From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!news.eternal-september.org!mx05.eternal-september.org!feeder.eternal-september.org!aioe.org!.POSTED!not-for-mail
From: "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>
Newsgroups: comp.lang.ada
Subject: Re: Queues just aren't scalable
Date: Wed, 26 Jun 2013 09:12:31 +0200
Organization: cbb software GmbH
Message-ID: <1bzu3qjknf5b.1dj7y7733p6uq$.dlg@40tude.net>
References: <kqcaii$s7l$5@dont-email.me>
 <30aafae8-8475-4beb-8ca0-e065488001c8@googlegroups.com>
Reply-To: mailbox@dmitry-kazakov.de
NNTP-Posting-Host: IenaDxMXK2hi7fvYcb+MlQ.user.speranza.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse@aioe.org
User-Agent: 40tude_Dialog/2.0.15.1
X-Notice: Filtered by postfilter v. 0.8.2
Xref: news.eternal-september.org comp.lang.ada:15938
Date: 2013-06-26T09:12:31+02:00
List-Id: <comp.lang.ada>

On Tue, 25 Jun 2013 15:34:28 -0700 (PDT), Shark8 wrote:

> On Tuesday, June 25, 2013 8:53:03 AM UTC-6, aminer wrote:
>> 
>> Queues just aren't scalable
> 
>> Read this:
>> http://concurrencyfreaks.blogspot.ca/2013_05_01_archive.html
> 
> ...isn't that obvious?

No, it is not. There is no obvious reason why the *same* number requests
submitted by one thread should be processed substantially slower when
submitted by several threads.

Let us ignore for a while that peer-to-peer (1-to-1) queues can be
implemented differently (e.g. without locking) than n-to-1 queues.

> A Queue, like a Stack, has implicit sequentiality -- anyone who's had to
> wait for someone else to get their plate from an all-you-can-eat buffet
> plate-stack has experienced this... anyone who has waited in line at the
> DMV has experienced this.

BTW, scalability is not what seems to be implied here. A system is scalable
in the parameter X and the outcome L if L depends on X more or less linear:

   L = a X + b

E.g.

X = number of threads
L = latency

Scalability does NOT mean that a = 0!

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de