Clusters of networked, off-the-shelf workstations are currently used for computationintensive, parallel applications. However, it is hardly possible to predict the timing behaviou...
This work provides a systematic study of the impact of communication performance on parallelapplications in a high performance network of workstations. We develop an experimental ...
Richard P. Martin, Amin Vahdat, David E. Culler, T...
By managing network resources at compile time, the compiled communication technique greatly improves the communication performance for communication patterns that are known at com...
—The current trend in clusters architecture leads toward a massive use of multicore chips. This hardware evolution raises bottleneck issues at the network interface level. The us...
For regular, sparse, linear systems, like those derived from regular grids, using High Performance Fortran (HPF) for iterative solvers is straightforward. However, for irregular ma...