A decentralized, scalable, and autonomous grid monitoring system

Authors:
Laurent Baduel;Satoshi Matsuoka
Affiliations:
Tokyo Institute of Technology, Tokyo, Japan;Tokyo Institute of Technology, Tokyo, Japan and National Institut of Informatics, Tokyo, Japan
Venue:
OPODIS'07 Proceedings of the 11th international conference on Principles of distributed systems
Year:
2007

Citing 7
Cited 0

On spreading a rumor

SIAM Journal on Applied Mathematics
The network weather service: a distributed resource performance forecasting service for metacomputing

Future Generation Computer Systems - Special issue on metacomputing
The Vision of Autonomic Computing

Computer
Grid Information Services for Distributed Resource Sharing

HPDC '01 Proceedings of the 10th IEEE International Symposium on High Performance Distributed Computing
A Gossip Protocol for Subgroup Multicast

ICDCSW '01 Proceedings of the 21st International Conference on Distributed Computing Systems
A taxonomy of grid monitoring systems

Future Generation Computer Systems
Enabling the p2p JXTA platform for high-performance networking grid infrastructures

HPCC'05 Proceedings of the First international conference on High Performance Computing and Communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Grid monitoring systems collect a substantial amount of information on the infrastructure's status in order to perform various tasks, more commonly to provide a better use of the grid's entities. Modern computational and data grids have become very complex by their size, their heterogeneity, their interconnection. Monitoring systems as any other grid's tools have to adapt to this evolution. In this paper we present a decentralized, scalable, and autonomous grid monitoring system able to tackle the growths of scale and complexity. System's components communications are hierarchically organized on a peer-to-peer overlay network. Fresh information is efficiently propagated thanks to an directed gossip protocol that limits the number of message. Automation of key management operations eases system administration and maintenance. This approach provides scalability and adaptability. The main properties of our application are presented and discussed. Performance measurements confirm the efficiency of our system.