cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jay Patel (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-7882) Memtable slab allocation should scale logarithmically to improve occupancy rate
Date Thu, 18 Sep 2014 22:13:34 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-7882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14139606#comment-14139606
] 

Jay Patel commented on CASSANDRA-7882:
--------------------------------------

Hey Benedict, I've a first cut on this. Need to do some testing before attaching a patch.

Should we scale back region size? If so, when? For now, I'm scaling back the region size logarithmically
in the setDiscard() method when all regions are deallocated. 

> Memtable slab allocation should scale logarithmically to improve occupancy rate
> -------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-7882
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7882
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Jay Patel
>            Assignee: Jay Patel
>              Labels: performance
>             Fix For: 2.1.1
>
>
> CASSANDRA-5935 allows option to disable region-based allocation for on-heap memtables
but there is no option to disable it for off-heap memtables (memtable_allocation_type: offheap_objects).

> Disabling region-based allocation will allow us to pack more tables in the schema since
minimum of 1MB region won't be allocated per table. Downside can be more fragmentation which
should be controllable by using better allocator like JEMalloc.
> How about below option in yaml?:
> memtable_allocation_type: unslabbed_offheap_objects
> Thanks.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message