From: Jonathan Morton <chromatix99@gmail.com>
To: Dave Taht <dave.taht@gmail.com>
Cc: Dan Siemon <dan@coverfire.com>, Cake List <cake@lists.bufferbloat.net>
Subject: Re: [Cake] Using cake to shape 1000’s of users.
Date: Sat, 28 Jul 2018 19:11:05 +0300 [thread overview]
Message-ID: <33C09185-34D2-4D96-9DE3-345D51D4D5C9@gmail.com> (raw)
In-Reply-To: <CAA93jw7JcYyAVgqjdiD1xEvKvO4fY=Wh-ON28pfgKb8Kt4KzHQ@mail.gmail.com>
> On 28 Jul, 2018, at 6:51 pm, Dave Taht <dave.taht@gmail.com> wrote:
>
> That's also pretty low end. On the high end nowadays there's stuff like this:
>
> https://www.amazon.com/Intel-Xeon-E5-2698-Hexadeca-core-Processor/dp/B00PDD1QES
Intel is no longer high-end for x86 CPUs. Not all of the market has realised that yet, but it's true.
Look at Threadripper 2 which scales up to 32 cores, 64 threads in a single socket at HEDT prices, and EPYC which just goes bonkers in terms of I/O capabilities and still costs less than its nearest Intel competitor. None of which has any serious concerns with the recent series of Meltdown/Spectre speculation bugs, unlike Intel.
AMD is moving to a 7nm silicon process which apparently works pretty well already, and is theoretically on par with Intel's 10nm process which they still haven't got working reliably after how many years of delays now? And they're already beating Intel over the head with a 14nm process which is theoretically *inferior* to Intel's 14nm process, which they'll be stuck with in practice for *at least* the next year even by their own wildly optimistic latest estimates.
The only place Intel temporarily holds a real advantage is in maximum single-threaded turbo clock speed. This is relevant to a shrinking minority of users these days. AMD's next CPUs are supposed to make that wholly irrelevant with a significant further jump in IPC - because they were designed to compete with 10nm Intel CPUs that Intel now looks very unlikely to be capable of manufacturing.
And is this relevant to "Super Mega Turbo Cake Edition XLRi"? Well, one of the nice things about having lots of users is that you can statistically multiplex them across multiple hardware queues more easily. Each subscriber's traffic can sanely end up on the same queue each time, and each queue can have a separate "Super Cake" instance allocated an even division of the total backhaul bandwidth, and in theory each of *those* can run on its own CPU core. Instant throughput boost.
- Jonathan Morton
next prev parent reply other threads:[~2018-07-28 16:11 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-17 7:24 Felix Resch
2018-07-17 16:59 ` Dave Taht
2018-07-26 15:46 ` Dan Siemon
2018-07-26 15:48 ` Dave Taht
2018-07-26 18:07 ` Dan Siemon
2018-07-28 15:51 ` Dave Taht
2018-07-28 16:11 ` Jonathan Morton [this message]
2018-07-28 16:36 ` Dave Taht
2018-07-26 17:42 ` Toke Høiland-Jørgensen
2018-07-26 18:10 ` Dan Siemon
2018-07-26 21:09 ` Toke Høiland-Jørgensen
2018-07-26 21:38 ` Jonathan Morton
2018-07-27 9:25 ` Pete Heist
2018-07-27 14:04 ` Dan Siemon
2018-07-27 18:58 ` Jonathan Morton
2018-07-28 8:56 ` Toke Høiland-Jørgensen
2018-07-28 15:04 ` Dave Taht
2018-07-28 16:19 ` Jonathan Morton
2018-07-28 16:39 ` Dave Taht
2018-07-28 17:01 ` Pete Heist
2018-07-28 17:37 ` Pete Heist
2018-07-28 17:52 ` Dave Taht
2018-07-28 17:56 ` Dave Taht
2018-07-28 18:12 ` Toke Høiland-Jørgensen
2018-07-29 0:17 ` Pete Heist
2018-07-29 19:14 ` Toke Høiland-Jørgensen
2018-07-30 9:14 ` Pete Heist
2018-07-30 10:09 ` Sebastian Moeller
2018-07-30 10:55 ` Toke Høiland-Jørgensen
2018-07-30 11:05 ` Pete Heist
2018-07-30 11:28 ` Toke Høiland-Jørgensen
2018-07-30 22:10 ` Pete Heist
2018-07-30 22:17 ` Toke Høiland-Jørgensen
2018-07-31 7:31 ` Jonathan Morton
2018-07-30 10:55 ` Pete Heist
2018-07-30 11:05 ` Jonathan Morton
2018-07-28 17:53 ` Jonathan Morton
2018-07-28 18:07 ` Dave Taht
2018-07-28 18:17 ` Toke Høiland-Jørgensen
2018-07-28 19:35 ` [Cake] 1000s " Dave Taht
2018-07-29 23:24 ` [Cake] Using cake to shape 1000’s " Dave Taht
2018-08-07 1:46 ` Dan Siemon
2018-07-28 7:18 ` Pete Heist
2018-07-28 8:06 ` Jonathan Morton
2018-07-28 16:41 ` Pete Heist
2018-07-28 17:32 ` [Cake] isp economics Dave Taht
2018-07-28 18:39 ` Pete Heist
2018-07-28 19:03 ` Dave Taht
2018-07-28 20:00 ` Pete Heist
2018-07-29 5:49 ` Loganaden Velvindron
2018-07-28 19:09 ` Dave Taht
-- strict thread matches above, loose matches on Subject: below --
2018-07-16 18:39 [Cake] Using cake to shape 1000’s of users Mike
2018-07-16 19:01 ` Jonathan Morton
2018-07-16 19:13 ` Michel Blais
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/cake.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=33C09185-34D2-4D96-9DE3-345D51D4D5C9@gmail.com \
--to=chromatix99@gmail.com \
--cc=cake@lists.bufferbloat.net \
--cc=dan@coverfire.com \
--cc=dave.taht@gmail.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