Ann
I understand, I thought true SMP was referring to assigning each client's
request to all CPU at once, instead of balacing the clients requests across
processors. So the current version (FB 3) is probably going to outperform the
FB 2.54 when there are concurrent requests passed to the server engine because
each request will go to a different processor, correct? The more concurrent
requests we have the more we are going to notice the difference, right?
Cheers
Fabian
----- Original Message -----
From: Ann Harrison aharri...@ibphoenix.com [firebird-support]
To: firebird-support@yahoogroups.com
Sent: Thursday, May 26, 2016 4:53 AM
Subject: Re: [firebird-support] Re: Is Firebird 3 ready for Production?
On Wed, May 25, 2016 at 1:11 PM, fabia...@itbizolutions.com.au
[firebird-support] <firebird-support@yahoogroups.com> wrote:
Now on the flip side, the performance sucks, it is worst than with FB
2.54, and when looking at the task manager on windows it appears only one
processor it doing the job, as if the code was not SMP enabled.... very
strange.
One possibility is that you're testing V3.0 SuperServer single user. In
V3.0, Firebird is multi-threaded at the client statement level. It does not
decompose queries and schedule the pieces on different processors. That means
that a full-table scan runs on only one processor. Two simultaneous full-table
scans will run on two processors.
Good luck,
Ann
