[Mep-dev] TCP Performance Issues

[Phil Karn Jr, KA9Q]

Paul Williamson wrote:

>     Store and forward includes conventional IP routing. A packet comes in,
>     it's stored in RAM, a routing decision is made, and it's sent back out.
>     There's no conceptual difference between this and a BBS other than the
>     time scale.
> 
> 
> That's a funny thing to hear from somebody who was just preaching the
> gospel of end-to-endness earlier in this same discussion.

Not at all. Both networks can have end-to-end flow control, ARQ error
control and sequencing atop a store and forward network layer. The only
difference is the time scale.

A formal transport layer is more common above IP, but there's nothing
that says you can't have one above email if you want. It's just that
most people are satisfied with email's native reliability. IP and email
really don't differ much conceptually. They differ in format, message
size and time scale, but that's it.

>     If you mean discontinuous networking, yes, that's a qualitatively
>     different class of problems that's defined by the lack of a working
>     end-to-end path for periods much longer than the end-to-end propagation
>     latency.
> 
> 
> That description would seem to apply pretty well to amateur radio! If
> there are solutions, maybe there would be some benefit for amateur radio.

Depends on the type of ham radio. We have continuous links and
discontinuous ones.

My point is that conceptually these links aren't as different as you
think. The differences require more tuning and transport protocol
awareness than fundamentally different protocols.

--Phil