March 2003
Intermediate to advanced
912 pages
27h 17m
English
In Chapter 15 we considered the effects of crashes and mechanisms that would provide crash resilience for a single operation invocation. Figure 17.6 reiterates the basic approach. In Chapter 16 we considered an operation on one host being invoked from another host in a distributed system and considered the possibility of crashes both in the invoking and invoked systems. We now generalize these ideas.

Suppose, in the simple example introduced in Section 17.5 above, that a crash occurs after the debit operation, which must have written its result to persistent store in order to ...
Read now
Unlock full access