[ https://issues.apache.org/jira/browse/CASSANDRA-8897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14564245#comment-14564245 ]
Stefania commented on CASSANDRA-8897: ------------------------------------- The "visits" seem to work for standard commands but not for user profiles. Anyway, the test result is [here|http://cstar.datastax.com/tests/id/4c4e9688-05c3-11e5-b028-42010af0688f]. On my machine I had about 4.5 GB of small sstables, aproximately 33MB each. I used a clustering of 10. I limited the file cache size to 64 MB but I didn't see much difference in GC activity for the read. Anything else I should try? The code is also ready for one more round of review. I implemented the chunk recycling as discussed. I had to flip the free slots via cas to protect against races. Let me know if you prefer to use a separate boolean or other approach. > Remove FileCacheService, instead pooling the buffers > ---------------------------------------------------- > > Key: CASSANDRA-8897 > URL: https://issues.apache.org/jira/browse/CASSANDRA-8897 > Project: Cassandra > Issue Type: Improvement > Components: Core > Reporter: Benedict > Assignee: Stefania > Fix For: 3.x > > Attachments: 8897.yaml, 9240_test_results.txt, > snapshot-1431582436640-cpu-backtraces.png, > snapshot-1431582436640-cpu-calltree-compression-8897.nps, > snapshot-1431582436640-cpu-calltree-compression-trunk.nps > > > After CASSANDRA-8893, a RAR will be a very lightweight object and will not > need caching, so we can eliminate this cache entirely. Instead we should have > a pool of buffers that are page-aligned. -- This message was sent by Atlassian JIRA (v6.3.4#6332)