Date: Fri, 8 Aug 2014 01:28:20 -0700 From: Doug Hardie <bc979@lafn.org> To: Odhiambo Washington <odhiambo@gmail.com> Cc: questions <questions@freebsd.org> Subject: Re: TCP/IP on the way out? Message-ID: <94DD6882-7A24-443B-85BF-090416562120@lafn.org> In-Reply-To: <CAAdA2WNmX628FLiPDKRrpwe9Pzov0JysiPetAKbVyHyFnttveQ@mail.gmail.com> References: <CAAdA2WNmX628FLiPDKRrpwe9Pzov0JysiPetAKbVyHyFnttveQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 8 August 2014, at 00:00, Odhiambo Washington <odhiambo@gmail.com> = wrote: > I trust that all is well with everyone. >=20 > I have seen this article which sounds much like a dream, but seems = true. >=20 > = http://www.networkworld.com/article/2459286/why-tcp/why-tcp/ip-is-on-the-w= ay-out.html >=20 > I'd love to hear the views of those who understand the network stack. A quick reading indicates that the protocol is almost identical to the = IMP protocol used by ARPANET. It is more efficient of bandwidth and = redundant paths. However, TCP/IP won out simply because it is a very = simple protocol that works well and can be easily implemented in very = small devices. What the article doesn't address is the path = determination process. That is where the bulk of the code and overhead = will be found. ARPANET used a static network structure that required = re-loading of all IMPs when the network changed in order to keep the = overhead down to something manageable. IPv6 is somewhat of a compromise = in that area, but it still incurs significantly more overhead than IPv4. = IPX was much worse yet. We never could get a LAN with 100 workstations = to be useable, where we could easily put 200 on a LAN using IPv4.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?94DD6882-7A24-443B-85BF-090416562120>