SUMMARY - FDDI and Suns, (story with a happy ending)

From: System Administrator (sysadmin@toshiba.tic.oz.au)
Date: Thu Jul 22 1993 - 09:56:44 CDT


On the 9th July I posted the following question about Suns and FDDI performance:

] Hi all. I am hoping to tap some NetWisdom here. We are currently contemplating
] introducing some FDDI at our site. The main aim is to connect our file servers
] to the central hub at high speed. I've been keeping an eye out for information
] on Suns & FDDI and most of what I hear is rather disturbing. I sounds as though
] the current implimentations for the Sun provides worse performance than
] ethernet (slow drivers, etc...) Can anyone comment on this?
] We do not have any other equipment (appart from Synoptics) that would be using
] the FDDI, so if the Suns do not use it properly there would be no real
] bennifit.
]
] Who out there is using FDDI on Suns? What sort of performance are you getting?
] Why did you choose (if you had a choice) to go that route? Are there any third
] party products that are worth looking at?
]
] For what it is worth, the server(s) will be either 2 x SS10 or 1 x SS1000 and
] the hub is a Synoptics with an EtherSwitch board bridging to 10base2 networks.
]
] Anyone care to comment?

Since then I have been inundated by responses both from users and from would-be
users. The quick synopsis is that Suns using FDDI give quite good performance.
People quote from 4 to 12 times faster (depending upon application) with the
average appearing to be around 25~30MBit/sec. No one reported any real problems
installing or using FDDI.

The types of cards in use include:

1 LANNET (LFD-104)
3 Crescendo
2 Sun
6 Network Peripherals
2 Interphase

The types of FDDI hubs included:

1 LANNET hubs
1 Digital
1 Synoptics
2 3Com
1 Interphase fiberhub 1600

Plus some people did not specify equipment, just stated they were very
pleased.

No one claimed any one board was vastly better than any other, though most
agreeded that Suns implimentation of FDDI was slower than some of the
competition (Still faster than ethernet).

I have individual repsonses from people and will provide a more detailed summary
if needed.

Thanks to the following people

wojsyl1@appli.mimuw.edu.pl (Wojtek Sylwestrzak)
miker@il.us.swissbank.com (Mike Raffety)
grover@vulcan1.hac.com (Dean Grover)
JanBerger.Henriksen@ii.uib.no
kevin@uniq.com.au (Kevin Sheehan)
Matt_Mauss_-T@hqdev.3mail.3com.com
root@ewi.ch (Christoph Rothlin)
jpack@fnoc.navy.mil (Jeff Pack)
casper@fwi.uva.nl (Casper Dik)
dan@bellcore.com (Daniel Strick)
beal@udgserv.cencar.udg.mx (Lupemaria Beal)
nick@dsd.es.com (Nick Nickerson)
shandelm@jpmorgan.com (Joel Shandelman)
lhume@macadam.mpce.mq.edu.au (Leigh Hume)
cal@soac.bellcore.com (Christian Lawrence)
breen@noaapmel.gov (David Borg-Breen)
aribi@geo.vu.nl (Arie Bikker)
myk@artel.com (Mike Steadman)
dwight@hyphen.com (Dwight Ernest)
dbm@mosaic.uncc.edu (dwayne mcneil)
exudnw@exu.ericsson.se (Dave Williams)
sid@Think.COM (Sid J. Stuart)
xcpgzm@atom.oryx.com (Phil Meyer)
S.C.Fitch@open.ac.uk (Steve Fitch)
robert_laskin@applenet.CMC.COM (Robert Laskin)

John
==============================================================================
= John Baird ACSnet: john@toshiba.tic.OZ.AU =
= Toshiba International Corporation Pty Ltd Phone: +61-2-428-2077 =
= Sydney, Australia Fax: +61-2-418-7791 =
==============================================================================



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:08:02 CDT