Sprite position statement: use distributed state for failurerecovery
Welch, B.
Baker, M.
Douglis, F.
Hartman, J.
Rosenblum, M.
Ousterhout, J.
Dept. of Electr. Eng. & Comput. Sci., California Univ., Berkeley, CA;
This paper appears in: Workstation Operating Systems, 1989., Proceedings of the Second Workshop on
Publication Date: 27-29 Sep 1989
On page(s): 130-133
Meeting Date: 09/27/1989 - 09/29/1989
Location: Pacific Grove, CA, USA
References Cited: 4
INSPEC Accession Number: 3601312
Digital Object Identifier: 10.1109/WWOS.1989.109282
Current Version Published: 2002-08-06
Abstract
The authors advocate keeping state in main memory instead of
logging state to disk, so that higher performance services can be
implemented. They are motivated by their distributed file system which
uses stateful servers to support a high-performance distributed caching
system. For reliability, a server's state is replicated in the main
memory of other hosts so that the system can recover from the failure of
a server. After a server reboots, its clients help it rebuild its
internal state. The authors point out that as networks and processors
get faster, but disks do not, relying on other hosts will be more
efficient than using disks
Index
Terms
Available to subscribers and IEEE members.
References
Available to subscribers and IEEE members.
Citing Documents
Available to subscribers and IEEE members.
Cart
