You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Feng Qu <ma...@gmail.com> on 2011/12/12 06:00:46 UTC

read/write counts

Hello,

When I use nodetool cfstas, I see read/write for both keyspace and column family. I assume both number are counted across the ring, but I saw different read/write counts showed on one node comparing to other 7 nodes.

node 1,2,4-8:
Keyspace: XXXXks
        Read Count: 44285565
        Read Latency: 1.4984792287509485 ms.
        Write Count: 161096052
        Write Latency: 0.006321300412750028 ms.
        Pending Tasks: 0
                Column Family: XXXXEvents
                Read Count: 44219534

                Read Latency: NaN ms.
                Write Count: 43245679
                Write Latency: 0.012 ms.

Node 3:
Keyspace: XXXXks

        Read Count: 44190641
        Read Latency: 1.738313580810018 ms.
        Write Count: 136735281
        Write Latency: 0.007389342133285994 ms.
                Column Family: XXXXEvents

                Read Count: 44125278

                Read Latency: NaN ms.
                Write Count: 36991005
                Write Latency: 0.010 ms.

So my questions are:
1) Are KS level counts and CF level counts for whole cluster or just for an individual node?
2) Why I see different counts from different nodes if counts are at KS level?
 
Feng

Re: read/write counts

Posted by Peter Schuller <pe...@infidyne.com>.
> 1) Are KS level counts and CF level counts for whole cluster or just for an
> individual node?

Individual node.

Also note that the CF level counts will refer to local reads/writes
submitted to the node, while the statistics you get from StorageProxy
(in JMX) are for requests routed. In general, you will see a
magnification by a factor of RF on the local statistics (in aggregate)
relative to the StorageProxy stats.

-- 
/ Peter Schuller (@scode, http://worldmodscode.wordpress.com)