Study on highly adaptable distributed control of interconnection networks
| Project/Area Number |
16K00068
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Computer system
|
|---|
| Research Institution | Utsunomiya University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2016-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2017: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2016: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
|---|
| Keywords | 相互結合網 / 並列計算機 / 輻輳制御 / 時間的最適化 / 空間的最適化 / 大規模最適化 / 論理配置 / 一斉同期通信 / 集合通信 / パケットスケジューリング / 並列計算機システム / 計算機システム |
|---|
| Outline of Final Research Achievements |
Interconnection network is an inevitable component in parallel computing systems. It largely affects the total performance of the parallel system. This program aims at a rapid control of the interconnection network by a highly distributed manner in autonomous routers.
Since an initial study of the program suggested severe difficulty in appropriate reaction to the vast communication traffic, this program has employed some essential discussions on the ideal situations. The program has introduced some artificial intelligence methodologies, which are novel approaches in this area, and temporal and spatial optimizations are deeply discussed. Through the research period, this program has produced many results such as journal papers.
|
| Academic Significance and Societal Importance of the Research Achievements |
並列計算機の通信機構に関しては、従来、様々な通信状況に対応できる機構的な工夫が中心だった。本課題で理想状態を解明することにより、制御の可否次第で大きく通信性能を向上できることが明らかになった。この成果は、過重な機構・装備を備えていない簡素な構成であっても、使いこなし方の工夫によって遜色ない性能を達成できる可能性があることを示している。
また、本課題の研究の過程では、粒子群最適化（PSO）、遺伝的アルゴリズム（GA）、機械学習（ML）、深層学習（DL）などの人工知能の手法を積極的に導入し、顕著な成果を得ることができたことから、当該分野への人工知能技術の適用が進むものと期待される。
|
Report
(5 results)
Research Products
(74 results)
