Re: bonnie++ uninterruptible under heavy I/O load

[Simone Piunno]

Alle 17:00, venerdì 11 marzo 2005, Jens Axboe ha scritto:

> > It was after running bonnie++.
>
> Are you sure? It lists only 190 commands run since it initialized,
> that's basically nothing.

Yes I'm sure!

Isn't this number the MAX value ever samplef for "Current # commands on 
controller"?  It doesn't seem to be the counter of how many commands were 
processed since start.

To doublecheck, I're run bonnie one more time.
This one was taken while bonnie was running:

cciss0: HP Smart Array 5i Controller
Board ID: 0x40800e11
Firmware Version: 2.56
IRQ: 18
Logical drives: 1
Current Q depth: 0
Current # commands on controller: 132
Max Q depth since init: 147
Max # commands on controller since init: 190
Max SG entries since init: 31

rereading the file, the only moving number was "Current # commands on 
controller", basically oscillating between 10 and 150.

> 6 was the highest depth, should not be enough to cause serious latency
> issues unless it was constantly at 6. 

As you see, after more runs Max Q raised to 147, but I'm sure it was 6 after 
that first run, and in that 1st run responsiveness was already sluggish.

Anyway, I wouldn't focus too much on CCISS, cause the only thing that's 
specific of this hardware is the long delay killing bonnie.  
Bad responsiveness has been found almost everywhere else.

-- 
Simone Piunno, chief architect
Wireless Solutions SPA - DADA group
Europe HQ, via Castiglione 25 Bologna
web:www.dada-ws.com tel:+39512966811 fax:+39512966800
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/