Skip to Main Content
Operating Systems: Concurrent and Distributed Software Design
book

Operating Systems: Concurrent and Distributed Software Design

by Jean Bacon, Tim Harris
March 2003
Intermediate to advanced content levelIntermediate to advanced
912 pages
27h 17m
English
Pearson Business
Content preview from Operating Systems: Concurrent and Distributed Software Design

21.5. Use of a recovery log

As noted in Section 21.1, it is assumed that a daily dump of the system's persistent state is taken and that the DBMS must make a record of subsequent activity if complete crash resilience is to be achieved at any time.

An outline of the general approach is that, on an invocation, the object value is updated in persistent store and a record of the invocation is made in a log. We have already assumed that every object keeps information (for concurrency control purposes) of the kind we require to be logged and we can require objects to write log records. In order to achieve a recoverable state we must consider when the update to persistent object state is made and when the corresponding log record is written to persistent ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Start your free trial

You might also like

Principles of Concurrent and Distributed Programming, Second Edition

Principles of Concurrent and Distributed Programming, Second Edition

M. Ben-Ari
Operating Systems in Depth

Operating Systems in Depth

Thomas W. Doeppner

Publisher Resources

ISBN: 0321117891Purchase book