[e2e] traffic engineering considered harmful

Jon Crowcroft J.Crowcroft at cs.ucl.ac.uk
Wed Jun 13 00:00:23 PDT 2001

In message <Pine.GSO.4.21.0106121701120.4684-100000 at rac5.wam.umd.edu>, Manish K
arir typed:

 >>definately a very e2e idea, but misses one crucial fact, that 
 >>the network providers are different from the consumers(the ends).
 >>the network providers can never be convinced to trust the ends to 
 >>do the right thing(did'nt somebody on this list bring up the issue of
 >>trust earlier...??) think about it...if you spend millions building 
 >>out your network, would you let somebody else control it??

you might want someone to USE it - this is a chance to maximise the
number of people using your network if its BETTER than someone

 >>On Tue, 12 Jun 2001, Jon Crowcroft wrote:
 >>> so here's a new topic
 >>> Why I think traffic engineering is a Bad Thing
 >>> 	by jon crowcroft
 >>> I think traffic engineering is a Bad Thing
 >>> because it is only a good thing for ISPs. 
 >>> traffic engineering is about maximising utilsiation of 
 >>> your resources - whether its fiddling with MPLS FEC or
 >>> hacking the metrics in OSPF, what you're trying to do is post hoc
 >>> rationalisation.
 >>> what would be better? what would be better would be to give the users
 >>> a choice - if loose source routing (or strict, or scalelble
 >>> multihoming, or IPv6 GSE) worked well, then users would choose the
 >>> best path for their traffic. if a user who sees a poor packet loss
 >>> (or ECN marking) rate on path a can try a different path, then a market 
 >>> in alternate paths would develope - this would obviate the need for
 >>> traffic engineering, and would make it an end2end selection (hence why
 >>> i am posting this here)
 >>> What's hard about this? routers. why is it hard. well, only coz we
 >>> havnt thought about optimising them for it - fact is it oughta be easy
 >>> coz they have LESS decision to make if we do this, since we take the
 >>> decision out of their silicon paws and puit it into the C-shape hands
 >>> of end system software....
 >>> thats what i think....but only for the next 3 nanoseconds...
 >>> chrz
 >>> jon 



More information about the end2end-interest mailing list