Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

backplane/switch fabric speed

Hi,

 

I've been looking for the backplane information of WS-C2960-48PST-L and WS-C2960S-48LPS-L. Nothing is indicated in the datasheet http://www.cisco.com/c/en/us/products/collateral/switches/catalyst-2960-series-switches/product_data_sheet0900aecd80322c0c.html. All I saw was switching bandwidth.

 

According to http://www.techrepublic.com/article/decision-support-choose-the-right-switch-for-your-network/ switching bandwidth is different from backplane/switch fabric speed.

 

Thanks.
 

4 REPLIES
Hall of Fame Super Gold

The only time you'll see

The only time you'll see "backplane" being mentioned is when you're talking about chassis-based switches like the 4500, 6500/7600 and 6800.  And the backplane speed is dependent on the supervisor card and chassis being used.

 

With 2960S, the stacking speed is set for 10 Gbps (full duplex).

Hello leoWould this

Hello leo - Joesph

Would this calculation you have stated apply to all platforms? - It would be nice to have something like this to use as a guide, I have never really thought of asking this question on these forums, so I might as well.

I usually take the port speed x 2( for return traffic) and x2 again for backplane ( if applicable)
example:
P1 -P2 - 100mbs x2 = 200mbs
P1-backplane = 100mbs
P2 backplane = 100mbs
total  = 400mbs

would this be correct?

res

Paul

Please don't forget to rate any posts that have been helpful. Thanks.
Super Bronze

DisclaimerThe Author of this

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

I believe it applies to all platforms with full duplex ports.

Forwarding PPS is just the aggregate of all ports' nominal bandwidth.  This because the forwarding decision is only made once, for ingress to egress.

Backplane/fabric bandwidth, is all ports' nominal bandwidth, x2.  This because we're counting the ingress and egress bandwidth.

Super Bronze

DisclaimerThe Author of this

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

In your first reference, see first two entries of table 4, "Forwarding bandwidth" and "Switching bandwidth".  The latter is typically what is quoted for fabric or backplane bandwidth.  (NB: the numbers for the 2960G are odd.)

Yes, switching bandwidth is different from backplane/fabric bandwidth.  The former depends on the PPS rate for a particular frame/packet size, the latter is independent of frame/packet size.  The PPS rates, for different 2960 models, is listed lower in the same table.

Most newer Enterprise class switches are wire-speed/line-rate non-blocking, older or non-Enterprise switches often are not.

To calculate non-blocking backplane/fabric bandwidth, take all your port bandwidths, and double.

To calculate wire-speed/line-rate for minimum size Ethernet packets, take port's Gbps times 1.488 Mpps.

In the table we see:

Switching bandwidth*

176 Gbps

32Gbps

32 Gbps (2960G)

Cisco Catalyst 2960G-8TC-L

11.9 mpps

Cisco Catalyst 2960G-24TC-L

35.7 mpps

Cisco Catalyst 2960G-48TC-L

39.0 mpps

8 Gig ports would need 16 Gbps fabric and about 11.9 Mpps.  (Both met.)

24 Gig ports would need 48 Gbps fabric and about 35.7 Mpps (Fabric bandwidth insufficient.)

48 Gig ports would need 96 Gbps fabric and about 71.4 Mpps (Neither sufficient.)

BTW, real production traffic rarely is 100% across all ports and/or all minimum size packets, so wire-speed/line-rate, non-blocking, is often not actually required.

614
Views
5
Helpful
4
Replies
CreatePlease login to create content