Date: Thu, 22 Feb 2007 05:41:27 -0800 (PST)
Message-Id: <20070222.054127.104035694.davem@davemloft.net>
To: johnpol@2ka.mipt.ru
Cc: arjan@infradead.org, mingo@elte.hu, drepper@redhat.com,
       linux-kernel@vger.kernel.org, torvalds@linux-foundation.org,
       hch@infradead.org, akpm@zip.com.au, alan@lxorguk.ukuu.org.uk,
       zach.brown@oracle.com, suparna@in.ibm.com, davidel@xmailserver.org,
       jens.axboe@oracle.com, tglx@linutronix.de
Subject: Re: [patch 00/13] Syslets, "Threadlets", generic AIO support, v3
From: David Miller <davem@davemloft.net>
In-Reply-To: <20070222123929.GA5208@2ka.mipt.ru>
References: <20070222113148.GA3781@2ka.mipt.ru>
	<1172145159.3531.253.camel@laptopd505.fenrus.org>
	<20070222123929.GA5208@2ka.mipt.ru>
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 968
Lines: 23

From: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Date: Thu, 22 Feb 2007 15:39:30 +0300

> It does not matter - even with threads cost of having thousands of
> threads is _too_ expensive. So, IMO, it is wrong to have to create 
> 20k threads for the simple web server which only sends one index page to
> 80k connections with 4k connections per seconds rate.
> 
> Just have that example in mind - more than 20k blocks in 80k connections 
> over gigabit lan, and it is likely optimistic result, when designing new 
> type of AIO.

I totally agree with Evgeniy on these points.

Using things like syslets and threadlets for networking I/O
is not a very good idea.  Blocking is more the norm than the
exception for networking I/O.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/