public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed
From: Rick Jones <rick.jones2@hp•com>
To: Shawn Bohrer <sbohrer@rgmadvisors•com>
Cc: David Miller <davem@davemloft•net>,
	Eric Dumazet <eric.dumazet@gmail•com>,
	tomk@rgmadvisors•com, netdev <netdev@vger•kernel.org>
Subject: Re: [net-next 2/3] udp: Add udp early demux
Date: Tue, 01 Oct 2013 13:12:18 -0700	[thread overview]
Message-ID: <524B2CA2.4090807@hp.com> (raw)
In-Reply-To: <1380656025-8847-3-git-send-email-sbohrer@rgmadvisors.com>

On 10/01/2013 12:33 PM, Shawn Bohrer wrote:
> The removal of the routing cache introduced a performance regression for
> some UDP workloads since a dst lookup must be done for each packet.
> This change caches the dst per socket in a similar manner to what we do
> for TCP by implementing early_demux.
>
> For UDP multicast we can only cache the dst if there is only one
> receiving socket on the host.  Since caching only works when there is
> one receiving socket we do the multicast socket lookup using RCU.
>
> Benchmark results from a netperf UDP_RR test:
> Before 90596.44 transactions/s
> After  91296.97 transactions/s

Were those measured with confidence intervals enabled?  It would be a 
Good Idea (tm) to either use that - I would suggest -I 99,1 -i 30,3 
added to the global portion of the netperf command line - or take 
several runs.  (If you've not already done so since those look more like 
"raw" netperf numbers rather than theaverage of several runs)

happy benchmarking,

rick jones

> Benchmark results from a fio 1 byte UDP multicast pingpong test
> (Multicast one way unicast response):
> Before 12.647us RTT
> After  12.497us RTT

  reply	other threads:[~2013-10-01 20:12 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-01 19:33 [net-next 0/3] Improve UDP multicast receive latency Shawn Bohrer
2013-10-01 19:33 ` [net-next 1/3] udp: Only allow busy read/poll on connected sockets Shawn Bohrer
2013-10-01 20:44   ` Eric Dumazet
2013-10-01 19:33 ` [net-next 2/3] udp: Add udp early demux Shawn Bohrer
2013-10-01 20:12   ` Rick Jones [this message]
2013-10-01 22:26     ` Shawn Bohrer
2013-10-01 20:52   ` Eric Dumazet
2013-10-02 17:34     ` Shawn Bohrer
2013-10-02 18:09       ` Eric Dumazet
2013-10-02 20:35         ` Shawn Bohrer
2013-10-02 21:08           ` Eric Dumazet
2013-10-02 21:24             ` Shawn Bohrer
2013-10-02 21:38               ` Eric Dumazet
2013-10-03 17:39                 ` Shawn Bohrer
2013-10-03 18:06                   ` Eric Dumazet
2013-10-01 19:33 ` [net-next 3/3] net: ipv4 only populate IP_PKTINFO when needed Shawn Bohrer
2013-10-01 20:42   ` Eric Dumazet
2013-10-01 22:29     ` Shawn Bohrer
2013-10-01 20:21 ` [net-next 0/3] Improve UDP multicast receive latency Veaceslav Falico

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=524B2CA2.4090807@hp.com \
    --to=rick.jones2@hp$(echo .)com \
    --cc=davem@davemloft$(echo .)net \
    --cc=eric.dumazet@gmail$(echo .)com \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=sbohrer@rgmadvisors$(echo .)com \
    --cc=tomk@rgmadvisors$(echo .)com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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