Atlanta-NAP's choice of switch

Paul J. Zawada zawada at ncsa.uiuc.edu
Sun Oct 27 23:06:13 UTC 1996


At 08:12 PM 10/26/96 -0700, Jonathan Heiliger wrote:
>On Sat, 26 Oct 1996, Darin Wayrynen wrote:
>What else would you suggest?  Gigabit Ethernet hasn't been standardized
>yet, Cisco doesn't make a HIPPI interface, and some people prefer to not
>use ATM.  FDDI has proven to very reliable, etc.  Having ISPs continue to
>grow egress bandwidth has shown to be a bigger problem than the switch
>fabric at the larger exchange points. 

HIPPI would probably make a poor technology to build a NAP out of.  The
set-up time that the HIPPI-SC protocol takes to set up a connection
introduces too much latency for small packets.  While HIPPI works great for
large bulk data transfers (using tuned applications, including tuned ftp
implementations), I think the performance would be pretty poor when it came
to exchanging tinygrams between NSPs.  HIPPI might be better than full
duplex FDDI, but something like OC12 ATM may be better, even after
considering the packet-shredding factor...

--zawada




Paul J. Zawada, RCDD     | Senior Network Engineer
zawada at ncsa.uiuc.edu     | National Center for Supercomputing Applications
+1 217 244 4728          | http://www.ncsa.uiuc.edu/People/zawada





More information about the NANOG mailing list