[ih] Separation of TCP and IP

[Toerless Eckert]

On Thu, Jun 23, 2022 at 10:37:37PM -0400, John Levine via Internet-history wrote:
> It appears that Jack Haverty via Internet-history <jack at 3kitty.org> said:
> >But I think the motivation for the TCP/IP split started much earlier 
> >than when it actually happened.  E.g., I have a faded copy of 
> >"Comparison of TCP and DSP" by Dave Clark (MIT Local Network Note #7, 
> >April 28, 1977).  ...
> 
> I've never understood what the motivation for splitting TCP and IP was,
> not that there wasn't one, but there were so many.

For someone who has not lived through the time of separation, it is
more like "whot ? why would anybody ever think to stick them together"

> It allows UDP for latency critical stuff. Was there per-flow state in
> the routers before the split or was allowing stateless routers another
> benefit? Other things?

Not knowing the history before end of the 80th, let me at least
add the fun experience that even into the 2000th, vendors tried
to sell equipment using per-5-tuple flow forwarding because TCAMs
where cheap back then at linerate and longest-prefix-match lookups where not.

These "routers" if you dare to call them that did explode in
the face of the sales people who pitched them to service providers
by angry engineers of said service providers.

Since then at least that one vendor i remember has become more careful into
what type of forwarding technologies qualify to have that device
be called "router" (hint: it most NOT only do per-flow forwarding).

Cheers
    Toerless

> 
> R's,
> John
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history

-- 
---
tte at cs.fau.de