cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gary Dusbabek (JIRA)" <j...@apache.org>
Subject [jira] Commented: (CASSANDRA-620) Add per-keyspace replication factor (possibly even replication strategy)
Date Wed, 13 Jan 2010 02:57:54 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-620?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12799546#action_12799546
] 

Gary Dusbabek commented on CASSANDRA-620:
-----------------------------------------

Jaakko,

1) As long as the 'replica_' member of ARS can safely be considered as a maximum, then there
is no reason we can't limit the number of instances.  The fact that it is used to populate
in 'getNaturalEndpoints' made me unsure.

2) No. This could be addressed by making 'TMD.pendingRanges' a ConcurrentMap (preferred) or
by synchronizing SS.calculatePendingRanges().  I suppose this doesn't address the fact that
the contents could change while whoever called getPendingRanges() is using the data, but that
we had that problem before.

3)  StorageService owns it.  I had a hard time following this at first too.  Every reference
to TMD can trace it's roots back to the one created in SS.  TMD is *so close* to being a singleton.
 I can't remember why I changed calculatePendingRanges() to retrieve it from the ARS instead
of grabbing the SS member though.

4) What if I change this so that the message becomes: "send me bootstrap data for this list
of tables" instead of: "for each table, send me bootstrap data."?  Then, as soon as a remote
node is finished, it can send an indication and the bootstrapping node can remove the remote
node from it's bootstrap set.  It seems like that would solve CASSANDRA-673 at the same time,
correct?

Thanks for the feedback!

> Add per-keyspace replication factor (possibly even replication strategy)
> ------------------------------------------------------------------------
>
>                 Key: CASSANDRA-620
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-620
>             Project: Cassandra
>          Issue Type: New Feature
>            Reporter: Jonathan Ellis
>            Assignee: Gary Dusbabek
>             Fix For: 0.9
>
>         Attachments: 0001-push-replication-factor-and-strategy-into-table-exce.patch,
0002-cleaned-up-as-much-as-possible-before-dealing-with-r.patch, 0003-push-table-names-into-streaming-expose-TMD-in-ARS.patch,
0004-fix-non-compiling-tests.patch, 0005-introduce-table-into-pending-ranges-code.patch, 0006-added-additional-testing-keyspace.patch,
0007-modify-TestRingCache-to-make-it-easier-to-test-speci.patch, 0008-push-endpoint-snitch-into-keyspace-configuration.patch,
0009-Marked-a-few-AntiEntropyServices-methods-as-private-.patch
>
>
> (but partitioner may only be cluster-wide, still)
> not 100% sure this makes sense but it would allow maintaining system metadata in a replicated-across-entire-cluster
keyspace (without ugly special casing), as well as making Cassandra more flexible as a shared
resource for multiple apps

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message