well, you have two flushes and a compaction there, but none of them lasts more than a few seconds (out of the half hour period) so i'm guessing that's not the problem.
On Thu, Jul 29, 2010 at 1:53 AM, Thorvaldsson Justus < justus.thorvalds...@svenskaspel.se> wrote: > I am testing on one node only right now and simply adding data and > reading it. > > There is no real problem but I feel I get differencing results from the > tests depending on something I would like to know what it is. > > What is happening between two “points in time” > > > > àHere is a point with Slower result > > INFO 08:12:02,113 Creating new commitlog segment > /srv/cassandra/commitlog/CommitLog-1280383922113.log > > INFO 08:12:06,872 Completed flushing > /srv/cassandra/data/MyGames/Game1-207-Data.db > > INFO 08:12:11,950 Compacted to > /srv/cassandra/data/MyGames/EssentialGame1-57-Data.db. 321936296/319836199 > bytes for 2000 keys. Time: 10659ms. > > INFO 08:13:23,221 Creating new commitlog segment > /srv/cassandra/commitlog/CommitLog-1280384003221.log > > INFO 08:13:24,401 Game1 has reached its threshold; switching in a fresh > Memtable at > CommitLogContext(file='/srv/cassandra/commitlog/CommitLog-1280384003221.log', > position=7180218) > > INFO 08:13:24,402 Enqueuing flush of memtable-ga...@2027049345(65162130 > bytes, 3145758 operations) > > INFO 08:13:24,406 Writing memtable-ga...@2027049345(65162130 bytes, > 3145758 operations) > > INFO 08:13:28,365 Completed flushing > /srv/cassandra/data/MyGames/Game1-208-Data.db > > INFO 08:13:35,327 Deleted > /srv/cassandra/data/MyGames/EssentialGame1-56-Data.db > > INFO 08:13:35,348 Deleted > /srv/cassandra/data/MyGames/EssentialGame1-52-Data.db > > INFO 08:13:35,369 Deleted > /srv/cassandra/data/MyGames/EssentialGame1-53-Data.db > > INFO 08:13:35,390 Deleted > /srv/cassandra/data/MyGames/EssentialGame1-55-Data.db > > INFO 08:38:47,378 Creating new commitlog segment > /srv/cassandra/commitlog/CommitLog-1280385527378.log > > INFO 08:49:03,295 Game1 has reached its threshold; switching in a fresh > Memtable at > CommitLogContext(file='/srv/cassandra/commitlog/CommitLog-1280385527378.log', > position=25079957) > > INFO 08:49:03,296 Enqueuing flush of memtable-ga...@1607706649(65162710 > bytes, 3145786 operations) > > INFO 08:49:03,296 Writing memtable-ga...@1607706649(65162710 bytes, > 3145786 operations) > > INFO 08:49:07,948 Completed flushing > /srv/cassandra/data/MyGames/Game1-209-Data.db > > INFO 08:49:07,949 Compacting > [org.apache.cassandra.io.SSTableReader(path='/srv/cassandra/data/MyGames/Game1-206-Data.db'),org.apache.cassandra.io.SSTableReader(path='/srv/cassandra/data/MyGames/Game1-207-Data.db'),org.apache.cassandra.io.SSTableReader(path='/srv/cassandra/data/MyGames/Game1-208-Data.db'),org.apache.cassandra.io.SSTableReader(path='/srv/cassandra/data/MyGames/Game1-209-Data.db')] > > INFO 08:49:23,920 Compacted to > /srv/cassandra/data/MyGames/Game1-210-Data.db. 443120374/443120374 bytes > for 898844 keys. Time: 15970ms. > > INFO 08:50:17,479 Creating new commitlog segment > /srv/cassandra/commitlog/CommitLog-1280386217479.log > > àHere is a point with faster result again > > > > This is no absolute position in time, but I would like to know because I > could probably configure Cassandra better. > > Regards from Sweden > > Justus Thorvaldsson > > AB SVENSKA SPEL > 106 10 Stockholm > Sturegatan 11, Sundbyberg > Växel +46 8 757 77 00 > http://svenskaspel.se > -- Jonathan Ellis Project Chair, Apache Cassandra co-founder of Riptano, the source for professional Cassandra support http://riptano.com
