| Project/Area Number |
14380135
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
計算機科学
|
|---|
| Research Institution | Utsunomiya University |
|---|
Principal Investigator |
BABA Takanobu Utsunomiya University, Faculty of Engineering, Professor, 工学部, 教授 (70092616)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YOKOTA Takashi Utsunomiya University, Faculty of Engineering, Associate Professor, 工学部, 助教授 (90334078)
OOTSU Kanemitsu Utsunomiya University, Faculty of Engineering, Research Associate, 工学部, 助手 (00292574)
YOSHINAGA Tsutomu The University of Electro-Communications, Associate Professor, 大学院・情報システム学研究科, 助教授 (60210738)
KATO Shigeo Utsunomiya University, Faculty of Engineering, Professor, 工学部, 教授 (00143529)
HASEGAWA Madoka Utsunomiya University, Faculty of Engineering, Associate Professor, 工学部, 助教授 (80322014)
|
|---|
| Project Period (FY) |
2002 – 2005
|
| Project Status |
Completed (Fiscal Year 2005)
|
| Budget Amount *help |
¥16,300,000 (Direct Cost: ¥16,300,000)
Fiscal Year 2005: ¥1,000,000 (Direct Cost: ¥1,000,000)
Fiscal Year 2004: ¥4,600,000 (Direct Cost: ¥4,600,000)
Fiscal Year 2003: ¥6,000,000 (Direct Cost: ¥6,000,000)
Fiscal Year 2002: ¥4,700,000 (Direct Cost: ¥4,700,000)
|
|---|
| Keywords | Receiving message prediction / Message passing interface / Speculative execution / NAS parallel benchmark / Message communication / Workstation cluster |
|---|
| Research Abstract |
This research proposes and evaluates the Receiving Message Prediction Method for high performance message passing. In this method, a node in the idle state predicts the next message reception and speculatively executes the message reception and user processes. This method is independent of underlying computer architecture and message passing libraries. We propose the algorithms for the message prediction, and evaluate them from the viewpoint of the success ratio and speed-ups. We use the NAS parallel benchmark programs as typical parallel applications running on two different types of parallel platforms, i.e., a workstation cluster and a shared memory multiprocessor. The experimental results show that the method can be applied to various platforms. The method can also be implemented just by changing the software inside their message passing libraries without any support from the underlying system software or hardware. This mean that we do not require any change of applications software that uses the libraries. The application of the method to the message passing interface libraries achieves a speed-up of 6.8 % for the NAS Parallel Benchmarks, and the static and dynamic selection of prediction methods based on profiling results improve the performance.
|
