cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vijay (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-3762) AutoSaving KeyCache and System load time improvements.
Date Tue, 24 Jan 2012 23:06:40 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-3762?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13192641#comment-13192641
] 

Vijay commented on CASSANDRA-3762:
----------------------------------

I dont think it is as bad as it looks.... We aren't doing a lot of random IO because with
this patch Keys are sorted and we will read the same blocks often and if it is mmapped it
will get the most advantage. Also most of the work load, the keys will not be from the same
SST's and 75% of the keys falling into a SST is not that common IMO (If they do they have
a bigger problem because all their reads are going to be loger and longer) the load time increases
if we have a lot of data in the disk.
I got around 180ms for 3K keys and thats far is the memory in my laptop :)

The other option is to redesign keycache and save the Index location when we store the keys
and then look it up and to fault fill the data which are not in the cache via (getPosition)....
Makes sense?
                
> AutoSaving KeyCache and System load time improvements.
> ------------------------------------------------------
>
>                 Key: CASSANDRA-3762
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-3762
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 1.2
>            Reporter: Vijay
>            Assignee: Vijay
>            Priority: Minor
>             Fix For: 1.2
>
>         Attachments: 0001-SavedKeyCache-load-time-improvements.patch
>
>
> CASSANDRA-2392 saves the index summary to the disk... but when we have saved cache we
will still scan through the index to get the data out.
> We might be able to separate this from SSTR.load and let it load the index summary, once
all the SST's are loaded we might be able to check the bloomfilter and do a random IO on fewer
Index's to populate the KeyCache.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message