« ISP Contention Ratios and average traffic per user – some data points | Main | Broadband capacity - the impact of congestion »

December 15, 2009



You could have a look at the work done by Aiko Pras, Michiel Mandjes and Remco van de Meent of Twente University on this topic. Van de Meent's Phd thesis was on the topic of link dimensioning. to have a look at some of his work see here http://eprints.eemcs.utwente.nl/6869/01/TR-CTIT-06-56.pdf


A maxim I heard a long time ago in the IEEE Ethernet standards effort:

"The most effective form of QoS is overprovisioning."


Thanks Rudolph, thats a very interesting article, and directly relevant to what I was going to write about next.

Juliet Bradman

Thanks you so much for sharing this useful information with us.I think people should know about it.There are many people who don't know these broadband capacity.


I work for a small ISP and see the same thing every day -- the links with a few dozen or hundreds of users is much more varied than our aggregated upstream. Similarly, other ISPs in our bandwidth-buying group who are much smaller have much greater variations in average to peak burst, which gives them the greater benefit in our aggregate purchasing than ourselves, which is more steady. They benefit more because the can meet the burst but are billed only on their average usage.

In summary, the smaller the number of end nodes, the greater the cost because (a) peak bandwidth per subscriber ratio and (b) lower purchase volumes.

Juliet Waugh

One thing I would like tell you that I really enjoyed reading your post that was too informative.I think this is a great information about broadband.

The comments to this entry are closed.

My Photo

Search this Blog

Subscribe by Email

March 2014

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          


Site Meter

Upcoming Travel & Conferences

Twitter Feed

Become a Fan