adelapena commented on code in PR #1891:
URL: https://github.com/apache/cassandra/pull/1891#discussion_r1129238419
##########
src/java/org/apache/cassandra/db/RangeTombstoneList.java:
##########
@@ -58,12 +58,12 @@ public class RangeTombstoneList implements
Iterable<RangeTombstone>, IMeasurable
private ClusteringBound<?>[] starts;
private ClusteringBound<?>[] ends;
private long[] markedAts;
- private int[] delTimes;
+ private long[] delTimes;
Review Comment:
IIRC a boolean array takes one byte per element, plus the array overhead.
That'd be 4+1 bytes per tombstone, lesser than the 8 bytes of using longs.
Although taking advantage of the scarcity of invalid values to just have a set
of invalid tombstone indexes sound good too.
I'm not yet convinced that trying to reduce the increase of the memtable
size is early optimization. Do the benchmark on jira include range tombstones?
--
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]
