Avneesh Pandey created CASSANDRA-19345:
------------------------------------------
Summary: Optimize disk seek using min/max column name meta data
when the LIMIT clause is used
Key: CASSANDRA-19345
URL: https://issues.apache.org/jira/browse/CASSANDRA-19345
Project: Cassandra
Issue Type: Improvement
Reporter: Avneesh Pandey
Attachments: Jira-Ticket.pdf
{code:java}
CREATE TABLE tracking_details (
tracker_public_id text,
updated_at timeuuid,
details text,
PRIMARY KEY (tracker_public_id, updated_at)
) WITH CLUSTERING ORDER BY (updated_at DESC){code}
__
{code:java}
INSERT INTO tracking_details (tracker_public_id, updated_at, details)
VALUES('trk_e0544f077f7f4140bf1b961493fa99fe', now(), '{"details":[{...1-3KB
}]}');{code}
__
__
{code:java}
nodetool flush trackingdata tracking_details{code}
{code:java}
INSERT INTO tracking_details (tracker_public_id, updated_at, details)
VALUES('trk_e0544f077f7f4140bf1b961493fa99fe', now(), '{"details":[{...1-3KB
}]}');
INSERT INTO tracking_details (tracker_public_id, updated_at, details)
VALUES('trk_e0544f077f7f4140bf1b961493fa99fe', now(), '{"details":[{...1-3KB
}]}');{code}
__
{code:java}
nodetool flush trackingdata tracking_details{code}
{code:java}
Compaction is disabled-
{'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy',
'enabled': 'false', 'max_threshold': '32', 'min_threshold': '4'}{code}
_Each node is reading from both sstables - *Merged data from memtables and 2
sstables.*_
_{*}{*}I thought this was fixed -
https://issues.apache.org/jira/browse/CASSANDRA-8180_
_But somehow C* doesn't optimizes the read path_
_Tracing attached_
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
