Yves Glodt wrote:

>
> We intensively queried our 4GB database from 8 connections (thus 8
> fb_inet_server processes were running). But, then we ended up with
> freezes (kernel panic), even once of the complete cluster. (It consists
> of one master and one node atm, running clusterknoppix.) So for now we
> forget the idea of FB on a cluster...

yes, I think that's right, though the open source distributed lock
manager may prove to be a workable solution. At the moment, none of the
developers has a system that we can use to test it, and making it work
will be non-trivial I'm sure, so it's not likely to be addressed
immediately.

>
> As I read Vulcan plans to introduce strong SMP capabilities. Will it
> also be able to run in a cluster?

Not without a distributed lock manager.

> Or is the preferable way to use big (8 or 16 CPUs) SMP machines in case
> where huge load is expected?

Yes.

>

Regards,


Ann