Merit Network
Can't find what you're looking for? Search the Mail Archives.
  About Merit   Services   Network   Resources & Support   Network Research   News   Events   Home

Discussion Communities: Merit Network Email List Archives

North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

Re: Intradomain Traffic Engineering

  • From: Robert Boyle
  • Date: Wed Jan 18 00:37:49 2006

At 12:06 AM 1/18/2006, you wrote:
(snip)
wrong prediction, the technique suffers very high MLU (as high as 140%).
Basically, I have the following two questions:
1. In the traces I have, there exist several intervals with a huge, sudden increase of traffic on some links. The prediction model I use cannot predict those 'big spikes'. Do these 'big spikes' really happen in operational networks? Or are they merely measurement errors? If they really happen, is there a gradual ramp up of traffic in smaller time scale, say, on the order of tens of seconds? Or do these 'big spikes' really occur very quickly, say, in a few seconds?
Nobody can predict them so you build your network with excess capacity from an overhead standpoint as well as a link standpoint. Here are several reasons for variation and unpredictability. This is not a comprehensive list and I'm sure others will add to it.

CNN or other major network coverage including major advertising events - super bowl, victoria's secret show, etc. (10s of seconds)
SQL Slammer / Code Red / Nimda / or other major fast moving outbreaks (10s of seconds - maybe. We saw the spread of SQL slammer within 2 seconds to many unmanaged colo customer machines)
depeering of any two or more large networks or routing mistakes or flapping thus dampening (a few seconds to 10s of seconds to hours)
major provider outage which moves flows to other paths (a few seconds to 10s of seconds)
fiber cuts / regional power outages (a few seconds to 10s of seconds)
significant events such as 9/11 & Katrina (a few seconds to many hours)

2. I have the option to make a tradeoff between average case performance and worst case performance guarantee, but I don't know which one is deemed more important by you. Are ISP networks currently optimized for worst case or average case performance? Is the trade-off between these two an appealing idea, or may the ISP networks are already doing it?
Each ISP makes their own decisions based on their business needs, budgets, and promised SLAs to customers

-Robert



Tellurian Networks - The Ultimate Internet Connection
http://www.tellurian.com | 888-TELLURIAN | 973-300-9211
"Well done is better than well said." - Benjamin Franklin





Discussion Communities


About Merit | Services | Network | Resources & Support | Network Research
News | Events | Contact | Site Map | Merit Network Home


Merit Network, Inc.