aurora-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David McLaughlin <da...@dmclaughlin.com>
Subject Re: Review Request 59699: Improve task history pruning by batch deleting tasks
Date Thu, 01 Jun 2017 00:57:22 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/59699/#review176568
-----------------------------------------------------------



Can you also provide metrics showing how much time is spent in the coalesce method? And show
before/after results using the scale test?

- David McLaughlin


On June 1, 2017, 12:18 a.m., Kai Huang wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/59699/
> -----------------------------------------------------------
> 
> (Updated June 1, 2017, 12:18 a.m.)
> 
> 
> Review request for Aurora, David McLaughlin and Santhosh Kumar.
> 
> 
> Bugs: AURORA-1929
>     https://issues.apache.org/jira/browse/AURORA-1929
> 
> 
> Repository: aurora
> 
> 
> Description
> -------
> 
> Improve task history pruning by batch deleting tasks.
> 
> The `'aurora_admin prune_tasks'` endpoint seems to be very slow when the cluster has
a large number of inactive tasks.
> 
> This CR batches all removeTasks operations and execute them all at once to avoid additional
cost of coalescing. The fix will also benefit implicit task history pruning since it has similar
underlying implementation. See https://issues.apache.org/jira/browse/AURORA-1929 for more
information and details.
> 
> 
> Diffs
> -----
> 
>   src/main/java/org/apache/aurora/scheduler/state/StateManagerImpl.java 73878210f9028901fda3b08e66c6a63c24260d35

> 
> 
> Diff: https://reviews.apache.org/r/59699/diff/1/
> 
> 
> Testing
> -------
> 
> ./build-support/jenkins/build.sh
> 
> 
> Thanks,
> 
> Kai Huang
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message