OpenSSH
[Top] [All Lists]

Re: Disk I/O with OpenSSH

To: Damien Miller <djm@mindrot.org>
Subject: Re: Disk I/O with OpenSSH
From: chris rapier <rapier@psc.edu>
Date: Mon, 29 Jan 2007 09:51:17 -0500
Cc: OpenSSH <openssh-unix-dev@mindrot.org>
Delivered-to: sp-com-lists@consult.net
Delivered-to: openssh-unix-dev-list1@securepoint.com
Delivered-to: openssh-unix-dev-tmda@mindrot.org
Delivered-to: openssh-unix-dev@mindrot.org
In-reply-to: <Pine.BSO.4.64.0701270945230.24614@fuyu.mindrot.org>
List-archive: <http://lists.mindrot.org/pipermail/openssh-unix-dev>
List-help: <mailto:openssh-unix-dev-request@mindrot.org?subject=help>
List-id: Development of portable OpenSSH <openssh-unix-dev.mindrot.org>
List-post: <mailto:openssh-unix-dev@mindrot.org>
List-subscribe: <http://lists.mindrot.org/mailman/listinfo/openssh-unix-dev>, <mailto:openssh-unix-dev-request@mindrot.org?subject=subscribe>
List-unsubscribe: <http://lists.mindrot.org/mailman/listinfo/openssh-unix-dev>, <mailto:openssh-unix-dev-request@mindrot.org?subject=unsubscribe>
References: <45BA4F85.5030900@psc.edu> <Pine.BSO.4.64.0701270945230.24614@fuyu.mindrot.org>
Sender: openssh-unix-dev-bounces+openssh-unix-dev-list1=securepoint.com@mindrot.org
User-agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 (ax)
Damien Miller wrote:
> On Fri, 26 Jan 2007, Chris Rapier wrote:
> 
> 
>>I was wondering if the disk I/O in OpenSSH is synchronous or 
>>asynchronous. If its not asynchronous is there any reason why doing at 
>>least asynch writes would be a bad idea? I'm just wondering if I'm over 
>>looking some obvious problem.
> 
> 
> What disk IO? Do you mean sftp/scp?


Yes. Mostly I'm wondering if a momentary hiccup on I/O could cause a 
stoppage along the entire IO chain and end up causing a delayed ACK 
which might push the connection back into slow start. Even with a 
reasonably fast disk on a fat enough connection seek time can exceed RTT 
and sometimes by several multiples. This just came up at a meeting at 
work and I had no worthwhile answer for them so I'm doing a little 
research. My feeling is that HW disk buffering will prevent it from 
being a factor in 99% of the time. However, I thought I would ask.


_______________________________________________
openssh-unix-dev mailing list
openssh-unix-dev@mindrot.org
http://lists.mindrot.org/mailman/listinfo/openssh-unix-dev

<Prev in Thread] Current Thread [Next in Thread>