General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: "Toke Høiland-Jørgensen" <toke@toke.dk>
To: Dave Taht <dave.taht@gmail.com>
Cc: Mikael Abrahamsson <swmike@swm.pp.se>,
	bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] [Cerowrt-devel] beating the drum for BQL
Date: Fri, 24 Aug 2018 19:58:42 +0200	[thread overview]
Message-ID: <877ekfbq99.fsf@toke.dk> (raw)
In-Reply-To: <CAA93jw7jMfT-jnWWN_yrdbg+n59oaG78ctvT-CD6dPt76c7FZQ@mail.gmail.com>

Dave Taht <dave.taht@gmail.com> writes:

> On Thu, Aug 23, 2018 at 2:01 PM Toke Høiland-Jørgensen <toke@toke.dk> wrote:
>>
>> Dave Taht <dave.taht@gmail.com> writes:
>>
>> > One of the things not readily evident in trying to scale up, is the
>> > cost of even the most basic routing table lookup. A lot of good work
>> > in this area landed in linux 4.1 and 4.2 (see a couple posts here:
>> > https://vincent.bernat.im/en/blog/2017-performance-progression-ipv4-route-lookup-linux
>> > )
>> >
>> > Lookup time for even the smallest number of routes is absolutely
>> > miserable for IPv6 -
>> > https://vincent.bernat.im/en/blog/2017-ipv6-route-lookup-linux
>>
>> The IPv6 routing lookup is on par with v4 these days. We got 7.2M pkts/s
>> in our XDP tests on a single core (although admittedly a fairly high-end
>> Intel one). Which allows you to route 10Gbps of 64-byte packets on two
>> cores...
>
> Call me cynical, call me grumpy...
>
> but did you get that result with testing 1,10,100,1000, 10,000, 100k,
> 1M routes? The best
> case performance on that test looked like .150us, the worst case
> 1.75us

Think those were single-route tests, or close to it. Don't have results
handy for large routing tables for IPv6, but IPv4 performance drops by
~33% when going from a single route to a full BGP table dump...

Can run some tests for v6 once by testbed is running again...

-Toke

  reply	other threads:[~2018-08-24 17:58 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-23  0:49 [Bloat] " Dave Taht
2018-08-23  5:47 ` Mikael Abrahamsson
2018-08-23 13:06   ` [Bloat] [Cerowrt-devel] " Sebastian Moeller
2018-08-23 13:57     ` Jan Ceuleers
2018-08-23 14:22       ` Sebastian Moeller
2018-08-23 15:32         ` Mikael Abrahamsson
2018-08-23 17:14           ` Rosen Penev
2018-08-23 17:50             ` Mikael Abrahamsson
2018-08-23 18:21               ` Dave Taht
2018-08-23 20:15                 ` Dave Taht
2018-08-24  7:00                   ` Mikael Abrahamsson
2018-08-24  8:06                     ` Dave Taht
2018-08-24 11:22                       ` Toke Høiland-Jørgensen
2018-08-24 11:46                       ` Jan Ceuleers
2018-08-24 13:43                         ` Jan Ceuleers
2018-08-24 13:44                           ` Mikael Abrahamsson
2018-08-24 13:56                             ` Jan Ceuleers
2018-08-24 13:36                       ` Mikael Abrahamsson
2018-08-24  7:05                   ` Mikael Abrahamsson
2018-08-24 11:24                     ` Toke Høiland-Jørgensen
2018-08-23 21:01                 ` Toke Høiland-Jørgensen
2018-08-24 14:44                   ` Dave Taht
2018-08-24 17:58                     ` Toke Høiland-Jørgensen [this message]
2018-08-23 23:35           ` Sebastian Moeller
2018-08-24  1:08             ` David Collier-Brown
2018-08-24  6:16             ` Jan Ceuleers
2018-08-24 11:27           ` Toke Høiland-Jørgensen
2018-08-24 12:46             ` Pedro Tumusok
2018-08-24 13:09             ` Michael Richardson
2018-08-24 13:37             ` Mikael Abrahamsson
2018-08-24 13:44               ` Toke Høiland-Jørgensen
2018-08-26  4:29           ` [Bloat] Marvell 385 Dave Taht
2018-08-26  4:32             ` Jonathan Morton
2018-08-26 16:28             ` Rosen Penev
2018-08-26 18:44             ` Mikael Abrahamsson
2018-08-26 20:53               ` Dave Taht
2018-08-26 20:58               ` Rosen Penev
2018-08-26 21:08                 ` Dave Taht
2018-08-27  0:55                   ` Rosen Penev
2018-08-27  1:21                     ` Dave Taht
2018-08-27  1:33                       ` Rosen Penev
2018-08-27  1:45                         ` Dave Taht
2018-08-27  1:53                           ` Rosen Penev
2018-08-27 11:33                   ` Mikael Abrahamsson
2018-08-23  8:26 ` [Bloat] beating the drum for BQL Pete Heist
2018-08-23 10:51   ` Mikael Abrahamsson
2018-08-23 11:38     ` Pete Heist
2018-08-24 17:13   ` Pete Heist
2018-08-24 17:30     ` Dave Taht
2018-08-24 18:43       ` Dave Taht

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

  List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=877ekfbq99.fsf@toke.dk \
    --to=toke@toke.dk \
    --cc=bloat@lists.bufferbloat.net \
    --cc=dave.taht@gmail.com \
    --cc=swmike@swm.pp.se \
    /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