Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 09 Mar 2004 15:09:53 -0800 (PST)
From:      jhanna@shaw.ca
To:        Brad Knowles <brad.knowles@skynet.be>
Cc:        freebsd-current@freebsd.org
Subject:   Re: Who wants SACK? (Re: was My planned work on networking stack)
Message-ID:  <XFMail.20040309150953.jhanna@shaw.ca>
In-Reply-To: <20040309214205.3EE2D5D07@ptavv.es.net>

next in thread | previous in thread | raw e-mail | index | archive | help

On 09-Mar-2004 Kevin Oberman wrote:
>> Date: Tue, 9 Mar 2004 20:13:11 +0100
>> From: Brad Knowles <brad.knowles@skynet.be>
>> Sender: owner-freebsd-current@freebsd.org
>> 
>> At 3:32 PM -0800 2004/03/08, Jeffrey Hsu wrote:
>> 
>> >  What Luigi says is absolutely correct.  It doesn't take a lot to
>> >  get this done.  I've talked to a number of companies about implementing
>> >  SACK for them and while there was interest, no one wanted to fund
>> >  it all themselves, potentially for the benefit of their competitors.
>> 
>>      Out of curiosity, can someone provide some pointers as to where 
>> SACK really helps?  Is this just for high-speed WANs and doesn't help 
>> on LANs, or is it useful in both contexts?  Also, at what 
>> speeds/packet sizes does SACK start to become really useful?
>> 
>>      I'm just wondering if there aren't a lot of people who could 
>> benefit from something like this, only they don't know it.  If they 
>> were to find out, it might help provide funding and other resources 
>> to spur development.
> 
> Selective ACKnowledgment (SACK) allows acknowledgment of received
> packets in a TCP window so that only the missing/damaged packet needs to
> be re-transmitted. This is normally of little value on a LAN where ACKs
> arrive quickly and windows are smaller and no use on slow circuits. On
> fat pipes with latency and big windows it is a huge win as it allows you to
> recover much faster from a packet drop. If you don't have SACK, you need
> to re-transmit all of the packets in flight within the window while
> with SACK, you need only retransmit the dropped packet(s). If you have a
> 10 or 20 MB window, this is a big deal.
> 
> Dynamic window sizing will make it of less significance in LANs as the
> windows will not be very large.

Radio links as well, with their latency and higher frame drop rates, can benefit
considerably. Cell phones and such may account for a large amount of garden variety
traffic as time goes on.

jhanna@shaw.ca



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.20040309150953.jhanna>