🚀 go-pugleaf

In article <phc8hcg@rhyolite.wpd.sgi.com> vjs@rhyolite.wpd.sgi.com (Vernon Schryver) writes:
>In article <1992Sep6.211409.16525@menudo.uh.edu>, stan@casc.math.uh.edu (Stan Hanks) writes:
>> See my paper from Winter '92 USENIX. We've run FDDI rings that are *BIG* --
>> right at 80 miles long with only 5 nodes. 

>What TTRT do you use?

The Evil 8 ms. (enter egg-on-face-mode) It's the one value I didn't screw 
with. Maybe next time I do testing. (enter normal-mode)

>What is the aggregate ring bandwidth that you use?  (in other words,
>(microseconds while some station is transmitting data) / 1000000)

Calculate from the above, plus ring latency. Should be a smidge under 80 Mb/s.

>What is the maximum TCP/IP speed you can do between any pair of
>stations?
>
>I wouldn't be surprised if no or few of your 5 stations do TCP/IP on
>their own behalf, so maybe this is the better question;
>what is the maximum forwarding rate (either bridging or routing) in
>Bytes/sec between any pair of stations?

Correct. These stations are packaging LAN traffic into closed user groups
built out of TCP/IP streams. Their transmission  rates are limited by input 
rates into them. I'm (supposedly) going to have an opportunity to test this 
stuff in a multi-VAX environment running LAVC over FDDI (plus some other 
stuff) Real Soon. Until then, I won't have any real "fully loaded testing"
info. The best I've seen to date is ~35 Mb/s end-to-end, and I know
that's a function of how the data is getting into the FDDI node (i.e.
the bottleneck is *known* to be elsewhere in the system).

I'll try to remember to let you know what I find out. If nothing else,
look me up at InterOp in/near the MFS Datanet booth.

--
Stanley P. Hanks      Principal Scientist,      Technology Transfer Associates
U.S. Snail:        P.O. Box 2087,   Bellaire    TX    77402-2087
FedEx/Shipping:    5303-F Glenmont, Houston     TX    77081
e-mail: stan@karazm.math.uh.edu   voice: (713) 661-2084     fax: (713) 661-0633