Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!touch!john
From: john@touch.touch.com (John Weald)
Newsgroups: comp.dcom.lans
Subject: Re: Wanted: FDDI LAN analyzers
Message-ID: <200@touch.touch.com>
Date: 18 Jan 91 18:02:15 GMT
References: <5670@exodus.Eng.Sun.COM> <122@thoreau.nsc.com> <42620@ut-emx.uucp>
Organization: Touch Communications Inc, Campbell, CA
Lines: 67

In article <42620@ut-emx.uucp> spurgeon@atget.cc.utexas.edu.UUCP (Charles E. Spurgeon) writes:
>In article <122@thoreau.nsc.com> macomber@thoreau.nsc.com ( =Robert Macomber) writes:
>>In article <5670@exodus.Eng.Sun.COM>, randolph@cognito.Eng.Sun.COM (Randolph Fritz) writes:
>>> I'm looking for recommendations for FDDI LAN analyzers, especially any
>>> that can generate test traffic at high speeds or have FDDI protocol
>>> verification suites.  Would you-folk please write or post if you know
>>> of any.
>>> 
>>Two companies making FDDI LAN analyzers that I know of are:
>>
>>	Digital Technology, Inc.
>>	tel: (513) 443-0412
>>	fax: (513) 226-0511
>>
>>	Tekelec
>>	tel: (818) 880-7817
>>	fax: (818) 880-6993
>>	
>>DTI has three products (that I know of) 1) an interface for an HP logic
>>analyzer, 2) a symbol generator (PC card) and a passive FDDI LAN
>>analyzer (uses fiber-optic splitters).  I've been told that the DTI
>>products don't use anybody's chipsets (NSC, AMD, ...) to avoid biasing
>>their monitor.  The DTI monitor is capable of generating detailed,
>>timestamped event traces from up to 4 places in an FDDI ring.  It is
>>especially useful for gathering information like line state changes or
>>"who claimed first".
>>
>
>I saw a demo of the DTI equipment yesterday.  They came by and hooked

We had a demo of the Tekelec ChameLAN 100 yesterday, and a demo of the DTI
a few months ago.

The general vote was that we perfered the Tekelec. Its cheaper, smaller and
has a better user interface.

They use an AT portable with a flat colour screen running Interactive's
386/ix UNIX. The user interface is Motif (thus you can X-window into it
from another X-server an access the user interface). The interface is
still not as nice as the Network General LAN Sniffer.

To capture they use a RISC chip front end processor and AMD chipset.
The front end has a 500MB bus.  It is completely transparent and 
passive on the ring (looks like a FDDI repeater at the PMD level) 
and connects via standard connectors. 
The capture buffer is either 5MB or 20MB. Usual triggers filters
etc.

It decodes TCP/IP today, but our need is to decode OSI. Thus we did not
go into its low level FDDI features that one might use for developing FDDI
drivers or hardware (IBM are giving us an FDDI card with software).

They are porting their OSI decode from one of their other X.25
products, apparently they have a decode language and a "compiler" 
that "compiles" into C. You can buy the compiler and source to their decode
stack for a (small) fee or buy the executable decode stack.

There is an optional simulation package that is not yet ready. It will allow
you to inject all kinds of errors and generate taffic.

Base price is $37,500 for the 5MB, 100mb harddisk or 
$46,000 for 20MB,  200mb harddisk. Decode software etc. extra.
-- 
John Weald, Touch Communications, Campbell, CA

uunet!touch!jweald
jweald@touch.com