Re: [foxboro] A mesh question

Tom....some answers

Question:
Does the seven switch per path rule apply to all MESH topologies, or just
inverted tree or ring topologies.

Answer: Yes. It applies to all topologies. This guarantees RSTP operation.
There are many ways to design a Mesh Network beyond even the 4 topologies we
publish. As long as the 7 switch limit is adhered to, your good.


Question:
Is there any downside to interconnecting the A&B edge switches locally to
each other using only 100 BASE FX MTRJ connections, via (P0972YJ)?

Answer: No downside when speaking I/A deliverable. The typical I/A system
communications as released only takes 2-3 % of the 100MB thruput. The 1GB
links are most commonly used for distance rather than thruput as an I/A
delivery. The two 1GB links should be used to connect the switch to your
Root and Backup Root Switches. The requirement for 1GB communications would
only be needed in extreme heavy traffic applications that would "clog" the
100MB pipe, this would not be a typical situation as applied today.
Then the edge switches can tie together if you wish by using the 100MB
standard ports. This would address the issue that you speak of loosing power
to your Root and Backup Root switches and the need to rely on the local edge
switches for communication between stations that need to play together.

Remember, as long as the Root or Root Backup are alive, the best path cost
is: 
Source Edge to Root, then Root to destination edge. If Roots disappear, then
path is edge to edge.



-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx] On
Behalf Of tom.vandewater@xxxxxxxxxxxxxx
Sent: Friday, November 18, 2005 4:18 PM
To: foxboro@xxxxxxxxxxxxx
Subject: Re: [foxboro] A mesh question

Russ,
        We plan to use a Star topology with two root switches, A&B,
connected to 5 sets, A&B, edge switches.  In some early depictions of
the STAR topology I saw the edge switches interconnected with their own
A or B partner.  The current STAR topology drawing in P0700AZ Rev. C
doesn't show the edge switches interconnected but based on what you said
below, it looks like a valid topology as long as you never have a
station to station path that passes through more than 7 switches.

Question:
Does the seven switch per path rule apply to all MESH topologies, or
just inverted tree or ring topologies.

        Our topology plans to use 1000 BASE LX uplinks, (P0972WU),
mounted in an uplink expansion module, (P0972YK), via existing single
mode fiber between our root switch location and our edge switch
locations. =20

Question:
Is there any downside to interconnecting the A&B edge switches locally
to each other using only 100 BASE FX MTRJ connections, via (P0972YJ)? =20

We could also easily interconnect the A & B edge switches locally via
1000 BASE SX MMF but would have to buy Mini GBIC modules, (P0972WT), to
populate the expansion module, (P0972YK).  This adds more cost than just
running an MTRJ Zip cord between two spare ports on the A&B edge blades
that support 100 BASE FX, (P0972YJ).
        Our root switches will be located in our Central Control Room
area and the 5 sets of edge switches strategically located in five
different field/"control and IO" areas of our plant.  If the edge switch
to root switch fiber path is broken we want the edge switches to
continue to be able to control within their own process area and a
direct link between each local A&B edge switch should add some
robustness.

Thanks for any insight.

Cheers,
Tom VandeWater

-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]
On Behalf Of Boulay, Russ
Sent: Thursday, November 17, 2005 10:41 AM
To: 'foxboro@xxxxxxxxxxxxx'
Subject: Re: [foxboro] A mesh question

Stan..

In the Mesh world there are only a few restrictions when connecting the
switched network.

Multiple connections between switches are acceptable.

You could have 15 connections between two switches if you like.

RSTP Rapid Spanning Tree Protocol determines the connectivity and will
block
and open ports between two switches at any given time.=20

Ports to use are determined by:
First)  Path "costs" as deemed by RSTP
Second) The actual Port #  1 vs 2 vs 3 if their path costs are equal.=20
Port 1 open, 2 & 3 blocked. If then connection to Port1 fails, Port 2
would
then open.

Most important rule is:
No more than 7 switches between stations.
So if an AW is to communicate with a particular CP the interconnection
between switches to get from AW to CP must be less than 7 switches.
So all possible paths from one station to another station must be
considered
when designing the layout.

-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]
On
Behalf Of stan
Sent: Thursday, November 17, 2005 9:59 AM
To: Foxboro List
Subject: [foxboro] A mesh question

Given an existing V6 architecture that spans 2 buildings like this:

Building "A" contains the AW, and the CP's
Building "B" contains 2 WP's most of the FBM's a Comm10, and the AB
integrates.

Presently we have fiber between the buildings for nodebus extender, and
filedbus.

I'm thinking that when we upgrade to mes, I should put switches in both
buildings.=20
Given that I will have at least 4 fiber pairs between the buildings, is
there some way
I can "cross connect" the switch pairs in the 2 buildings to increase my
redundancy.
Assuming of course that the CP -> FBM communications moves to the
nodebus.

--=20
U.S. Encouraged by Vietnam Vote - Officials Cite 83% Turnout Despite
Vietcong Terror=20
- New York Times 9/3/1967
=20
=20
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
=20
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Djoin
to unsubscribe:      =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Dleave
=20

=20
=20
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
=20
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Djoin
to unsubscribe:      =
mailto:foxboro-request@xxxxxxxxxxxxx?subject=3Dleave
=20
 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
 

 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: