The communication challenge for MPP: Intel Paragon and Meiko CS-2
Parallel Computing
IBM Systems Journal
The SP2 high-performance switch
IBM Systems Journal
The communication software and parallel environment of the IBM SP2
IBM Systems Journal
Computer architecture (2nd ed.): a quantitative approach
Computer architecture (2nd ed.): a quantitative approach
Modeling Communication Overhead: MPI and MPL Performance on the IBM SP2
IEEE Parallel & Distributed Technology: Systems & Technology
Assessing Fast Network Interfaces
IEEE Micro
Parallel simulation of message routing networks
PDP '95 Proceedings of the 3rd Euromicro Workshop on Parallel and Distributed Processing
MPI programming environment for IBM SP1/SP2
ICDCS '95 Proceedings of the 15th International Conference on Distributed Computing Systems
Message-Passing Performance of Various Computers
Message-Passing Performance of Various Computers
An Analytical Method for Predicting the Performance of Parallel Image Processing Operations
The Journal of Supercomputing
Data Locality Exploitation in the Decomposition of Regular Domain Problems
IEEE Transactions on Parallel and Distributed Systems
Sensitivity of Performance Prediction of Message Passing Programs
The Journal of Supercomputing
Four-Ary Tree-Based Barrier Synchronization for 2D Meshes without Nonmember Involvement
IEEE Transactions on Computers - Special issue on the parallel architecture and compilation techniques conference
Modeling of interconnection subsystems for massively parallel computers
Performance Evaluation
Performance Prediction of Oblivious BSP Programs
Euro-Par '01 Proceedings of the 7th International Euro-Par Conference Manchester on Parallel Processing
| Hi-index | 0.00 |
IBM has recently launched an upgrade of the communication subsystem of its SP2 parallel computer. This change affects both hardware and software elements: high-performance switch, message interface adapters, and a new implementation of the MPI message-passing library. To characterize to what extent these changes will affect the execution times of parallel applications, these authors have run a collection of benchmarks on a SP2 with the old communication subsystem and on the same machine after upgrade. These benchmarks include point-to-point and collective communication tests as well as a set of complete parallel applications. The performance indicators are the latency and throughput exhibited by the basic communication tests and the execution time in the case of real applications. Result indicate that only under certain circumstances does a significant performance increase result.