|
>
> there's two answers here
>
> o just treat the cluster as if it were a single server and
> run some test (this is what you've done). For n realservers
> you hope to get n times the throughput of 1 realserver.
>
> o there's no good tool for showing that the realservers
> in an LVS are working together. For this reason I did
> my tests on LVS performance
>
>
http://www.linuxvirtualserver.org/Joseph.Mack/performance/single_realserver_performance.html
>
> with only 1 realserver.
>
> The best tool I know for showing that a cluster is clustering is
> polygraph
>
>
http://www.austintek.com/LVS/LVS-HOWTO/HOWTO/LVS-HOWTO.performance.html#performance_testing_tools
>
> although it is difficult to setup.
>
>
Another cool tool for monitoring the Real Servers is Ganglia. (With version
2) you run gmond monitoring daemons on each RS and a single gmetad daemon to
poll the gmonds on a server (that is running Apache) outside the cluster.
Then with the Ganglia Web Frontend you get great color graphs that help you
to find "hot spots". You can then write your own gmetric script to create
your own custom graph for anything happening on the Real Servers (I suppose
you could cull the Apache logs for "GET" operations--check out the Gmetric
Script Repository).
Ganglia Page:
http://ganglia.sourceforge.net/
Sample Ganglia page:
http://meta.rocksclusters.org/Rocks-Network/
Gmetric Script Repository:
http://ganglia.sourceforge.net/gmetric/
Incidentally, you can also add the gexec program to submit batch jobs (like
cron jobs) to the least loaded real server or to all nodes simultaneously.
I know there are a lot of tools like this out there, but this one works
great for me in production.
--Karl
|
