From: Rick Jones <rick.jones2@hp.com>
To: Chris Snook <csnook@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>,
David Miller <davem@davemloft.net>,
netdev@vger.kernel.org
Subject: Re: RFC: Nagle latency tuning
Date: Mon, 22 Sep 2008 15:47:30 -0700 [thread overview]
Message-ID: <48D82082.7010803@hp.com> (raw)
In-Reply-To: <48D81A95.5080207@redhat.com>
> Indeed. Setting tcp_delack_min to 0 completely eliminated the undesired
> latencies, though of course that would be a bit dangerous with naive
> apps talking across the network.
What did it do to the packets per second or per unit of work? Depending
on the nature of the race between the ACK returning from the remote and
the application pushing more bytes into the socket, I'd think that
setting the delayed ack timer to zero could result in more traffic on
the network (those bare ACKs) than simply setting TCP_NODELAY at the source.
And since with small packets and/or copy avoidance an ACK is
(handwaving) just as many CPU cycles at either end as a data segment
that also means a bump in CPU utilization.
rick jones
next prev parent reply other threads:[~2008-09-22 22:47 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-08 21:56 RFC: Nagle latency tuning Christopher Snook
2008-09-08 22:39 ` Rick Jones
2008-09-09 5:10 ` Chris Snook
2008-09-09 5:17 ` David Miller
2008-09-09 5:56 ` Chris Snook
2008-09-09 6:02 ` David Miller
2008-09-09 10:31 ` Mark Brown
2008-09-09 12:05 ` David Miller
2008-09-09 12:09 ` Mark Brown
2008-09-09 12:19 ` David Miller
2008-09-09 6:22 ` Evgeniy Polyakov
2008-09-09 6:28 ` Chris Snook
2008-09-09 13:00 ` Arnaldo Carvalho de Melo
2008-09-09 14:36 ` Andi Kleen
2008-09-09 18:40 ` Chris Snook
2008-09-09 19:07 ` Andi Kleen
2008-09-09 19:21 ` Arnaldo Carvalho de Melo
2008-09-11 4:08 ` Chris Snook
2008-09-09 19:59 ` David Miller
2008-09-09 20:25 ` Chris Snook
2008-09-22 10:49 ` David Miller
2008-09-22 11:09 ` David Miller
2008-09-22 20:30 ` Andi Kleen
2008-09-22 22:22 ` Chris Snook
2008-09-22 22:26 ` David Miller
2008-09-22 23:00 ` Chris Snook
2008-09-22 23:13 ` David Miller
2008-09-22 23:24 ` Andi Kleen
2008-09-22 23:21 ` David Miller
2008-09-23 0:14 ` Andi Kleen
2008-09-23 0:33 ` Rick Jones
2008-09-23 2:12 ` Andi Kleen
2008-09-23 1:40 ` David Miller
2008-09-23 2:23 ` Andi Kleen
2008-09-23 2:28 ` David Miller
2008-09-23 2:41 ` Andi Kleen
2008-09-22 22:47 ` Rick Jones [this message]
2008-09-22 22:57 ` Chris Snook
2008-09-09 16:33 ` Rick Jones
2008-09-09 16:54 ` Chuck Lever
2008-09-09 17:21 ` Arnaldo Carvalho de Melo
2008-09-09 17:54 ` Rick Jones
2008-09-08 22:55 ` Andi Kleen
2008-09-09 5:22 ` Chris Snook
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=48D82082.7010803@hp.com \
--to=rick.jones2@hp.com \
--cc=andi@firstfloor.org \
--cc=csnook@redhat.com \
--cc=davem@davemloft.net \
--cc=netdev@vger.kernel.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).