The Unix Heritage Society mailing list
 help / color / mirror / Atom feed
From: "Theodore Y. Ts'o" <tytso@mit.edu>
To: "Perry E. Metzger" <perry@piermont.com>
Cc: TUHS main list <tuhs@minnie.tuhs.org>
Subject: Re: [TUHS] PDP-11 legacy, C, and modern architectures
Date: Thu, 28 Jun 2018 18:29:54 -0400	[thread overview]
Message-ID: <20180628222954.GD8521@thunk.org> (raw)
In-Reply-To: <20180628170317.14d65067@jabberwock.cb.piermont.com>

On Thu, Jun 28, 2018 at 05:03:17PM -0400, Perry E. Metzger wrote:
> 
> Tens of thousands of machines is a lot more than one. I think the
> point stands. This is the age of distributed and parallel systems.

This is the age of distributed systems, yes.  I'm not so sure about
"parallel".  And the point remains that for many problems, you need
fewer strong cores, and a crapton of weak cores is not as useful.

Of course we should parllelize work where we can.  The point is that
very often, we can't.  And if you are really worried about potential
problems with Spectre and Meltdown, what that means is that sharing
caches is perilous.  So if you have 128 wimpy cores, you need 128
separate I and D cacaches.  If you have 32 stronger cores, you need 32
separate I and D caches.

And the fact remains that humans really suck at parallel programming.
Use a separate core for each HTTP request, with a load balancer to
split the incoming request to tens of hundreds servers?  Sure!  But
using a several dozen cores for each HTTP request?  That's a much
bigger lift.

You're conflating "distributed" and "parllel" computing, and they are
really quite different.

      	    	     	  	    	   - Ted

  reply	other threads:[~2018-06-28 22:30 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-26 17:54 Nelson H. F. Beebe
2018-06-26 18:03 ` Cornelius Keck
2018-06-26 21:21   ` Nelson H. F. Beebe
2018-06-26 21:56   ` Kurt H Maier
2018-06-26 18:52 ` Ronald Natalie
2018-06-26 19:01 ` Ronald Natalie
2018-06-26 21:16   ` Arthur Krewat
2018-06-26 21:50     ` Larry McVoy
2018-06-26 21:54       ` Ronald Natalie
2018-06-26 21:59         ` Larry McVoy
2018-06-26 22:20           ` Bakul Shah
2018-06-26 22:33             ` Arthur Krewat
2018-06-26 23:53               ` Bakul Shah
2018-06-27  8:30             ` Tim Bradshaw
2018-06-26 22:33           ` Andy Kosela
2018-06-27  0:11             ` Bakul Shah
2018-06-27  6:10               ` arnold
2018-06-27  2:18           ` [TUHS] PDP-11 legacy, C, and modern architectTures Theodore Y. Ts'o
2018-06-27  2:22             ` Theodore Y. Ts'o
2018-06-28 14:36             ` Steffen Nurpmeso
2018-06-27 11:26         ` [TUHS] PDP-11 legacy, C, and modern architectures Tony Finch
2018-06-27 14:33           ` Clem Cole
2018-06-27 14:38             ` Clem Cole
2018-06-27 15:30             ` Paul Winalski
2018-06-27 16:55               ` Tim Bradshaw
2018-06-27  6:27     ` arnold
2018-06-27 16:00 ` Steve Johnson
2018-06-28  4:12   ` Bakul Shah
2018-06-28 14:15     ` Theodore Y. Ts'o
2018-06-28 14:40       ` Larry McVoy
2018-06-28 14:55         ` Perry E. Metzger
2018-06-28 14:58           ` Larry McVoy
2018-06-28 15:39             ` Tim Bradshaw
2018-06-28 16:02               ` Larry McVoy
2018-06-28 16:41                 ` Tim Bradshaw
2018-06-28 16:59                   ` Paul Winalski
2018-06-28 17:09                   ` Larry McVoy
2018-06-29 15:32                     ` tfb
2018-06-29 16:09                       ` Perry E. Metzger
2018-06-29 17:51                       ` Larry McVoy
2018-06-29 18:27                         ` Tim Bradshaw
2018-06-29 19:02                         ` Perry E. Metzger
2018-06-28 20:37                 ` Perry E. Metzger
2018-06-28 15:37         ` Clem Cole
2018-06-28 20:37           ` Lawrence Stewart
2018-06-28 14:43       ` Perry E. Metzger
2018-06-28 14:56         ` Larry McVoy
2018-06-28 15:07           ` Warner Losh
2018-06-28 19:42           ` Perry E. Metzger
2018-06-28 19:55             ` Paul Winalski
2018-06-28 20:42             ` Warner Losh
2018-06-28 21:03               ` Perry E. Metzger
2018-06-28 22:29                 ` Theodore Y. Ts'o [this message]
2018-06-29  0:18                   ` Larry McVoy
2018-06-29 15:41                     ` Perry E. Metzger
2018-06-29 18:01                       ` Larry McVoy
2018-06-29 19:07                         ` Perry E. Metzger
2018-06-29  5:58                   ` Michael Kjörling
2018-06-28 20:52             ` Lawrence Stewart
2018-06-28 21:07               ` Perry E. Metzger
2018-06-28 16:45       ` Paul Winalski
2018-06-28 20:47         ` Perry E. Metzger
2018-06-29 15:43         ` emanuel stiebler
2018-06-29  2:02       ` Bakul Shah
2018-06-29 12:58         ` Theodore Y. Ts'o
2018-06-29 18:41           ` Perry E. Metzger
2018-06-29  1:02 Noel Chiappa
2018-06-29  1:06 Noel Chiappa

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=20180628222954.GD8521@thunk.org \
    --to=tytso@mit.edu \
    --cc=perry@piermont.com \
    --cc=tuhs@minnie.tuhs.org \
    /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).