btw, cassandra cluster is more stable with turned off multithread compaction.
One node have more keys than other nodes normal node Keyspace: Social Read Count: 65530294 Read Latency: 2.010432367020969 ms. Write Count: 183948607 Write Latency: 0.04994240148825917 ms. Pending Tasks: 0 Table: home_timeline SSTable count: 505 SSTables in each level: [1, 10, 102/100, 392, 0, 0, 0, 0, 0] Space used (live), bytes: 80346741272 Space used (total), bytes: 80365804136 SSTable Compression Ratio: 0.35779981803726807 Number of keys (estimate): 808064 Memtable cell count: 1209392 Memtable data size, bytes: 270605395 Memtable switch count: 763 Local read count: 65530294 Local read latency: 1.807 ms Local write count: 183948608 Local write latency: 0.053 ms Pending tasks: 0 Bloom filter false positives: 12109 Bloom filter false ratio: 0.00633 Bloom filter space used, bytes: 897296 Compacted partition minimum bytes: 73 Compacted partition maximum bytes: 962624926 Compacted partition mean bytes: 629847 Average live cells per slice (last five minutes): 51.0 Average tombstones per slice (last five minutes): 0.0 node with mad gc Keyspace: Social Read Count: 746862 Read Latency: 144.83174594905083 ms. Write Count: 6387984 Write Latency: 2.4869556636334718 ms. Pending Tasks: 0 Table: home_timeline SSTable count: 625 SSTables in each level: [132/4, 10, 99, 384, 0, 0, 0, 0, 0] Space used (live), bytes: 109408488817 Space used (total), bytes: 109408549785 SSTable Compression Ratio: 0.3814217358554331 !!!!! Number of keys (estimate): 5950208 Memtable cell count: 130084 Memtable data size, bytes: 66904752 Memtable switch count: 52 Local read count: 746862 Local read latency: 33.855 ms Local write count: 6387984 Local write latency: 0.050 ms Pending tasks: 0 Bloom filter false positives: 3307633 Bloom filter false ratio: 0.07478 Bloom filter space used, bytes: 3575936 Compacted partition minimum bytes: 73 Compacted partition maximum bytes: 2874382626 Compacted partition mean bytes: 546236 Average live cells per slice (last five minutes): 51.0 Average tombstones per slice (last five minutes): 0.0 -- View this message in context: http://cassandra-user-incubator-apache-org.3065146.n2.nabble.com/Cassandra-mad-GC-tp7592248p7592300.html Sent from the cassandra-u...@incubator.apache.org mailing list archive at Nabble.com.