siddharthteotia commented on code in PR #9957:
URL: https://github.com/apache/pinot/pull/9957#discussion_r1050124592
##########
pinot-spi/src/main/java/org/apache/pinot/spi/utils/CommonConstants.java:
##########
@@ -291,6 +291,21 @@ public static class QueryOptionKey {
public static final String ORDER_BY_ALGORITHM = "orderByAlgorithm";
+ // Number of rows limit to select in leaf stage.
+ // It has to be greater than 0.
+ // If it is invalid number format, or the number is smaller than or
equal to zero, we use
+ // DEFAULT_LEAF_NODE_LIMIT.
+ public static final String MS_LEAF_LIMIT = "multiStageLeafLimit";
+
+ // Override numGroupLimit used by v1 engine.
+ public static final String NUM_GROUP_LIMIT = "numGroupLimit";
+
+ // Override maxInitialResultCap used by v1 engine.
+ public static final String MAX_INITIAL_RESULT_HOLDER_CAPACITY =
"maxInitialResultCap";
+
+ // Override groupByTrimThreshold used by v1 engine.
+ public static final String GROUP_BY_TRIM_THRESHOLD =
"groupByTrimThreshold";
Review Comment:
I think we have typically played with `numGroupsLimit` vast majority of the
times and for some of our production use cases have set it higher than default
100K to get more accurate results for high cardinality GROUP BY.
I don't remember configuring the `resultHolderCapacity` for any use case.
We have configured `trimThreshold` in one case for performance reasons
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
