RE: : New ID : Diameter Compression

"STURA Marco Consultant" <[email protected]> Fri, 23 Sep 2005 09:43:07 +0200
Newsgroups gmane.ietf.aaa
Message-ID <[email protected]>
I also think that the problem the draft is trying to solve is not clear.
After all compressing/decompressing data has a cost, which presumably
doesn't make sense when there is a 1GB connection between Diameter peers
as already mentioned by Miguel. 

My opinion is also that there seems to be no requirements for this
draft, if one really wants to compress data between Diameter peers can
still use TLS compression that is presumably more effective than just
compressing some AVP.

Regards
Marco

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf
Of Miguel Garcia
Sent: Friday, September 23, 2005 9:24 AM
To: Bernard Aboba
Cc: [email protected]
Subject: Re: [AAA-WG]: New ID : Diameter Compression

I also have similar thoughts at a first glance at the I-D.

I think that the draft focuses on how to negotiate some compression 
algorithm, but it does not describe:

a) What is actually compressed? Is it the whole Diameter message, or 
just one or more AVPs within a message.

b) A clear justification of the problem. In essence, Diameter will be 
typically used in real-time environments characterized also by the 
presence of high bandwidth (this is a network-to-network protocol, so it

is not going to reach low-bandwidth interfaces such as a mobile phone). 
So I wouldn't like that Diameter peers spend a very precious time 
compressing and decompressing a few kilobytes of data when there is a 1 
GB/s Ethernet connection among them. It will be faster and cheaper to 
send the data uncompressed.

My opinion is that there is no need for this draft. At least, I cannot 
see the applicability nor the requirements.

/Miguel


Bernard Aboba wrote:

> Diameter already supports compression at the ttransmission layer via
TLS and IKE. 
> 
> Does anyone understand why this "extension" is necessary?
> 
> On Thu, 22 Sep 2005, Rajith wrote:
> 
> 
>>All
>>
>>	A draft about "Diameter Compression" is available now.
>>
>>	Please review the same and give your comments.
>>
>>Warmest Regards
>>
>>Rajith
>>
>>______________________________________________________________________
______
>>______________________
>>
>>A New Internet-Draft is available from the on-line Internet-Drafts
>>directories.
>>
>>
>>		 Title		 		 : Diameter Compression
>>		 Author(s)		 : H. Zhangtao, R. Rajith
>>		 Filename		 :
draft-zhangtao-aaa-compression-00.txt
>>		 Pages		 		 : 11
>>		 Date		 		 : 2005-9-14
>>
>>  This document describes the negotiation of compression algorithm
>>  between two Diameter peers and the indication of data compression
>>  using a new AVP [Attribute Value Pair] flag.
>>
>>A URL for this Internet-Draft is:
>>http://www.ietf.org/internet-drafts/draft-zhangtao-aaa-compression-00.
txt
>>
>>
> 
> 

-- 
Miguel A. Garcia           tel:+358-50-4804586
sip:[email protected]
Nokia Research Center      Helsinki, Finland