Re: Handling Packet Misordering?

[Craig Partridge <[e-mail]>]

>My point is that the real alternative to accepting resolution of reordering 
>at the endpoints is doing something in the network while will guarantee 
>sequential packet ordering. That is something that could be pretty 
>interesting to design and implement in terms of complexity and brittleness 
>added to the network, and very possible has the same effects.

Just adding to Fred's comment.

There are a few known ways to use multiple paths for *one* hop and maintain
packet ordering.  Most are *not* work conserving so you lose some capacity
(i.e., capacity grows as .9 * n * <mumble> rather than n * <mumble>).

If you want to keep order when paths diverge for multiple hops, you've
got an exciting research project (esp. if you want to be close to work
conserving....)

Craig