[jira] [Comment Edited] (KUDU-1587) Memory-based backpressure is insufficient on seek-bound workloads

["Todd Lipcon (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/KUDU-1587?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15677427#comment-15677427 ] 

Todd Lipcon edited comment on KUDU-1587 at 11/18/16 7:53 PM:
-------------------------------------------------------------

Seeing this same issue apparently with Impala workloads. The apply time on one server is creeping up into the 10s of seconds and eventually causes a timeout.

!graph.png!


was (Author: tlipcon):
Seeing this same issue apparently with Impala workloads. The apply time on one server is creeping up into the 10s of seconds and eventually causes a timeout.

> Memory-based backpressure is insufficient on seek-bound workloads
> -----------------------------------------------------------------
>
>                 Key: KUDU-1587
>                 URL: https://issues.apache.org/jira/browse/KUDU-1587
>             Project: Kudu
>          Issue Type: Bug
>          Components: tserver
>    Affects Versions: 0.10.0
>            Reporter: Todd Lipcon
>            Priority: Critical
>         Attachments: graph.png, queue-time.png
>
>
> I pushed a uniform random insert workload from a bunch of clients to the point that the vast majority of bloom filters no longer fit in buffer cache, and the compaction had fallen way behind. Thus, every inserted row turns into 40+ seeks (due to non-compact data) and takes 400-500ms. In this kind of workload, the current backpressure (based on memory usage) is insufficient to prevent ridiculously long queues.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)