From: Paolo Valente <paolo.valente@unimore.it>
To: Pavel Machek <pavel@ucw.cz>
Cc: axboe@kernel.dk, linux-kernel@vger.kernel.org
Subject: Re: [RESEND][RFC] BFQ I/O Scheduler
Date: Thu, 17 Apr 2008 08:14:35 +0200 [thread overview]
Message-ID: <4806EACB.7040408@unimore.it> (raw)
In-Reply-To: <20080416184441.GA3923@ucw.cz>
Pavel Machek ha scritto:
>
>> In the first type of tests, to achieve a higher throughput than CFQ
>> (with the default 100 ms time slice), the maximum budget for BFQ
>> had to be set to at least 4k sectors. Using the same value for the
>>
>
> Hmm, 4k sectors is ~40 seconds worst case, no? That's quite long...
>
Actually, in the worst case among our tests, the aggregate throughput
with 4k sectors was ~ 20 MB/s, hence the time for 4k sectors ~ 4k * 512
/ 20M = 100 ms.
About these test cases we repeated several measures of the aggregate
throughput with simultaneous file reads from 2 to 5 (and other varying
parameters). The lowest aggregate throughput for 4k sectors (~ 20 MB/s)
was achieved in case of 2, 128 MB long, files, lying on two slices at
the maximum distance from each other (more details on the experiments,
testbed and so on at http://algo.ing.unimo.it/people/paolo/disk_sched/).
I hope I didn't misunderstand your point.
next prev parent reply other threads:[~2008-04-17 6:14 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-01 15:29 [RESEND][RFC] BFQ I/O Scheduler Fabio Checconi
2008-04-15 8:22 ` Jens Axboe
2008-04-15 9:11 ` Fabio Checconi
2008-04-15 12:42 ` Jens Axboe
2008-04-15 18:08 ` Fabio Checconi
2008-04-16 6:48 ` Paolo Valente
2008-04-18 1:26 ` Aaron Carroll
2008-04-16 18:44 ` Pavel Machek
2008-04-17 6:14 ` Paolo Valente [this message]
2008-04-17 7:10 ` Jens Axboe
2008-04-17 8:26 ` Paolo Valente
2008-04-17 8:30 ` Jens Axboe
2008-04-17 9:24 ` Paolo Valente
2008-04-17 9:27 ` Jens Axboe
2008-04-17 10:19 ` Aaron Carroll
2008-04-17 10:21 ` Jens Axboe
2008-04-17 11:30 ` Fabio Checconi
2008-04-17 15:19 ` Avi Kivity
2008-04-17 15:47 ` Paolo Valente
2008-04-17 15:51 ` Avi Kivity
2008-04-17 18:12 ` Paolo Valente
2008-04-17 23:44 ` Aaron Carroll
2008-04-17 10:24 ` Aaron Carroll
2008-04-17 11:14 ` Fabio Checconi
2008-04-17 12:14 ` Aaron Carroll
2008-04-17 13:54 ` Jens Axboe
2008-04-17 15:18 ` Paolo Valente
2008-04-17 8:48 ` Pavel Machek
2008-04-17 8:57 ` Jens Axboe
2008-04-17 9:14 ` Fabio Checconi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4806EACB.7040408@unimore.it \
--to=paolo.valente@unimore.it \
--cc=axboe@kernel.dk \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.