[
https://issues.apache.org/jira/browse/HDFS-8594?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Walter Su resolved HDFS-8594.
-----------------------------
Resolution: Not A Problem
> Erasure Coding: cache ErasureCodingZone
> ---------------------------------------
>
> Key: HDFS-8594
> URL: https://issues.apache.org/jira/browse/HDFS-8594
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Walter Su
> Assignee: Walter Su
>
> scenario 1:
> We have 100m files in a EC zone. Everytime we open a file, we need to get
> ECSchema(need get ECZone first). So getting EC zone is frequent.
> scenario 2:
> We have a EC zone "/d1". We have a file in "/d1/d2/d3/.../dN". We have to
> search from xAttrs from dN, dN-1, ..., d3, d2, d1, until we find a EC zone
> from d1's xAttr.
> It's better we cache the EC zone, like EncryptionZoneManager#encryptionZones
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
