From: Vladislav Bolkhovitin <vst-d+Crzxg7Rs0@public.gmane.org>
To: Bart Van Assche <bart.vanassche-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: James.Bottomley-JuX6DAaQMKPCXq6kfMZ53/egYHeGw8Jk@public.gmane.org,
stgt-devel-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org,
linux-scsi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
FUJITA Tomonori
<fujita.tomonori-Zyj7fXuS5i5L9jVzuh4AOg@public.gmane.org>,
scst-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
Subject: Re: Performance of SCST versus STGT
Date: Tue, 22 Jan 2008 18:23:25 +0300 [thread overview]
Message-ID: <47960A6D.6010602@vlnb.net> (raw)
In-Reply-To: <e2e108260801220432l353b1d76xd2707b5e6f336aef-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Bart Van Assche wrote:
> On Jan 22, 2008 12:33 PM, Vladislav Bolkhovitin <vst-d+Crzxg7Rs0@public.gmane.org
> <mailto:vst-d+Crzxg7Rs0@public.gmane.org>> wrote:
>>
>> What are the new SRPT/iSER numbers?
>
> You can find the new performance numbers below. These are all numbers
> for reading from the remote buffer cache, no actual disk reads were
> performed. The read tests have been performed with dd, both for a block
> size of 512 bytes and of 1 MB. The tests with small block size learn
> more about latency, while the tests with large block size learn more
> about the maximal possible throughput.
If you want to compare performance of 512b vs 1MB blocks, your
experiment isn't fully correct. You should use "iflag=direct" dd option
for that.
> .............................................................................................
>
> . . STGT read SCST read . STGT
> read SCST read .
> . . performance performance .
> performance performance .
> . . (0.5K, MB/s) (0.5K, MB/s) . (1 MB,
> MB/s) (1 MB, MB/s) .
> .............................................................................................
> . Ethernet (1 Gb/s network) . 77 78 .
> 77 89 .
> . IPoIB (8 Gb/s network) . 163 185 .
> 201 239 .
> . iSER (8 Gb/s network) . 250 N/A .
> 360 N/A .
> . SRP (8 Gb/s network) . N/A 421 .
> N/A 683 .
> .............................................................................................
>
> My conclusion from the above numbers: the performance difference between
> STGT and SCST is small for a Gigabit Ethernet network. The faster the
> network technology, the larger the difference between SCST and STGT.
This is what I expected
> Bart.
next prev parent reply other threads:[~2008-01-22 15:23 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-17 9:27 Performance of SCST versus STGT Bart Van Assche
2008-01-17 9:40 ` FUJITA Tomonori
2008-01-17 9:48 ` Vladislav Bolkhovitin
2008-01-17 10:05 ` FUJITA Tomonori
[not found] ` <20080117190558K.fujita.tomonori-Zyj7fXuS5i5L9jVzuh4AOg@public.gmane.org>
2008-01-17 10:34 ` Vladislav Bolkhovitin
[not found] ` <478F2F46.9040103-d+Crzxg7Rs0@public.gmane.org>
2008-01-17 12:29 ` Robin Humble
2008-01-17 13:44 ` [Scst-devel] [Stgt-devel] " Vladislav Bolkhovitin
[not found] ` <20080117122956.GA3567-Td5ZOp7sT3Xw02mFwxTg32+DJq1SqhBbsOSz5zK2v9k@public.gmane.org>
2008-01-17 14:43 ` Bart Van Assche
2008-01-17 17:45 ` Pete Wyckoff
[not found] ` <20080117174542.GC29650-pxmRpbKlMIQ@public.gmane.org>
2008-01-18 10:30 ` Bart Van Assche
2008-01-18 12:08 ` Vladislav Bolkhovitin
2008-01-20 9:36 ` Bart Van Assche
[not found] ` <e2e108260801200136g7f17b8a0g89a54cc1d73bbc34-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2008-01-21 12:07 ` Vladislav Bolkhovitin
2008-01-22 3:26 ` FUJITA Tomonori
[not found] ` <20080122122657R.fujita.tomonori-Zyj7fXuS5i5L9jVzuh4AOg@public.gmane.org>
2008-01-22 7:50 ` Bart Van Assche
2008-01-22 11:33 ` Vladislav Bolkhovitin
[not found] ` <4795D479.1080805-d+Crzxg7Rs0@public.gmane.org>
2008-01-22 11:48 ` FUJITA Tomonori
2008-01-22 12:20 ` Vladislav Bolkhovitin
2008-01-22 15:14 ` Bart Van Assche
2008-01-22 10:04 ` Bart Van Assche
2008-01-22 11:33 ` Vladislav Bolkhovitin
[not found] ` <4795D4A7.5000105-d+Crzxg7Rs0@public.gmane.org>
2008-01-22 12:32 ` Bart Van Assche
[not found] ` <e2e108260801220432l353b1d76xd2707b5e6f336aef-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2008-01-22 15:23 ` Vladislav Bolkhovitin [this message]
2008-01-24 7:06 ` Robin Humble
2008-01-24 10:36 ` [Stgt-devel] " Bart Van Assche
2008-01-24 11:10 ` Vladislav Bolkhovitin
[not found] ` <4798720E.4020802-d+Crzxg7Rs0@public.gmane.org>
2008-01-24 11:40 ` Robin Humble
[not found] ` <20080124114027.GC26751-Td5ZOp7sT3Xw02mFwxTg32+DJq1SqhBbsOSz5zK2v9k@public.gmane.org>
2008-01-24 12:41 ` Vladislav Bolkhovitin
[not found] ` <e2e108260801240236o2273be0bw24a2a61dcc781222-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2008-01-24 11:32 ` Robin Humble
[not found] ` <20080124113215.GB26751-Td5ZOp7sT3Xw02mFwxTg32+DJq1SqhBbsOSz5zK2v9k@public.gmane.org>
2008-01-24 12:40 ` Vladislav Bolkhovitin
2008-01-24 16:16 ` [Stgt-devel] " Bart Van Assche
2008-01-24 19:54 ` Vladislav Bolkhovitin
2008-01-25 7:24 ` Bart Van Assche
2008-01-22 12:33 ` Bart Van Assche
2008-01-17 14:22 ` Erez Zilber
[not found] ` <478F64A0.6020201-hKgKHo2Ms0F+cjeuK/JdrQ@public.gmane.org>
2008-01-17 14:32 ` Vladislav Bolkhovitin
[not found] ` <478F6708.30604-d+Crzxg7Rs0@public.gmane.org>
2008-01-17 14:46 ` Erez Zilber
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=47960A6D.6010602@vlnb.net \
--to=vst-d+crzxg7rs0@public.gmane.org \
--cc=James.Bottomley-JuX6DAaQMKPCXq6kfMZ53/egYHeGw8Jk@public.gmane.org \
--cc=bart.vanassche-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=fujita.tomonori-Zyj7fXuS5i5L9jVzuh4AOg@public.gmane.org \
--cc=linux-scsi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=scst-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org \
--cc=stgt-devel-0fE9KPoRgkgATYTw5x5z8w@public.gmane.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).