallel components for PDEs and optimization: some issues and experiences
Parallel Computing - Special issue: Advanced environments for parallel and distributed computing
Toward a Common Component Architecture for High-Performance Scientific Computing
HPDC '99 Proceedings of the 8th IEEE International Symposium on High Performance Distributed Computing
A Component Architecture for High-Performance Scientific Computing
International Journal of High Performance Computing Applications
Hi-index | 0.00 |
Component-based programming has been recognized as an effective technique to manage the increasing complexity of high performance scientific code. Yet the adoption of the component approach introduces new challenges that are unique to the parallel and distributed high performance scientific computing domain. The Common Component Architecture (CCA) is an ongoing effort to develop a component model targeted specifically to the needs of high-performance scientific computing, and to study the issues involved in developing, deploying, and using such an infrastructure in the HPC environment. In this paper, we present an overview of our investigation into incorporating message passing systems, namely MPI and PVM, into CCA component-based applications and frameworks. We discuss the architectural and performance issues of different design options, and their impact on developing new components as well as on the process of componentizing existing codes. These ideas are based on experiences and insights gained from developing a number of scientific simulations within the prototype parallel Ccaffeine framework.