You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "David King (JIRA)" <ji...@apache.org> on 2010/08/11 20:23:19 UTC

[jira] Created: (CASSANDRA-1379) Uncached row reads may block cached reads

Uncached row reads may block cached reads
-----------------------------------------

                 Key: CASSANDRA-1379
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1379
             Project: Cassandra
          Issue Type: Bug
          Components: Core
    Affects Versions: 0.6.4
            Reporter: David King


The cap on the number of concurrent reads appears to cap the *total* number of concurrent reads instead of just capping the reads that are bound for disk. That is, given N concurrent readers if all of them are busy waiting on disk, even reads that can be served from the row cache will block waiting for them.

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

[jira] Commented: (CASSANDRA-1379) Uncached row reads may block cached reads

Posted by "Chris Goffinet (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-1379?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12918316#action_12918316 ] 

Chris Goffinet commented on CASSANDRA-1379:
-------------------------------------------

It will also block BF access. BF is fast, but if you ran into the situation where you needed to check if a large percentage of keys don't exist in the system, this read path could block if you had uncached reads happening as well. Not sure it's possible to get around this unless you work on CASSANDRA-1576

> Uncached row reads may block cached reads
> -----------------------------------------
>
>                 Key: CASSANDRA-1379
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1379
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: David King
>            Priority: Minor
>             Fix For: 0.7.1
>
>
> The cap on the number of concurrent reads appears to cap the *total* number of concurrent reads instead of just capping the reads that are bound for disk. That is, given N concurrent readers if all of them are busy waiting on disk, even reads that can be served from the row cache will block waiting for them.

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

[jira] Updated: (CASSANDRA-1379) Uncached row reads may block cached reads

Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/CASSANDRA-1379?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jonathan Ellis updated CASSANDRA-1379:
--------------------------------------

           Issue Type: New Feature  (was: Bug)
        Fix Version/s: 0.7.1
    Affects Version/s:     (was: 0.6.4)
             Priority: Minor  (was: Major)

Are you suggesting splitting the read stage into check-cache and fetch-data stages?

> Uncached row reads may block cached reads
> -----------------------------------------
>
>                 Key: CASSANDRA-1379
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1379
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: David King
>            Priority: Minor
>             Fix For: 0.7.1
>
>
> The cap on the number of concurrent reads appears to cap the *total* number of concurrent reads instead of just capping the reads that are bound for disk. That is, given N concurrent readers if all of them are busy waiting on disk, even reads that can be served from the row cache will block waiting for them.

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

[jira] Commented: (CASSANDRA-1379) Uncached row reads may block cached reads

Posted by "David King (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/CASSANDRA-1379?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12897430#action_12897430 ] 

David King commented on CASSANDRA-1379:
---------------------------------------

Not being familiar with the implementation I'm not sure I'm suggesting anything, but that sounds sane. Or checking the cache first in the stage previous

> Uncached row reads may block cached reads
> -----------------------------------------
>
>                 Key: CASSANDRA-1379
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1379
>             Project: Cassandra
>          Issue Type: New Feature
>          Components: Core
>            Reporter: David King
>            Priority: Minor
>             Fix For: 0.7.1
>
>
> The cap on the number of concurrent reads appears to cap the *total* number of concurrent reads instead of just capping the reads that are bound for disk. That is, given N concurrent readers if all of them are busy waiting on disk, even reads that can be served from the row cache will block waiting for them.

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