Improve ExplicitTags queries with FuzzyFilters #1896
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This patch reworks how FuzzyFilters are used in the case of ExplicitTags queries:
It brings noticeable performance improvement when filtering on tag value for metrics with high cardinality (eg. a query for metric for a single host, with host tag cardinality of 20000 which took 9s is down to 250ms)
The list of fuzzy filters is constructed by a new method
QueryUtil.buildFuzzyFilters(row_key_literals)
.It is used for filtering all combinations of tags provided by the caller.
So for 3 tags
cluster
,service
andvm
, with filtering values:cluster=dev|qa|test
,service=mysvc
andvm=web|db
, a list of 8 (= 3 x 1 x 2) FuzzyFilters covering all combinations will be returned.A few remarks:
QueryUtil.getRowKeyUIDRegex(row_key_literals, explicit_tags)
: simplified variant stripped of everything related to FuzzyFiltersQueryUtil.getRowKeyUIDRegex(group_bys, row_key_literals)
andQueryUtil.getRowKeyUIDRegex(group_bys, row_key_literals, explicit_tags, fuzzy_key, fuzzy_mask)
are left unchanged as these are public API (but not used anymore)