The IP over ATM Mailing List Archive by date

Cell Relay Retreat>List Archive>month:1995-Oct> msg00085



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Header Compression.

  • From: Mikael Latvala <Mikael.Latvala@lmf.ericsson.se>
  • Date: Mon, 23 Oct 1995 15:29:35 +0100

Grenville Armitage's and Keith Adams' article appearing
in IEEE Network, January/February 1995 talks about this. It also
talks about the TCP/IP compression specified in RFC1144.

I have also read some other studies which seem to agree on those
traffic statistics.

If you just use common sense it's very easy to convince yourself that
70% of TCP/IP's payload contain < 10 octets. All the ACKs plus 
packets send by applications like telnet or rlogin are < 10 octets.

However, new multimedia applications like WWW-browsers can
alter the current traffic statistics in a major way. 

Mikael Latvala 

>     I would be interested in reading the studies you mentioned.  Any 
>     pointers would be appreciated.  Local studies here indicate 150-200 to 
>     be the norm but as Curtis said, acks are always around 40 depending on 
>     fields used in the header.  I'm not sure that any amount of 
>     compression would ensure ack's were less than 53 bytes but considering 
>     the volume of acks, 50% would chop the total number of packets 
>     considerably.
>     
>     Boz
>
>

>>     
>>     
>>I was reading some studies which indicate up to 70 % of TCP/IP 
>>packets are < 10 bytes.  For ATM that means 2 cells per packet.
>>If we were using header compression, then these would only need one cell. 
>>I have seen some papers mentioning this.  I also saw studies indicating 
>>that the 64KB window size will be too small.
>>     
>>So I had the following questions?
>>     
>>Can RFC1144 be used over ATM?  If the final destination
>>lies on the other side of some router, will the negotiation just
>>fail for the TCP connection and just not allow compression and not cause 
>>anything
>>to break?
>     
>     
>window size.  If RFC1106 size is used (this uses TCP options), can this be
used 
>along with header compression?
>     
>     
>     
>Thanks,
>Pat Medved
>     
>medved@npr.legent.com
>
>