From: Jake Reynolds (JREYNOLDS@uscentral.org)
Date: Mon Apr 14 2003 - 15:35:26 GMT-3
From what I understand the byte count is arbitrarily chosen to a point
unless explicitly defined. If the scenario doesn't define byte counts
then I just use 10K as my byte count and base everything off of that.
So using 10K 50% would be 5K and so on. You don't want to use too big
of a byte count of course because bandwidth distribution will get
jerky. Also, too small of a byte count will byte you as well. Just
understand how to figure proper byte counts to match bandwidth
percentages when they are given. This formula has been discussed at
length on the list and can be found easily in the archives.
Jake Reynolds
Systems Engineer - Information Systems
CCIE #11224, MCSE NT4 & W2K, CCNA, CCNP, A+
US Central Credit Union
9701 Renner Blvd.
Lenexa, KS 66219
Office- 913.227.6122
Cell- 816.305.6785
-----Original Message-----
From: Danny Andaluz [mailto:dannyandaluz@comcast.net]
Sent: Monday, April 14, 2003 11:10 AM
To: ccielab@groupstudy.com
Subject: Custom Queueing
I have seen countless definitions on what it does, how it works,
etc... What I don't understand is how the byte-count for the queues
is
derived. I have seen examples say to give 25% to this protocol in
this
queue or 50% to this protocol in that queue. 25% of what??? I was
under the impression that the bandwidth configured on the interface is
what gets carved up. I saw one example that had two queues getting
50%
of the bandwidth each. The byte-count was set to 1000 on each. This
means that the total is 2000, but where does the 2000 come from??
Thanks,
Danny
This archive was generated by hypermail 2.1.4 : Thu May 01 2003 - 13:35:52 GMT-3