[
https://issues.apache.org/jira/browse/BOOKKEEPER-464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13503463#comment-13503463
]
Fangmin Lv commented on BOOKKEEPER-464:
---------------------------------------
Actually this is the implementer of revisit garbage collection algorithm
proposed in BOOKKEEPER-249, there is a detailed design doc gc_revisit.pdf on
BOOKKEEPER-249 which explains different garbage collector algorithms and their
performance comparison. I'm afraid it will be superfluous to rewrite the doc
about this feature, so here I simply describe the algorithm I used in the
improved garbage collector:
1. When delete ledger, we will write the deleted ledgers to it's ensembles'
/ledgers/deleted/Bi/ nodes. Then the bookie's garbage collector thread will
read his own /ledgers/deleted/Bi/ node to get the deleted ledgers list, finally
bookie will delete the ledgers and remove ledger metadata in meta storage
according to the list. This is the same as 4.3 section Detail Design 2 in
gc_revisit.pdf.
2. To avoid zombie entries, we will trigger polling based garbage collector
when bookie's disk went out.
> Provide an improved GC algorithm
> --------------------------------
>
> Key: BOOKKEEPER-464
> URL: https://issues.apache.org/jira/browse/BOOKKEEPER-464
> Project: Bookkeeper
> Issue Type: Sub-task
> Components: bookkeeper-server
> Reporter: Sijie Guo
> Assignee: Fangmin Lv
> Fix For: 4.2.0
>
> Attachments: BOOKKEEPER-464.patch
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
