Re: question regarding queue depths and workers

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 28 February 2018 at 00:23, Jared Walton <jawalking@xxxxxxxxx> wrote:
> If I correctly understand how fio spawns workers then: it takes the
> given queue depth and splits it among workers such that a queue depth
> of 32 with 16 workers results in 16 workers each maintaining a queue
> depth of 2...

I'm afraid this is incorrect. If I have something like

[global]
filename=/dev/nullb0
iodepth=32
ioengine=libaio
direct=1

[worker]
numjobs=16
rw=read

Each worker is trying to achieve a depth of 32 so if they are all
running simultaneously etc. they will generate a total depth of 512 (32 * 16).

> But what happens when a queue depth of 16 is specified with 32 workers?

In this case each worker is trying to achieve a depth of 16 so if all 32
are running simultaneously etc. they will still generate a total depth
of 512.

-- 
Sitsofe | http://sucs.org/~sits/
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Kernel]     [Linux SCSI]     [Linux IDE]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux SCSI]

  Powered by Linux