> Greg Lehey wrote:
> > 
> > FWIW, I was never happy with the removal of block devices either.  I
> > was shouted down with "can you point to any one use they are?", to
> > which I replied "just because I don't know of one doesn't mean there
> > isn't one, or that there will never be one in the future".  This is an
> > example where they could presumably be useful.
> Doesn't Oracle run MUCH better when given raw block disk devices to store
> data on?

Oracle wants to cache it's own data, it doesn't want the buffer cache 
behind it.

>  Could this have lead to some of the poor performance Mike Smith
> was seeing when testing this summer?

No, we were layering over the filesystem, however filesystem load was 
insignificant.  The Oracle performance issues are well known, and this 
isn't one of them.

... every activity meets with opposition, everyone who acts has his
rivals and unfortunately opponents also.  But not because people want
to be opponents, rather because the tasks and relationships force
people to take different points of view.  [Dr. Fritz Todt]

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Reply via email to