[
https://issues.apache.org/jira/browse/HDFS-15658?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
chuanjie.duan updated HDFS-15658:
---------------------------------
Attachment: HDFS-15658-branch-2.7.patch
Status: Patch Available (was: Open)
> Improve datanode capability balancing
> -------------------------------------
>
> Key: HDFS-15658
> URL: https://issues.apache.org/jira/browse/HDFS-15658
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs
> Reporter: chuanjie.duan
> Priority: Major
> Attachments: HDFS-15658-branch-2.7.patch
>
>
> How about adjust the order of choosing replication to deletion?
> Is there any other meaning, choosing "oldestHeartbeatStorage" first?
>
> public DatanodeStorageInfo chooseReplicaToDelete(
> Collection<DatanodeStorageInfo> moreThanOne,
> Collection<DatanodeStorageInfo> exactlyOne,
> final List<StorageType> excessTypes,
> Map<String, List<DatanodeStorageInfo>> rackMap) {
> ......
> final DatanodeStorageInfo storage;
> if (minSpaceStorage != null) {
> storage = minSpaceStorage;
> } else if (oldestHeartbeatStorage != null) {
> storage = oldestHeartbeatStorage;
> } else {
> return null;
> }
> excessTypes.remove(storage.getStorageType());
> return storage;
> }
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
