aurora-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hudson (JIRA)" <>
Subject [jira] [Commented] (AURORA-122) Introduce a task store secondary index on slave host
Date Mon, 27 Jan 2014 22:59:39 GMT


Hudson commented on AURORA-122:

SUCCESS: Integrated in Aurora #73 (See [])
AURORA-122: Add an abstraction for secondary indices, introduce an index on slave host. (wfarner:
rev 4efa6748c949d76ea1c118fc7b19bb2490398219)
* src/main/java/org/apache/aurora/scheduler/async/
* src/main/java/org/apache/aurora/scheduler/storage/mem/
* src/main/java/org/apache/aurora/scheduler/http/
* src/test/java/org/apache/aurora/scheduler/storage/mem/
* src/main/java/org/apache/aurora/scheduler/base/

> Introduce a task store secondary index on slave host
> ----------------------------------------------------
>                 Key: AURORA-122
>                 URL:
>             Project: Aurora
>          Issue Type: Story
>          Components: Reliability, Scheduler
>            Reporter: Bill Farner
>            Assignee: Bill Farner
>              Labels: O2
> We've observed the {{resourceOffers}} callback taking longer than desired in large clusters.
 Timing has indicated that they take as long as 100ms regularly, with occasional calls taking
as long as a minute.  We've narrowed down the majority of time spent to be performing expensive
task store queries by slave host (in {{GcExecutorLauncher}}.  When lots of hosts and tasks
are present, these calls should be considerably faster if we maintain a index on slave host
to avoid full task store scans.

This message was sent by Atlassian JIRA

View raw message