cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From A J <>
Subject OOM
Date Wed, 02 Nov 2011 14:12:40 GMT

For a single node of cassandra(1.0 version) having 15G of data+index,
48GB RAM, 8GB heap and about 2.6G memtable threshold, I am getting OOM
when I have 1000 concurrent inserts happening at the same time.
I have kept concurrent_writes: 128 in cassandra.yaml as there are a
total of 16 cores (suggestion is to keep 8 * number_of_cores).

Can someone give pointers on what needs to be tuned.

Thanks, AJ.

ERROR 00:10:00,312 Fatal exception in thread Thread[Thread-3,5,main]
java.lang.OutOfMemoryError: unable to create new native thread
        at java.lang.Thread.start0(Native Method)
        at java.lang.Thread.start(
        at java.util.concurrent.ThreadPoolExecutor.addWorker(
        at java.util.concurrent.ThreadPoolExecutor.execute(
        at org.apache.cassandra.thrift.CustomTThreadPoolServer.serve(
        at org.apache.cassandra.thrift.CassandraDaemon$

View raw message