[ https://issues.apache.org/jira/browse/CASSANDRA-14227?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17619357#comment-17619357 ]
Berenguer Blasi edited comment on CASSANDRA-14227 at 10/18/22 9:58 AM: ----------------------------------------------------------------------- Hi, I have been asked to run some profiling to see the impact 'longs' would have. I have tested on a single node doing inserts with a TTL + selects against that for 4m at around 100Kops/s after a warm up run. The results are virtual identical to both trunk and 14227 both at jfr and stress tool perf reporting Trunk: !screenshot-1.png! 14227 !screenshot-2.png! If anything 14227's number are slightly better but probably just test env noise. GC pauses, total times, latencies, etc are all identical as well. The test CQL was: {noformat} CQL|INSERT INTO test.test (id, type, text) VALUES ($RANDOM_20000, $RANDOM_10, 'TTL Profiling') USING TTL $RANDOM_500000 CQL|SELECT id, type, text FROM test.test WHERE id=$RANDOM_20000 {noformat} where RANDOM_X means a random number up to X. Here we can see inserts, inserts colliding and selects. I will be happy to repeat the test if anybody has any suggestions. I am not posting the jfr for security reasons as the env vars section contains sensitive data (call me paranoid) but I can share them with known people on request. EDIT: I have been asked to confirm both runs are under {{memtable_allocation_type: heap_buffers}} so it is on heap. was (Author: bereng): Hi, I have been asked to run some profiling to see the impact 'longs' would have. I have tested on a single node doing inserts with a TTL + selects against that for 4m at around 100Kops/s after a warm up run. The results are virtual identical to both trunk and 14227 both at jfr and stress tool perf reporting Trunk: !screenshot-1.png! 14227 !screenshot-2.png! If anything 14227's number are slightly better but probably just test env noise. GC pauses, total times, latencies, etc are all identical as well. The test CQL was: {noformat} CQL|INSERT INTO test.test (id, type, text) VALUES ($RANDOM_20000, $RANDOM_10, 'TTL Profiling') USING TTL $RANDOM_500000 CQL|SELECT id, type, text FROM test.test WHERE id=$RANDOM_20000 {noformat} where RANDOM_X means a random number up to X. Here we can see inserts, inserts colliding and selects. I will be happy to repeat the test if anybody has any suggestions. I am not posting the jfr for security reasons as the env vars section contains sensitive data (call me paranoid) but I can share them with known people on request. > Extend maximum expiration date > ------------------------------ > > Key: CASSANDRA-14227 > URL: https://issues.apache.org/jira/browse/CASSANDRA-14227 > Project: Cassandra > Issue Type: Bug > Components: Legacy/Local Write-Read Paths > Reporter: Paulo Motta (Deprecated) > Assignee: Berenguer Blasi > Priority: Urgent > Fix For: 4.x > > Attachments: screenshot-1.png, screenshot-2.png > > > The maximum expiration timestamp that can be represented by the storage > engine is > 2038-01-19T03:14:06+00:00 due to the encoding of {{localExpirationTime}} as > an int32. > On CASSANDRA-14092 we added an overflow policy which rejects requests with > expiration above the maximum date as a temporary measure, but we should > remove this limitation by updating the storage engine to support at least the > maximum allowed TTL of 20 years. -- This message was sent by Atlassian Jira (v8.20.10#820010) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org