To see if you have the SFM, do a 'show module' from enabled mode and you'll
see output like:

Mod Ports Card Type                              Model              Serial
No.
--- ----- -------------------------------------- ------------------ --------
---
  1    2  Catalyst 6000 supervisor 2 (Active)    WS-X6K-SUP2-2GE
SAD05080DZA
  2    2  Catalyst 6000 supervisor 2 (Standby)   WS-X6K-SUP2-2GE
SAD0522001V
  3   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL0508444Q
  4   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL05084NJS
  5    0  Switching Fabric Module-128 (Active)   WS-C6500-SFM
SAD0447006C
  6   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL050958JE
  7   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL05084R0Y
  8   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL05073WX4
  9   48  48 port 10/100 mb RJ45                 WS-X6348-RJ-45
SAL05073WXL

You'll see (in the example) there's a SFM in slot 5 that's active.  If you
have 2, one will be active and the other in standby (or inactive).....

As far as the GBIC cards overrunning the 32Gbps backplane, it's a
possibility that's happening and here's why.......

>From the URL I had in my previous post:

http://www.cisco.com/warp/public/cc/pd/si/casi/ca6000/prodlit/k6kfy_wp.htm

"The key difference between the two line cards is that the fabric-enabled
cards use a single local switching bus with a bandwidth capacity of 16 Gbps.
The fabric-only line cards use two local switching buses, each clocked at 16
Gbps. Both line cards can support distributed forwarding. The DFC daughter
card is available as an add-on for the fabric-enabled cards. The fabric-only
line cards have the DFC embedded in the system."  (Note this said "both line
cards" meaning fabric-enabled and fabric-only... the DFC is NOT available on
non-fabric-enabled cards)

"The key difference in the DFC-enabled system is that a switching decision
is made on the input module instead of centrally at the Supervisor Engine.
The CEF table used for switching, however, is still calculated centrally at
the Supervisor Engine and then downloaded to the CEF table local to the line
card. Therefore, the Supervisor CEF table and the local CEF tables are
always synchronized."

The Distributed Forwarding Card (DFC, which is basically the same as the PFC
that's on the Sup/MSFC card) is what would allow a card to make it's own
forwarding decisions at layer 2 and/or 3.  If you have non-fabric enabled
cards, then they can not switch between ports on the same card without the
frames traveling along the 32Gbps bus to the Supervisor Engine to make the
switching decision.  If you have fabric-enabled cards (with optional DFC) or
fabric-only cards (that have DFC built-in), then all L2 and L3
switching/routing decisions are made on the card itself, so then switching
and or routing between ports on the same card will happen locally.  Also
cards with the DFC, when making L2/L3 decisions, if the destination is
another fabric-enabled card, send the frame directly to the other card via
the SFM... this is where you can realize the performance of the SFM.  IMHO,
if you have fabric-enabled modules, make sure you get the DFC to go on it to
unlock the full potential of the system.

Check your line cards, because the 16GBIC line card we have is NOT fabric
enabled, so not only does it have to use the 32Gbps bus to send frames
through the Sup. just to deliver traffic between ports on the same card, but
it needs to use the 32Gbps bus to communicate with any other card in the
chassis.

Mike W.

"Larry Letterman"  wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> not necessarily, a lot of the processing for port to port is done on the
> line cards..
>
>
> Larry Letterman
> Cisco Systems
> [EMAIL PROTECTED]
> ----- Original Message -----
> From: "Kris Keen"
> To:
> Sent: Tuesday, April 30, 2002 10:42 PM
> Subject: Re: why SFM? [7:42877]
>
>
> > Can we determine if we have SFM cards?
> >
> > We have 2 x 6509's with Sup1A and MSFC2/PFC. We have dual 16GBIC line
> cards
> > (32 in total) and we are using ALL of them. If we have the 32gbps
> backplane,
> > and our 32 sockets maxed out (this isnt including the 4 x 48port
ethernet
> > line cards we have) then we would be overloading our backplane yes?




Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=42965&t=42877
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]

Reply via email to