From: Andrew Lee Lissitz (alissitz@corvil.com)
Date: Mon May 30 2005 - 23:26:37 GMT-3
Hello Everyone,
Cisco's compression is pretty poor on anything that is already compressed;
it will add overhead to the packets and slow the transmission down
significantly. It is easy to test Cisco's compression. Be sure to test
this before rolling it out if you decide to go forward.
If you have the budget, look for Expand, Riverbed, Packeteer, or Peribit
(now juniper). Again, do not believe the marketing and test each device.
Simple question: why are you looking for compression? If you are having
application quality problems, compression may not help that much depending
on the app... you may need more bandwidth or a TCP optimization product...
(long distance links)
Kindest Regards all,
Andrew Lee Lissitz
908.303.4762
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
simon hart
Sent: Monday, May 30, 2005 6:43 AM
To: Quetta Walla; ccielab@groupstudy.com
Subject: RE: Compression on low band link
The only compression algorithm you can use on an HDLC link is stac (the
Lempel-Ziv compression algorithm).
Enter the command compression stac directly on the interface at both ends of
the link.
The stac algorithm is heavy on the CPU, therefore if you are using a lot of
CPU already, use with caution. If this is the case, then you can change the
link encapsulation to PPP and then use predictor. The trade off here is
that you will not use so much CPU, but will have to have enough memory.
HTH
Simon
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
Quetta Walla
Sent: 30 May 2005 10:49
To: ccielab@groupstudy.com
Subject: Compression on low band link
Hey,
What compression method is recommended on say 256k hdlc link?
And why?
Thanks
--
This archive was generated by hypermail 2.1.4 : Fri Jun 03 2005 - 10:12:03 GMT-3