Performance Regression or Improvement: gbk_python_batch_load_test_2gb_of_100B_records:runtime

[github-actions]

Performance change found in the
test: gbk_python_batch_load_test_2gb_of_100B_records for the metric: runtime.

For more information on how to triage the alerts, please look at
Triage performance alert issues section of the README.

Test description: GroupByKey Python Load test - 2GB of 100B records python | GBK | Medium records (100B) Test Dashboard - http://metrics.beam.apache.org/d/UYZ-oJ3Zk/gbk-load-tests?orgId=1&var-processingType=batch&var-sdk=python&from=now-90d&to=now&viewPanel=3 Test link -

beam/.test-infra/jenkins/job_LoadTests_GBK_Python.groovy

Line 60 in 5e38dec

test : 'apache_beam.testing.load_tests.group_by_key_test',

timestamp: Sun Oct  1 13:54:18 2023, metric_value: 150.00
timestamp: Sat Sep 30 13:54:07 2023, metric_value: 146.00
timestamp: Fri Sep 29 13:56:16 2023, metric_value: 172.00
timestamp: Thu Sep 28 13:55:29 2023, metric_value: 162.00
timestamp: Wed Sep 27 13:55:39 2023, metric_value: 151.00 <---- Anomaly
timestamp: Tue Sep 26 14:02:14 2023, metric_value: 208.00
timestamp: Mon Sep 25 13:57:29 2023, metric_value: 215.00
timestamp: Sun Sep 24 14:00:47 2023, metric_value: 211.00
timestamp: Sat Sep 23 13:55:23 2023, metric_value: 149.00
timestamp: Fri Sep 22 13:58:35 2023, metric_value: 236.00
timestamp: Thu Sep 21 13:57:12 2023, metric_value: 150.00
timestamp: Wed Sep 20 13:55:46 2023, metric_value: 173.00
timestamp: Tue Sep 19 13:58:41 2023, metric_value: 166.00
timestamp: Mon Sep 18 14:03:45 2023, metric_value: 195.00
timestamp: Sun Sep 17 14:02:30 2023, metric_value: 218.00

[AnandInguva]

Noisy data? Looking at the graph, I see that the data has noise.