<< | Thread Index | >> ]    [ << | Date Index | >> ]

Subject: Re: Proposal: Compression of large packets: fragmentation
From: Josef Drexler <jdrexler,AT,josefsbox,DOT,cjb,DOT,net>
Date: Sun, 16 Dec 2001 23:41:43 +0100
In-reply-to: <Pine.LNX.4.21.0112161158590.10380-100000@raid.kaico.com>

On Sun, 16 Dec 2001 ewheeler,AT,kaico,DOT,com wrote:
> When we are talking about time to compress, we also need to keep in mind
> latency.  If there is a slow compression algorithm, it could impact the
> transfer delay;  I don't know if this will be in us or ms.  I wonder how
> difficult it would be to implement several compression algorithms (bzip,
> gzip, lzo).

Well latency will be mainly important for smaller packets (syn, ack,
telnet characters, those kinds of things).  For larger packets it's mostly
bandwidth that's important, not latency, and so by only compressing larger
packets it wouldn't make a difference for latency but would improve
bandwidth by up to 100%.

Allan Latham's version of Cipe with compression uses an algorithm to
compress the tunnelled IP headers, which makes small packets tiny, and
makes serious impact to improve latency and responsiveness of telnet or
ssh sessions.

   Josef Drexler                 |    http://publish.uwo.ca/~jdrexler/
 Please help Conserve Gravity    | Email address is *valid*.
 Don't do push ups               | Don't remove the "nospam" part.

<< | Thread Index | >> ]    [ << | Date Index | >> ]