[
https://issues.apache.org/jira/browse/HDFS-9345?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15939854#comment-15939854
]
SammiChen commented on HDFS-9345:
---------------------------------
Some thoughts so far. The motivation of this JIRA is to help end user tune
performance of EC by skip encoding calculation and decoding calculation(?). So
without EC calculation, the left factors that impact the performance is data
block number, parity block number and cell size. So both dummy schema and dummy
EC policy are required. And for each real EC policy, it's better to have a
corresponding dummy policy.
> Erasure Coding: create dummy coder and schema
> ---------------------------------------------
>
> Key: HDFS-9345
> URL: https://issues.apache.org/jira/browse/HDFS-9345
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Rui Li
> Assignee: SammiChen
> Labels: hdfs-ec-3.0-nice-to-have
>
> We can create dummy coder which does no computation and simply returns zero
> bytes. Similarly, we can create a test-only schema with no parity blocks.
> Such coder and schema can be used to isolate the performance issue to
> HDFS-side logic instead of codec, which would be useful when tuning
> performance of EC.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
