RE: IBM 8230 Token-Ring concentrator

HI Simon

I do not dispute that Token ring is a better technology than Ethernet. But I
would compare this to VHS and Beta max we know who one, it was not the
technology leader.  The world is going Ethernet, and Switched Routed 100mb
Ethernet with a properly designed network works fine even in large networks.
You mention 100mg Token Ring. To move up to this you will probably have to
replace your nic's and MAU's. While I will I agree that Token ring is a
better technology, I would still recommend Ethernet. As far as price,
Ethernet is less expensive due to economies of scale. There is just more
manufactures, more competition and more users that are keeping costs down.
Look at how much Ethernet equipment is sold by Cisco Bay and Cabletron not
to mention all the little guys vs Token-Ring.

-----Original Message-----
From: owner-midrange-l@midrange.com
[mailto:owner-midrange-l@midrange.com]On Behalf Of Simon Coulter
Sent: Friday, January 01, 1999 7:35 AM
To: MIDRANGE-L@midrange.com
Subject: RE: IBM 8230 Token-Ring concentrator


Hello Russ,

*** Warning, warning, flame war approaching **************

I take issue with your statements.  I do agree with the sugesstion to
evaluate the entire infrastructure but
you seem to consider Ethernet to be an upgrade from Token-Ring.  Why?  The
ONLY advantage to Ethernet is its
price.  It is much cheaper than TR.  But then you get what you pay for --
Ethernet cards have much less
intelligence on board than TR cards.  Ethernet is good for light-to-medium
network loads; TR is good for
medium-to-heavy network loads.

TR uses a much more civilised protocol than Ethernet which is why a lower
rated TR network will perform
better than a higher rated Ethernet network.  At high traffic levels 4Mbs TR
will thrash 10Mbs Ethernet, and
16Mbs TR will give 100Mbs Ethernet a good run for its money.  (I have just
seen information suggesting that
100Mbs TR is available -- must look into this.)  Here are some graphs
showing what happens:

:font style=fixed.

T
h      |
r      |
o  1.0 |                            **********************
u      |                        ***
g   .8 |                     **      @@@@@@@@@@@@@@@@@@@@@
h      |                  **     @
p   .6 |               **    @
u      |            **   @
t   .4 |         **  @ ########
       |      ** @ ###          ####
S   .2 |   * @ #                     ##
       |*@#                             #######
       |_______________________________________###########
           |     |     |     |      |      |      |      |
           .2    .4    .6    .8    1.0    1.2    1.4    1.6
                                Offered Load G

* = Ideal
@ = Token-Ring
# = Ethernet

Throughput S = The amount of data successfully transmitted between nodes in
the network.
Offered Load G = Includes not only the node-to-node data, but also any
acknowledged packets, packets
retransmitted due to errors, and packets associated with network management.
The offered load is always
greater than or equal to the throughput.

Source: Design and Organization of Computing Structures by James H. Herzog.

Please make allowances for the poor quality graphics however you should get
the idea.

:efont.

The basic difference is that TR allows only the device with the token to
transmit; Ethernet lets everyone try
and hopes only one device is _actually_ trying -- they call it collision
detection.

A TR device says "I've got the token, you wait your turn", "OK, I'm done,
it's your turn now".  Ethernet says
"Ooh, nobody's using the LAN, here I go. Oh, damn, packets have collided,
flush the network, everybody try
again".

I realise the protocols are a bit more complicated than I suggest but that
is the gist of it.

The only real problem with TR is that one device may not release the token
so other devices complain -- this
is called beaconing. The naughty device is dropped from the ring and another
device takes over and issues a
new token.  This can be a time-consuming operation and the network suffers
because of it, however, in a well
managed network with good quality hardware and reliable operating systems,
beaconing should be minimal.

Regards,
Simon Coulter.

//----------------------------------------------------------
// FlyByNight Software         AS/400 Technical Specialists
// Phone: +61 3 9419 0175      Mobile: +61 0411 091 400
// Fax:   +61 3 9419 0175      E-mail: shc@flybynight.com.au
//
// Windoze should not be open at Warp speed.
//--- forwarded
letter -------------------------------------------------------
> X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0
> Date: Wed, 30 Dec 98 09:58:54 -0500
> From: "rpopeil" <russ.popeil@ac.avnet.com>
> To: MIDRANGE-L@midrange.com
> Reply-To: MIDRANGE-L@midrange.com
> Subject: RE: IBM 8230 Token-Ring concentrator
> Importance: Normal

>
> Since you are considering new equipment why not look at a total
> infrastructure refresh.  It might be a good time to upgrade the
> infrastructure and perhaps change to fast switched Ethernet.
>
> Russ Popeil
> IBM Certified Specialist
> Avnet Computer- Intergrated Solutions. Bottom-Line Results.
> Office: 516-677-9346 Fax: 516-677-0296
> Pager: 800-759-8888 Pin: 1651954
> Pager eMail: 1651954@skytel.com
> eMail: Russ.Popeil@avnet.com
> http://www.avnetcomputer.com

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator:
david@midrange.com
+---

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---