Hi Mazin, Thanks. I'm totally agreed with your calculation, but just to emphasize that how you treat the remaining bandwidth in % calculation. As long the *remaining* is *equally* shared, then i believe we got the point.
And you got the point. :) TIA Shingei On Fri, Oct 22, 2010 at 4:10 AM, Mazin Ahsan <[email protected]> wrote: > Thanks Shingei, > > Lets say that 4:4 ratio shares the 50% bandwidth among the two queues. > According to the task we also have to have 20% bandwidth with priority-queue > . > > The 3750 have two input queues. If we are to give one queue priority with > 20% bandwidth according to the question. Then we have to share the rest > between the two queues equally, we are giving 4:4 ratio to the queues. > > Lets say we have Fasthethernet link > > Queue 1 : 20% Prioritized. =============== 20Mbps > Queue 1 : 50% Bandwidth of 80 Mbps======= 40Mbps > Queue 2 : 50% Bandwidth of 80 Mbps======= 40Mbps > > Queue 1 will be prioritized queue and will get priority upto 20% bandwidth > after which the remaining of the traffic will be shared with the ratio of > 4:4 . Meaning 50% to each queue. > > > HTH > Mazin > > - > --------------------------------- > > > > Hi Greg, > > I think Amr is referring to the WB question that require the remaining > bandwidth of 80% (after deducted the 20% for PQ on queue1) to be equally > shared between Q1 and Q2. > In this case, neither Q1 4/(4+4) nor Q2 4/(4+4) will get 50%. Q1 and Q2 > will > get 40% each, i believe this achieve what the question asked which equally > shared between the remaining bandwidth. > > If configured Q1 5/(5+5) and Q2 5/(5+5) and the result multiply by 80%, you > will still get the 40% each queue, so mean it doesn't a matter what value u > configured, as long both of them are same. > > For me is quite straight, if the question ask to disable the PQ, then > n/(n+y) or y/(n+y) the valuable n and y can be ANY value that add up total > to be 100, which mean it can be (25+75), (40+60)...etc. > But if the the question asking to turn on PQ and equally shared the > remaining bandwidth, i think the same value of n and y is applied. > > Please correct me if wrong. > IMHO. > Shingei. > > > _______________________________________________ > For more information regarding industry leading CCIE Lab training, please > visit www.ipexpert.com > >
_______________________________________________ For more information regarding industry leading CCIE Lab training, please visit www.ipexpert.com
