[DocDB] Create kLatest Metric Aggregation Function #25437
Labels
area/docdb
YugabyteDB core features
kind/enhancement
This is an enhancement of an existing feature
priority/medium
Medium priority issue
status/awaiting-triage
Issue awaiting triage
Comments
yusong-yan
added
area/docdb
yugabyte-ci
added
kind/enhancement
yusong-yan
added a commit
that referenced
this issue
Dec 26, 2024
…etrics from Sum to Max Summary: `kSum` aggregation is currently used for many xCluster metrics, but this leads to overflows and inaccuracies for metrics like `last_read_hybridtime`, which should instead use maximum aggregation `kMax`. Next Step [[ #25437 | #25437 ]]: For optimal accuracy in the future, we should implement a `kLatest` aggregation function, which would require metric storing both the latest value and its corresponding timestamp. Jira: DB-14654 Test Plan: Jenkins Reviewers: hsunder, xCluster, jhe Reviewed By: jhe Subscribers: rthallam, ybase, ycdcxcluster, slingam Differential Revision: https://phorge.dev.yugabyte.com/D40873
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Jira Link: DB-14674
Description
As discussed in #25422, many xCluster metrics use
kSumaggregation, which leads to overflows and inaccuracies, particularly for metrics likelast_read_hybridtime, As a short-term solution, we are switching to use maximum aggregationkMax.This issue proposes a longer-term solution: implementing a
kLatestaggregation function to provide the most accurate aggregated value. This function will also require metrics to store both the latest value and the corresponding timestamp.Issue Type
kind/enhancement
Warning: Please confirm that this issue does not contain any sensitive information
The text was updated successfully, but these errors were encountered: