From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@9fans.net
Subject: Re: [9fans] running plan9 : an ideal setup?
Date: Sat, 22 Nov 2014 13:06:15 -0500 [thread overview]
Message-ID: <4a5f7359099c5bdfc05c341e19f17490@lilly.quanstro.net> (raw)
In-Reply-To: <0841E3F3-400C-4E19-811F-37B0595006BD@bitblocks.com>
On Fri Nov 21 12:31:13 EST 2014, bakul@bitblocks.com wrote:
> This paper is well worth reading:
> http://groups.csail.mit.edu/ana/Publications/PubPDFs/1988Analysis%20TCP%20Processing%20Overhead.pdf
>
> While the traditional BSD implementation uses mbufs that complicate things, actual tcp processing can be done quite cheaply.
- ignores tcp checksum -- it alone will take a couple cycles per byte.
- ignores locking
and i believe they've used the BKL to avoid locking any data structures, otherwise
how could they process ip in 61 instructions? actually there's proof of this in the
timer instruction count -- 17. that's not enough to acquire a lock.
- asserts that 300 instructions of x86 code -> 400 instructions of risc code, conservatively
absolutely not if one of them is rep; movb (which they appear to use)
in short, i see signs that this paper is not realistic.
furthermore, this sunny picture assumes an environment where tcp isn't giving any
benefit. if you're not retransmitting a bit, you're not getting anything out of tcp.
i think this was the original point.
- erik
next prev parent reply other threads:[~2014-11-22 18:06 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-18 13:29 Mayuresh Kathe
2014-11-18 13:53 ` dante
2014-11-18 14:11 ` Richard Miller
2014-11-18 14:28 ` dante
2014-11-27 20:57 ` Dante
2014-11-28 6:10 ` erik quanstrom
2014-11-28 6:54 ` David du Colombier
2014-11-28 8:42 ` Dante
2014-11-28 9:12 ` Mats Olsson
2014-11-28 9:18 ` Dante
2014-11-28 9:17 ` Richard Miller
2014-11-28 9:26 ` Dante
2014-11-28 13:20 ` erik quanstrom
2014-11-28 13:45 ` David du Colombier
2014-11-18 15:42 ` Kurt H Maier
2014-11-18 16:14 ` dante
2014-11-18 17:02 ` Aram Hăvărneanu
2014-11-18 20:29 ` Richard Miller
2014-11-18 21:28 ` Mats Olsson
2014-11-18 22:09 ` dante
2014-11-19 8:56 ` Mats Olsson
2014-11-26 17:16 ` Mats Olsson
2014-11-26 17:41 ` Dante
2014-11-26 17:56 ` Mats Olsson
2014-11-26 18:16 ` Mats Olsson
2014-11-26 18:41 ` Dante
2014-11-18 22:11 ` Skip Tavakkolian
2014-11-18 22:23 ` Steve Simon
2014-11-19 1:57 ` erik quanstrom
2014-11-19 5:36 ` Skip Tavakkolian
2014-11-19 5:59 ` lucio
2014-11-19 14:36 ` erik quanstrom
2014-11-19 15:34 ` Aram Hăvărneanu
2014-11-20 6:02 ` Anthony Sorace
2014-11-20 14:37 ` erik quanstrom
2014-11-20 18:43 ` Anthony Sorace
2014-11-21 14:34 ` erik quanstrom
2014-11-21 14:44 ` Anthony Sorace
2014-11-21 17:31 ` Bakul Shah
2014-11-22 18:06 ` erik quanstrom [this message]
2014-11-25 6:59 ` Bakul Shah
2014-11-25 11:10 ` erik quanstrom
2014-11-25 11:14 ` erik quanstrom
2014-11-25 13:52 ` Anthony Sorace
2014-11-25 14:26 ` erik quanstrom
2014-11-19 14:33 ` erik quanstrom
2014-11-19 20:05 ` Bakul Shah
2014-11-19 20:40 ` Tom Ivar Helbekkmo
2014-11-21 6:34 ` Harri Haataja
2014-11-19 2:04 ` erik quanstrom
2014-11-19 3:50 ` lucio
2014-11-19 9:40 ` Steve Simon
2014-11-19 9:50 ` dante
2014-11-19 10:18 ` Steve Simon
2014-11-19 10:27 ` dante
2014-11-19 18:55 ` Quintile
2014-11-19 10:36 ` lucio
2014-11-20 6:05 ` Anthony Sorace
2014-11-20 6:13 ` lucio
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=4a5f7359099c5bdfc05c341e19f17490@lilly.quanstro.net \
--to=quanstro@quanstro.net \
--cc=9fans@9fans.net \
/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).