A failure recovery mechanism for distributed metadata servers in DCFS2

Author

Fan, Zhihua ; Xiong, Jin ; Ma, Jie

Author_Institution

Nat. Res. Center for Intelligent Comput. Syst., Chinese Acad. of Sci., China

fYear

2004

fDate

20-22 July 2004

Firstpage

2

Lastpage

8

Abstract

Distributed metadata servers are required for cluster file system´s scalability. However, how to distribute the file system metadata among multiple metadata servers and how to make the file system reliable in case of server failures are two difficult problem. We present a journal-based failure-recovery mechanism for distributed metadata servers in the dawning cluster file system-DCFS2. The DCFS2 metadata protocol exploits a modified two-phase commit protocol which ensures consistent metadata updates on multiple metadata servers even in case of one server´s failure. We focus on the logging policy and concurrent control policy for metadata updates, and the failure recovery policy. The DCFS2 metadata protocol is compared with the two phase commit protocol and some virtues are shown. Some results of performance experiments on our system are also presented.

Keywords

distributed databases; file servers; meta data; network operating systems; protocols; system recovery; DCFS2 metadata protocol; concurrent control policy; dawning cluster file system; distributed metadata servers; failure recovery; logging policy; phase commit protocol; server failures; Collaboration; Computers; Distributed computing; File servers; File systems; Intelligent systems; Power capacitors; Protocols; Robustness; Scalability;

fLanguage

English

Publisher

ieee

Conference_Titel

High Performance Computing and Grid in Asia Pacific Region, 2004. Proceedings. Seventh International Conference on

Print_ISBN

0-7695-2138-X

Type

conf

DOI

10.1109/HPCASIA.2004.1324010

Filename

1324010