You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Junping Du (JIRA)" <ji...@apache.org> on 2015/03/04 18:58:06 UTC
[jira] [Updated] (YARN-3039) [Aggregator wireup] Implement ATS
app-appgregator service discovery
[ https://issues.apache.org/jira/browse/YARN-3039?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Junping Du updated YARN-3039:
-----------------------------
Attachment: YARN-3039-v2-incomplete.patch
Update a patch (haven't finished yet) to reflect some of discussions above.
Including:
+ maintain the app aggregator info in RMApp, with event model (done)
+ aggregator update in NM-RM heartbeat (done)
+ aggregator update in AM-RM allocation request/response (done)
+ Persistent aggregator update in RMStateStore (fix previous patch)
+ a new API to ResourceTrackerService to register app aggregator to RM (done)
+ adding a new protocol between aggregator and NM
- new proto file (and proto structure for request and response) -- done.
- interfaces: (protocol, request, response)
- AggregatorNodemanagerProtocol (done)
- AggregatorNodemanagerProtocolPBClientImpl (TODO)
- NMAggregatorService (TODO, server impl)
- AggregatorNodemanagerProtocolPB (done)
- AggregatorNodemanagerProtocolPBServiceImpl (done)
- ReportNewAggregatorsInfoRequest/Response (and PBs) (done)
- ReportNewAggregatorsInfoRequestPBImpl (done)
- ReportNewAggregatorsInfoResponse (done)
- ReportNewAggregatorsInfoResponsePBImpl (done)
- AppAggregatorsMap (done)
AppAggregatorsMapPBImpl (done)
Not included yet:
+ NM hosting new protocol
+ Aggregator call new protocol client
+ aggregator info get recovered during NM restart
+ make TimelineClient Observer pattern to observe the change of aggregator address.
Will update the proposal afterwards.
> [Aggregator wireup] Implement ATS app-appgregator service discovery
> -------------------------------------------------------------------
>
> Key: YARN-3039
> URL: https://issues.apache.org/jira/browse/YARN-3039
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: timelineserver
> Reporter: Sangjin Lee
> Assignee: Junping Du
> Attachments: Service Binding for applicationaggregator of ATS (draft).pdf, YARN-3039-no-test.patch, YARN-3039-v2-incomplete.patch
>
>
> Per design in YARN-2928, implement ATS writer service discovery. This is essential for off-node clients to send writes to the right ATS writer. This should also handle the case of AM failures.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)