Re: [SLUG] NFS and bandwidth bind nics

From: Ed Centanni (ecentan1@tampabay.rr.com)
Date: Thu May 09 2002 - 06:09:07 EDT


I think someone will write an IP spec for IP over just about anything.
My personal favorite is:

http://rfc.sunsite.dk/rfc/rfc1149.html A Standard for the Transmission
of IP Datagrams on Avian Carriers -- also known as CPIP: Carrier Pidgeon
Internet Protocol. This was actually implemented on Linux and tested.
http://www.blug.linux.no/rfc1149/ Here's the ping log:

vegard@gyversalen:~$ ping -i 900 10.0.3.1
PING 10.0.3.1 (10.0.3.1): 56 data bytes
64 bytes from 10.0.3.1: icmp_seq=0 ttl=255 time=6165731.1 ms
64 bytes from 10.0.3.1: icmp_seq=4 ttl=255 time=3211900.8 ms
64 bytes from 10.0.3.1: icmp_seq=2 ttl=255 time=5124922.8 ms
64 bytes from 10.0.3.1: icmp_seq=1 ttl=255 time=6388671.9 ms

--- 10.0.3.1 ping statistics ---
9 packets transmitted, 4 packets received, 55% packet loss
round-trip min/avg/max = 3211900.8/5222806.6/6388671.9 ms
vegard@gyversalen:~$ exit

I see that it has recently been extended
http://rfc.sunsite.dk/rfc/rfc2549.html to allow commercial airlines as a
carrier for the primary transport media, round-robin queueing, fair
weight queueing, traffic shaping, encapsulation, quality of service
levels and many other neat features.

Ed

Mario Lombardo wrote:

> Jeepers, I never thought I'd somebody bother with a spec for IP over
> 1394. That's neat, but I don't know how much demand there is when
> there's gigabit Ethernet.
>
> I think I'll work on a spec for IP over concrete; that way, people can
> get their email through their tennis shoes. How about IP over
> plywood?...plywood is good.
>
> Mario
>
>
>> On Wed, 2002-05-08 at 13:40, Patrick (at work) wrote:
>>
>>> Question is how can it be done with Linux.
>>>
>>> From: "Mario Lombardo" <mario@alienscience.com>
>>> > I didn't know you could bind ports like that. Cool!
>>
>>
>> You *can* bind ports like that, and it's a feature of the native 2.4
>> kernel, but it's not trivial to do and does require a decent ethernet
>> switch that supports ethernet "bonding".
>>
>> If the network is your bottleneck though, this is an option.
>>
>> Another option is to use IP over IEEE-1394 firewire.
>> 400Mbps vs 100Mbps. Unfortunately, cables are a bit beyond cat5, and I
>> can't say I've ever seen a firewire end crimper for making your own
>> cables ;)
>>
>> - Ian
>
>
>



This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 18:02:47 EDT