incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jonathan Ellis <jbel...@gmail.com>
Subject Re: memory question
Date Wed, 24 Mar 2010 19:08:13 GMT
did you check jmx to see if a compaction is going on?

On Mon, Mar 22, 2010 at 5:14 PM, Todd Burruss <bburruss@real.com> wrote:
> after running my cluster for a while performance has become unacceptable,
> 200+ ms for reads.  if running well, i see reads <10ms.  when i run iostat
> the disk is being hammered by reads.  seems like i/o caching isn't even
> being used
>
> avg-cpu:  %user   %nice %system %iowait  %steal   %idle
>            2.81    0.00    1.41   13.62    0.00   82.16
> Device:         rrqm/s   wrqm/s   r/s   w/s    rMB/s    wMB/s avgrq-sz
> avgqu-sz   await  svctm  %util
> cciss/c0d0p1
>                0.00     0.00  0.00  0.00     0.00     0.00    
0.00
> 0.00    0.00   0.00   0.00
> cciss/c0d1p1
>                0.00     0.00 848.50  0.00    13.66     0.00   
32.98
> 21.50   25.23   1.18 100.05
>
> i run top and i see cassandra's memory usage as follows:
>
>   PID USER     PR NI VIRT  RES SHR S %CPU %MEM    TIME+ COMMAND
> 31510 bburruss 19  0 359g  37g 27g S 48.8 80.1  2137:30 java
>
>
> i set -Xmx10g so it isn't java using the memory.  is it mmap i/o?  what
> would be causing the huge memory usage?
> it seems reasonable that the performance is bad because the i/o cache can't
> be used properly.
>

Mime
View raw message