Hello,

I have a large application that will require me to perform an average of
4000-5000 TPS with peaks as high as double that.  This application will run
around the clock and can't have any downtime.  As a result I was thinking
that I need a cluster...

but I'm not very familiar with the options.

I've seen PGCluster and CyberCluster.  I think they started in the same
place, but I'm not sure which to choose.  Do you have any thoughts on that
subject?

Also, it seems like the cluster will allow me to scale to the number of
reads that I need, but it seems like I would expect the speed of writes to
the database to decrease in cluster.  Is that right?

Please help me ask the right questions, if these don't make sense...

BTW, I've already got a database up and running.  I've also written some
simple benchmarking scripts in Python that indicate I can achieve about 2100
TPS to a single server (not in cluster).

Thanks,
Daniel
_______________________________________________
Pgcluster-general mailing list
[email protected]
http://pgfoundry.org/mailman/listinfo/pgcluster-general

Reply via email to