[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference 7.286::fddi

Title:FDDI - The Next Generation
Moderator:NETCAD::STEFANI
Created:Thu Apr 27 1989
Last Modified:Thu Jun 05 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2259
Total number of notes:8590

1914.0. "4 stations and concentrator ?" by ANNECY::BONNIER (Jean-Luc ... CSS Annecy France) Fri Dec 22 1995 11:10

We are proposing the following FDDI config for a customer:

	2 AlphaServer 1000 4/266 with each a DEFPA-DA
	2 AlphaServer 2100 4/233 with each a DEFEA-DA

These 4 machines are running Digital UNIX V3.2C or later.
They communicate over DECnet or TCP/IP.

Do we need a FDDI concentrator for the dual-ring to operate properly ?

Which type of cables between the FDDI controllers ?

Thanks for your support.
Jean-Luc
T.RTitleUserPersonal
Name
DateLines
1914.1NETCAD::STEFANIMachines to humanizeThu Dec 28 1995 09:0438
>>We are proposing the following FDDI config for a customer:
>>
>>	2 AlphaServer 1000 4/266 with each a DEFPA-DA
>>	2 AlphaServer 2100 4/233 with each a DEFEA-DA
>>
>>These 4 machines are running Digital UNIX V3.2C or later.
>>They communicate over DECnet or TCP/IP.
>>
>>Do we need a FDDI concentrator for the dual-ring to operate properly ?

    No.  If you're connecting all four (4) devices on the dual-ring, then
    you do not need a concentrator.
    
>>Which type of cables between the FDDI controllers ?
    
    The DEFEA-DA uses ANSI MIC connectors while the DEFPA-DA uses SC
    connectors.  Depending on the order in which you connect the systems
    you'll need two or more SC-MIC cables.  For example, if you were wiring
    the systems as:
    
    			1000
    
    		2100		2100
    
    			1000
    
    you'll need 4 SC-MIC cables.  If instead, you were wiring the systems
    as:
    
    			1000
    
    		1000		2100
    
    			2100
    
   you'll need 1 SC-SC, 1 MIC-MIC, and 2 SC-MIC cables.
    
    - Larry
1914.235356::RABAHYdtn 471-5160, outside 1-810-347-5160Thu Dec 28 1995 11:098
    re .0:
    
    Consider using a concentrator or two anyway.  It will be more flexible
    in terms of cabling and more tolerant of systems being shutdown.
    
    Also, are the systems very near to each other?  Or will they be spread
    out somewhat?  Chances are, if they are spread out, they will be cabled
    to an LIU which has ST connectors.
1914.3NETCAD::STEFANIMachines to humanizeThu Dec 28 1995 16:319
    >>Consider using a concentrator or two anyway.  It will be more flexible
    >>in terms of cabling and more tolerant of systems being shutdown.
    
    If you're connecting dual-attach stations to the dual-ring, then there
    is NO reason to purchase a concentrator.  What would you connect to it,
    anyway?  Now, if instead you're going to use single-attach stations
    connected to a concentrator, that would make sense.
    
    /l
1914.4more details on cables ?ANNECY::BONNIERJean-Luc ... CSS Annecy FranceFri Dec 29 1995 06:0025
Thanks for your reply.

The four machines are integrated into a dual-cabinet config.
The 2 2100s are rack-mounted in a cabinet and 2 1000s in the other cabinet.

Then the best topology is the second one you suggested in .1

	2100 (MIC)		1000 (SC)


	2100 (MIC)		1000 (SC)

Now let's look at the cables we need and the associated part numbers:
	2 cables SC-SC		BN34B-03	(3 meter-cable is enough)	
	4 cables MIC-SC		BN34D-03
	2 cables MIC-MIC	which part number ??

Are you OK with the number of cables and part numbers ?
We didn't find the MIC-MIC cable reference in the catalogs ... can you help ?

Maybe I can find these infos in the FDDI Network Configuration Guidelines
(EK-DFDDI-CG) manual. Is this guide available on Easynet and where ?

Thanks again.
JLuc
1914.5BN24B47259::FARRAUTOGutta caveat lapidemFri Dec 29 1995 07:555
    BN24B-01/03/10/20 is the ANSI MIC to ANSI MIC cable
    
    Regards
    
    Fab
1914.635356::RABAHYdtn 471-5160, outside 1-810-347-5160Fri Dec 29 1995 10:475
    re .3:
    
    Attaching the A port and the B port of a DAS to M ports on two
    different concentrators is an excellent way to connect a system to an
    FDDI ring.  The failure of either concentrator is fully tolerated.
1914.7NETCAD::STEFANIMachines to humanizeFri Dec 29 1995 14:0910
    >>Attaching the A port and the B port of a DAS to M ports on two
    >>different concentrators is an excellent way to connect a system to an
    >>FDDI ring.  The failure of either concentrator is fully tolerated.
    
    I'm well familiar with dual-homing.  However, given the size of the
    proposed network (only 4 nodes) I'd suggest that it's overkill.  My
    main point in .1 was to dispute the outdated and incorrect assumption
    that you MUST have concentrators in your network. 
    
    /l
1914.8FDDI config. manual on the net ?ANNECY::BONNIERJean-Luc ... CSS Annecy FranceWed Jan 03 1996 05:3010
Thanks for your reply.

Anyway I was thinking that the "FDDI Network Configuration Guidelines"
(EK-DFDDI-CG) manual would be useful. 

Then I reiterate my request ... Is this guide available on Easynet and where ?

Thanks again.
JLuc

1914.9Available on TNPUBS clusterCX3PST::HTKING::M_SHELDONThu Jan 04 1996 09:3311
Directory INTER::DISK$USER657:[PUBLIC.HRD.STOW]

EK-DFDDI-CG-001.PS;1
                        1388/1389     12-NOV-1991 13:17:02.00  (RWED,RWED,RWED,R
E)

Total of 1 file, 1388/1389 blocks.
$

						Mike Sheldon
						CSC/CS Network Support
1914.1049766::HOTZGregor; MCS, NaC Support, SchweizFri Jan 05 1996 03:371
The hardcopy I have is -002 and from Jun 1992.
1914.11number of FDDI cables ?ANNECY::BONNIERJean-Luc ... CSS Annecy FranceWed Jan 10 1996 10:1620
Coming back to reply .4

Given the AlphaServers connections, I'm wondering about the number of 
FDDI cables I need to order ...

I was proposing
	2 cables SC-SC		BN34B-03
	4 cables MIC-SC		BN34D-03
	2 cables MIC-MIC	BN24B-03

But I think I doubled the number of cables. Is it correct to say that
each cable is made of two wires (one for transmission, the other for reception)?

And in fact I need only
	1 cable SC-SC		BN34B-03
	2 cables MIC-SC		BN34D-03
	1 cable MIC-MIC		BN24B-03

Thanks for your quick reply.
JLuc
1914.12NETCAD::STEFANIMachines to humanizeWed Jan 10 1996 11:3513
>>But I think I doubled the number of cables. Is it correct to say that
>>each cable is made of two wires (one for transmission, the other for reception)?
>>
>>And in fact I need only
>>	1 cable SC-SC		BN34B-03
>>	2 cables MIC-SC		BN34D-03
>>	1 cable MIC-MIC		BN24B-03
    
    Yes.  Each cable contains a fiber optic strand for receive and another
    for transmit along with the appropriate connector on either end, so you
    should only need to purchase 4 (total) cables.
    
    /l
1914.13OBR device ?ANNECY::BONNIERJean-Luc ... CSS Annecy FranceThu Jan 11 1996 02:5411
Thanks a lot for your reply, Larry.

I have another point to clarify now.

What happen in the dual ring if one AlphaServer is switched off ?
Do I need an Optical Bypass Relay device connected to each DEFEA/DEFPA to 
make the ring continuity ?
Which type of device ? Any idea of price ?

Thanks again.
JLuc
1914.1435356::RABAHYdtn 471-5160, outside 1-810-347-5160Thu Jan 11 1996 10:0611
re .13:

>What happen in the dual ring if one AlphaServer is switched off ?

After a very brief transition, the ring will wrap - no loss in performance or
function.

A second non-adjacent break partitions the ring.

Optical bypasses are not required unless you want to tolerate two non-adjacent
breaks.
1914.15OBR device price/reference ?ANNECY::BONNIERJean-Luc ... CSS Annecy FranceThu Jan 11 1996 10:515
Can you give references and price of OBR devices working with Digital 
FDDI controllers ?

Thanks 
JLuc
1914.16NETCAD::STEFANIMachines to humanizeThu Jan 11 1996 11:088
    There's a note somewhere in this conference that describes the AMP
    part.  I'd contact AMP and get a catalog of their fiber optic gear. 
    I've found that they make OBR's with both ANSI MIC and SC
    connectors/cables, so you can get the exact parts you'll need for each
    node.
    
    -Larry
    
1914.17refer 1706.2ANNECY::BONNIERJean-Luc ... CSS Annecy FranceFri Jan 12 1996 05:014
Thanks Larry

I presume you are referring note 1706.2