From: "Richard Croucher" <richard-jNDFPZUTrfRJpuwtbJ71GdBPR1lH4CV8@public.gmane.org>
To: 'Tanin' <fanqielee-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
'linux-rdma' <linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
ewg-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org
Subject: Re: iWARP tuning
Date: Fri, 4 Mar 2011 08:59:45 -0000 [thread overview]
Message-ID: <005701cbda4a$82f9b8b0$88ed2a10$@com> (raw)
In-Reply-To: <AANLkTimhJv2Mdj3disORwbyudRZWF=DLvG8PCtg577a+-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
[-- Attachment #1.1: Type: text/plain, Size: 1439 bytes --]
Is your NIC using TOE or a kernel based TCP/IP stack.
TCP requires the window size to match the B/W delay product. By default, it
is only set to support high B/W LAN or low B/W WAN. If you have a high B/W
WAN, which you would need for RDMA, then you need to increase the TCP
buffers and window size settings accordingly. If it is TOE, then you will
be constrained by whether they expose these settings and whether there is
sufficient memory on the card to reach your B/W delay product.
From: ewg-bounces-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org
[mailto:ewg-bounces-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org] On Behalf Of Tanin
Sent: 04 March 2011 01:21
To: linux-rdma; ewg-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org
Subject: [ewg] iWARP tuning
Dear all,
We had conducted some tests on long distance data transfer using
iWARP. But the performance is bad. Does anyone has some idea on how to tune
the iWARP for better performance? Does iWARP support paralell data transfer?
Does the buffer size of the card tunable?
Thanks!
--
Best regards,
----------------------------------------------------------------------------
-------------------
Li, Tan
PhD Candidate & Research Assistant,
Electrical Engineering,
Stony Brook University, NY
Personal Web Site: https://sites.google.com/site/homepagelitan/Home
Email: fanqielee-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
[-- Attachment #1.2: Type: text/html, Size: 4734 bytes --]
[-- Attachment #2: Type: text/plain, Size: 176 bytes --]
_______________________________________________
ewg mailing list
ewg-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
prev parent reply other threads:[~2011-03-04 8:59 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-04 1:21 iWARP tuning Tanin
[not found] ` <AANLkTimhJv2Mdj3disORwbyudRZWF=DLvG8PCtg577a+-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2011-03-04 8:59 ` Richard Croucher [this message]
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='005701cbda4a$82f9b8b0$88ed2a10$@com' \
--to=richard-jndfpzutrfrjpuwtbj71gdbpr1lh4cv8@public.gmane.org \
--cc=ewg-ZwoEplunGu1OwGhvXhtEPSCwEArCW2h5@public.gmane.org \
--cc=fanqielee-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@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