Tagged queueing = lower write throughput?

[P.C. Uiterlinden]

Albert,

Thanks for replying to my mail. Now I at least know I'm not the only one
experiencing this kind of behavior.

Still I think it's strange that tagged queuing is able to slow things
down. I can't think of an obvious reason. Anyone cares to come up with
one?

Meanwhile, I'll download the manual of the hard disk to see if I can
discover something for my self.

Regards,

Paul.


Albert Max Lai wrote:
> 
> On Thursday, 2 September 1999, P.C. Uiterlinden wrote:
> 
> > While playing with RAID0, I found out that enabling tagged queueing
> > deteriorates write throughput considerably. The MD driver seems not
> 
> I have noticed that one has to play around with the queueing depth to
> get the highest throughput. With my IBM DGVSO9U (9ZX), the optimal
> queuing depth is between 12-14. With my Quantum Viking 4.5 NSE, having
> tagged queueing turned on slows it down.


To Unsubscribe: send mail to majordomo at FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message