Hello Jeff,
On Fri, Sep 24, 2010 at 7:12 AM, Schweiger, Jeff - ES
<Jeff.Schweiger@stripped> wrote:
> Thanks for the response... One follow up question:
>
>>You know, 60k queries/min is not that large for cluster although it
>>depends on the queries. How is the load of your servers at 60k/s? Can
>>you estimate what will be the first bottleneck?
>
> I guess that is part of my problem... I would like to get an idea how many
> queries cluster *can* handle. At my current query rate, the load on my servers is
> very low. I know performance is highly variable and is app dependant, hardware
> dependant and query type dependant. For example, currently I am running on old Sun
> 2000's, and know I will get a boost if I go to a newer Intel multi core platform.
> But I think I have tuned by app and cluster to perform well.
>
> So given a 4 data node 2 mysqld cluster, can I expect 500,000 queries per second, can
> I expect more? I would love to know what type of performance people have seen.
Due to all the specifics of any particular deployment, the only way to
answer such a thing is to obtain more information about how your
cluster is performing.
Measure your loads (cpu/diskio), key query response times at
60,000/min. Then double your test load to 120,000 and measure the same
data, and repeat until you either find a performance problem or a
configuration limitation. This way you should get some profiling
information and specific details about how your system will perform,
with your app. Its simply not possible to compare generic benchmarks
to your application.
Some benchmark tools allow custom sets of queries to be run if you are
unable to easily simulate your application loads.
Adam
