You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues-all@impala.apache.org by "Surya Hebbar (Jira)" <ji...@apache.org> on 2023/06/20 09:24:00 UTC
[jira] [Comment Edited] (IMPALA-12182) Add CPU utilization time series graph for RuntimeProfile's sampled values
[ https://issues.apache.org/jira/browse/IMPALA-12182?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17735203#comment-17735203 ]
Surya Hebbar edited comment on IMPALA-12182 at 6/20/23 9:23 AM:
----------------------------------------------------------------
The CPU utilization graph has been implemented as a stacked line chart, combining the basis point values for each the node's sys, user and iowait.
The utilization can be shown only by specific nodes as required.
Also, scaling the timeline along with the chart is supported.
This can be used in combination with finding text within browsers, to help with query debugging or exporting the query plan's timeline.
[^query_timeline.mkv]
was (Author: JIRAUSER299620):
The CPU utilization graph has been implemented as a stacked line chart, combining the basis point values from all the nodes.
The utilization can be shown only by specific nodes as required.
Also, scaling the timeline along with the chart is supported.
This can be used in combination with finding text within browsers, to help with query debugging or exporting the query plan's timeline.
[^query_timeline.mkv]
> Add CPU utilization time series graph for RuntimeProfile's sampled values
> -------------------------------------------------------------------------
>
> Key: IMPALA-12182
> URL: https://issues.apache.org/jira/browse/IMPALA-12182
> Project: IMPALA
> Issue Type: New Feature
> Reporter: Surya Hebbar
> Assignee: Surya Hebbar
> Priority: Major
> Attachments: cpu_utilization.png, cpu_utilization_test-1.png, cpu_utilization_test-2.png, query_timeline.mkv, three_nodes.png, three_nodes_zoomed_out.png, timeseries_cpu_utilization_line_plot.mkv, two_nodes.png
>
>
> The RuntimeProfile contains samples of CPU utilization metrics for user, sys and iowait clamped to 64 values (retrieved from the ChunkedTimeSeriesCounter, but sampled similar to SamplingTimeSeriesCounter).
> It would be helpful to see the recent aggregate CPU node utilization samples for each of the different nodes.
> These are sampled every `periodic_counter_update_period_ms`.
> AggregatedRuntimeProfile used in the Thrift profile contains the complete series of values from the ChunkedTimeSeriesCounter samples. But, as this representation is difficult to provide in the JSON, they have been downsampled to 64 values.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscribe@impala.apache.org
For additional commands, e-mail: issues-all-help@impala.apache.org