A Study on Monitoring for Distributed Information System and its Application
Grant-in-Aid for Scientific Research (C)
|Research Institution||University of Electro-Communications|
HAKOZAKI Katsuya Graduate school of Information Systems, Professor, 大学院・情報システム学研究科, 教授 (10262362)
KANAI Hideaki Graduate school of Information Systems, Research associate, 大学院・情報システム学研究科, 助手 (90282920)
|Project Fiscal Year
1996 – 1998
Completed(Fiscal Year 1998)
|Budget Amount *help
¥2,500,000 (Direct Cost : ¥2,500,000)
Fiscal Year 1998 : ¥800,000 (Direct Cost : ¥800,000)
Fiscal Year 1997 : ¥800,000 (Direct Cost : ¥800,000)
Fiscal Year 1996 : ¥900,000 (Direct Cost : ¥900,000)
|Keywords||network monitoring / network configuration management / network performance evaluation / traffic measurement / ネットワーク監視 / ネットワーク構成管理 / ネットワーク性能評価 / トラフィック測定 / プロトコル / ネットワーク性能管理 / システム測定 / 経路追跡 / インターネット|
The purpose of this study is to establish new monitoring methodology for distributed computing system corresponding to system debugging, network obstruction monitoring and performance evaluation of distributed computing system. There are three process steps in the methodology ; Step 1 : monitoring system operations on the network, Step 2 : restoring each operation and Step 3 : replaying the behavior of the whole system.
We investigated network monitoring technique for watching system behavior on network level, visualization technique for presenting the obtained data in visual form and reconstructing technique for restoring system behavior by analyzing the data.
There are three topics in our research results ;
(a) Distributed computing system monitor
We develop a network monitoring system for system behavior on the Internet. The system stores system operations information and visualizes the information using 3D graphics. The system can detect nodes as bottleneck of network traffic and unusu
al slow access nodes on our campus network. The system is useful for the improvement of network system configuration.
(b) Performance analyzing and evaluating for the distributed computing system
One is WWW access evaluation generational replacement schemes suitable for patterns of access to World Wide Web (WWW) proxy server caches. With our schemes, the hit rate per page is improved, and the number of cache misses is reduced. Other is adaptive load balancing method on distributed computing system. We confirm the effectiveness by computer simulation experiments effectiveness of adaptive and static load balancing on distributed computing system.
(c) Application systems based on our research products
They are an application management framework for user oriented system management from a viewpoint of application quality of services in distributed computing system, a virtual library system for personal use of information resources distributed on the Internet, WWW information resource updating managements based on WWW cashing log data, and Ad Hoc type workflow system using e-mail system and WWW platform. Less
Research Output (6results)