| Project/Area Number |
07680450
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
社会システム工学
|
|---|
| Research Institution | TOKYO METROPOLITAN UNIVERSITY |
|---|
Principal Investigator |
IWASAKI Kazuhiko Faculty of Engineering Professor, Tokyo Metropolitan University, 工学部, 教授 (40232649)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
HASHIMOTO Akihiro Assistant Professor Chiba University Information Processing Center., 総合情報処理センター, 講師 (60164779)
|
|---|
| Project Period (FY) |
1995 – 1996
|
| Project Status |
Completed (Fiscal Year 1996)
|
| Budget Amount *help |
¥2,300,000 (Direct Cost: ¥2,300,000)
Fiscal Year 1996: ¥500,000 (Direct Cost: ¥500,000)
Fiscal Year 1995: ¥1,800,000 (Direct Cost: ¥1,800,000)
|
|---|
| Keywords | network / failure / fault-tolerance / maximally connected graph / TCP / IP / checksum / Statistics / マルチメディアネットワーク / フォールトトレランス技術 / セキュリティ技術 / 物理層 / 落雷 / データリンク層 / ネットワーク層 / 障害事例 |
|---|
| Research Abstract |
The purpose of this project is to analyze failures on campus networks and to develop fault-tolerance for the networks to localize the failures. Another purpose of this work is to develop statistical data analysis technique.
Network fault-tolerance was theoretically analyzed. A network consists of nodes and links that connect nodes. If we use many links, the distance between the nodes will decrease and the much hardware will be required. In this project, we designed graphs of constant degrees five and six that have lower cost, where cost is defined as degree * diameter. It is also shown that the proposed graphs are maximally connected graph and have good faulttolerance.
The TCP / IP is widely used in current Internet communications. In this protocol, an error-control scheme called "checksum" is used. We analyzed the double and triple-bit error detecting probability for the checksum. Since more than four-bit error rarely occurs, the upper bound for checksum can be derived.
Failures that were observed on a campus was categorized in the following viewpoints. That is, a physical factor / human factor, failure range (from whole campus to very limited area) , failure level (out of operation, intermittent, negligible) , and so on. Fault-tolerance for the failures are not necessarily design carefully. These remind for further research.
The relationship between students'scores in high school and those in University are analyzed using a statistic technique. Parameters to maximize scores in a University were theoretically derived. A entrance examination technique is proposed that will maximize the score in the University.
|
