By Topic

A Novel Collaborative Fault-Tolerance Mechanism for Metadata Server Cluster

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Li, Z.X. ; Data Storage Inst., Singapore ; Wei, Q.S. ; Wang, Y.H.

In object-based storage system, metadata residing in metadata server (MDS) is crucial to the survival of overall system. Keeping metadata always available is becoming one of the key issues to be addressed for object-based storage system design and implementation. This paper presents a novel cooperative fault-tolerance mechanism for metadata server cluster, called Co-MDS. The Co-MDS consolidates disk space of individual MDS into a global metadata storage space and provides high fault-tolerance by distributing metadata and its parity data across collaborative metadata server cluster environment. With this mechanism, metadata is still available even when a part of metadata servers fail. Our design is transparent for applications and complies with OSD T10 standard. The experiments demonstrate that the proposed collaborative mechanism can offer high fault-tolerance and significant performance.

Published in:

Networks, 2007. ICON 2007. 15th IEEE International Conference on

Date of Conference:

19-21 Nov. 2007