Re: bandwidth percent

From: Arden Packeer (contactme@ardenpackeer.com)
Date: Thu Jul 10 2008 - 18:28:00 ART


Thanks Joseph!

Nortic: As Joseph stated, we have the total reservable bandwidth of
100% of the interface with the "max-reservable-bandwidth 100 command"
on the interface.

The bandwidth percent, priority percent, and max-reservable-bandwidth
command makes reservations from this total reservable bandwidth. After
these reservations have been made the available bandwidth is whats
left over. The bandwidth remaining percent command makes reservations
from this available bandwidth (whats left over of the total reservable
bandwidth after the other reservations is treated as 100%).

Check out the tutorial for a few examples of this:

http://ardenpackeer.com/qos-voip/tutorial-what-is-the-difference-between-banddwidth-percent-and-bandwidth-remaining-percent/

---
Arden Packeer
Cisco Certified Internetwork Expert (CCIE #20716 R&S)
Cisco Systems Instructor (CCSI #31090)

contactme@ardenpackeer.com http://ardenpackeer.com

On Fri, Jul 11, 2008 at 12:26 AM, Joseph Brunner <joe@affirmedsystems.com> wrote: > > Wow harsh dude! > > class-map class1 > match protocol http > > policy-map remaining > > class class1 > priority percent 40 > > class class-default > bandwidth remaining percent 100 > > int s0/0 > max-reservable-bandwidth 100 <- don't forget this... otherwise > you are "maxing" at say 1152 Kbps on a T-1 (75%) > > >Now the question is should "bandwidth remaining percent 100" not be the > >prefered solution as the whole idea behind MQC is to make things simpler > >when having to add/remove configs? > > The IE WB VOL 2 has a million dollar couple of pages on absolute vs > approximate reservations... I'd look at that if your doing the wookbook. > I prefer not using percentages in real world qos configurations. its harder > to undo/redo them when bandwidth becomes scarce- and absolute resv. are > easier to read in the config when a group of people is doing it. > > Part 2: falls under the "consistent units" rule, for those joining in; > > I think if you read this... > > http://ardenpackeer.com/qos-voip/tutorial-what-is-the-difference-between-ban > dwidth-percent-and-bandwidth-remaining-percent/ > > you'll have the answer to part 2 pretty much cleared up. > > > > -----Original Message----- > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of > nortic @hackermail.com > Sent: Thursday, July 10, 2008 6:37 AM > To: ccielab@groupstudy.com > Subject: bandwidth percent > > Hi group, > > Question regarding bandwidth percent, which would/should be the prefered > method: > > One class for x type traffic to be prioritized 40% of the bandwidth then the > default class has to be guaranteed the remaining bandwidth and make sure > 100% of the bandwidth is being used. > > Ok so there are two classes, class-map X and the default class. Class X is > configured with priority percent 40 and class-default configured with > bandwidth percent 60. > > Now the question is should "bandwidth remaining percent 100" not be the > prefered solution as the whole idea behind MQC is to make things simpler > when having to add/remove configs? > > Part2: That said while it works with priority percent it is not possible to > have one class configured with bandwidth percent x and then another class > configured with bandwidth remaining percent 100, so then what is the point > of the bandwidth remaining percent? > > Joe B need not reply. > > > > -- > _______________________________________________ > Get your free email from http://www.hackermail.com > > > _______________________________________________________________________ > Subscription information may be found at: > http://www.groupstudy.com/list/CCIELab.html > > > _______________________________________________________________________ > Subscription information may be found at: > http://www.groupstudy.com/list/CCIELab.html



This archive was generated by hypermail 2.1.4 : Mon Aug 04 2008 - 06:11:54 ART