From: Eric Dumazet <eric.dumazet@gmail.com>
To: David Laight <David.Laight@ACULAB.COM>,
'Paolo Abeni' <pabeni@redhat.com>,
Jesper Dangaard Brouer <brouer@redhat.com>
Cc: 'Marek Majkowski' <marek@cloudflare.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
network dev <netdev@vger.kernel.org>,
kernel-team <kernel-team@cloudflare.com>
Subject: Re: epoll_wait() performance
Date: Fri, 29 Nov 2019 17:07:20 -0800 [thread overview]
Message-ID: <c46e43d1-ba7d-39d9-688f-0141931df1b0@gmail.com> (raw)
In-Reply-To: <1265e30d04484d08b86ba2abef5f5822@AcuMS.aculab.com>
On 11/28/19 2:17 AM, David Laight wrote:
> From: Eric Dumazet
>> Sent: 27 November 2019 17:47
> ...
>> A QUIC server handles hundred of thousands of ' UDP flows' all using only one UDP socket
>> per cpu.
>>
>> This is really the only way to scale, and does not need kernel changes to efficiently
>> organize millions of UDP sockets (huge memory footprint even if we get right how
>> we manage them)
>>
>> Given that UDP has no state, there is really no point trying to have one UDP
>> socket per flow, and having to deal with epoll()/poll() overhead.
>
> How can you do that when all the UDP flows have different destination port numbers?
> These are message flows not idempotent requests.
> I don't really want to collect the packets before they've been processed by IP.
>
> I could write a driver that uses kernel udp sockets to generate a single message queue
> than can be efficiently processed from userspace - but it is a faff compiling it for
> the systems kernel version.
Well if destinations ports are not under your control,
you also could use AF_PACKET sockets, no need for 'UDP sockets' to receive UDP traffic,
especially it the rate is small.
next prev parent reply other threads:[~2019-11-30 1:07 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-22 11:17 epoll_wait() performance David Laight
2019-11-27 9:50 ` Marek Majkowski
2019-11-27 10:39 ` David Laight
2019-11-27 15:48 ` Jesper Dangaard Brouer
2019-11-27 16:04 ` David Laight
2019-11-27 19:48 ` Willem de Bruijn
2019-11-28 16:25 ` David Laight
2019-11-28 11:12 ` Jesper Dangaard Brouer
2019-11-28 16:37 ` David Laight
2019-11-28 16:52 ` Willy Tarreau
2019-12-19 7:57 ` Jesper Dangaard Brouer
2019-11-27 16:26 ` Paolo Abeni
2019-11-27 17:30 ` David Laight
2019-11-27 17:46 ` Eric Dumazet
2019-11-28 10:17 ` David Laight
2019-11-30 1:07 ` Eric Dumazet [this message]
2019-11-30 13:29 ` Jakub Sitnicki
2019-12-02 12:24 ` David Laight
2019-12-02 16:47 ` Willem de Bruijn
2019-11-27 17:50 ` Paolo Abeni
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=c46e43d1-ba7d-39d9-688f-0141931df1b0@gmail.com \
--to=eric.dumazet@gmail.com \
--cc=David.Laight@ACULAB.COM \
--cc=brouer@redhat.com \
--cc=kernel-team@cloudflare.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marek@cloudflare.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
/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).