You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-issues@hadoop.apache.org by "Hairong Kuang (JIRA)" <ji...@apache.org> on 2010/07/06 21:29:50 UTC
[jira] Created: (HADOOP-6850) Allow sequence file to be created
with different bytes per checksum
Allow sequence file to be created with different bytes per checksum
-------------------------------------------------------------------
Key: HADOOP-6850
URL: https://issues.apache.org/jira/browse/HADOOP-6850
Project: Hadoop Common
Issue Type: New Feature
Components: io
Affects Versions: 0.20-append
Reporter: Hairong Kuang
Assignee: Hairong Kuang
Fix For: 0.22.0, 0.20-append
This jira allows a sequence file to be created with a configured bytes per checksum. So we could experiment how this parameter could effect the performance of random reads of a sequence file.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-6850) Allow sequence file to be created
with different bytes per checksum
Posted by "Hairong Kuang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-6850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12886218#action_12886218 ]
Hairong Kuang commented on HADOOP-6850:
---------------------------------------
It looks that I have to add a new parameter "bytesPerChecksum" to FileSystem#create interface especially that we need this in 0.20. Is there any way that I could get around this FileSystem interface change?
> Allow sequence file to be created with different bytes per checksum
> -------------------------------------------------------------------
>
> Key: HADOOP-6850
> URL: https://issues.apache.org/jira/browse/HADOOP-6850
> Project: Hadoop Common
> Issue Type: New Feature
> Components: io
> Affects Versions: 0.20-append
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.22.0, 0.20-append
>
>
> This jira allows a sequence file to be created with a configured bytes per checksum. So we could experiment how this parameter could effect the performance of random reads of a sequence file.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-6850) Allow sequence file to be created
with different bytes per checksum
Posted by "Hairong Kuang (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-6850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12886359#action_12886359 ]
Hairong Kuang commented on HADOOP-6850:
---------------------------------------
In the use case of HBASE-2478, we want to configure HLog & HFile with different values of bytes per checksum.
> Allow sequence file to be created with different bytes per checksum
> -------------------------------------------------------------------
>
> Key: HADOOP-6850
> URL: https://issues.apache.org/jira/browse/HADOOP-6850
> Project: Hadoop Common
> Issue Type: New Feature
> Components: io
> Affects Versions: 0.20-append
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.22.0, 0.20-append
>
>
> This jira allows a sequence file to be created with a configured bytes per checksum. So we could experiment how this parameter could effect the performance of random reads of a sequence file.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HADOOP-6850) Allow sequence file to be created
with different bytes per checksum
Posted by "dhruba borthakur (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-6850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12886326#action_12886326 ]
dhruba borthakur commented on HADOOP-6850:
------------------------------------------
typically, io.bytes.per.checksum is a server side configuration because an administrator usually wants all the files in his/her cluster to have to the same value for this config. Will this be sufficient for you?
> Allow sequence file to be created with different bytes per checksum
> -------------------------------------------------------------------
>
> Key: HADOOP-6850
> URL: https://issues.apache.org/jira/browse/HADOOP-6850
> Project: Hadoop Common
> Issue Type: New Feature
> Components: io
> Affects Versions: 0.20-append
> Reporter: Hairong Kuang
> Assignee: Hairong Kuang
> Fix For: 0.22.0, 0.20-append
>
>
> This jira allows a sequence file to be created with a configured bytes per checksum. So we could experiment how this parameter could effect the performance of random reads of a sequence file.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.