Line card utilization - 4506

bberry · ‎09-09-2008

I have a 4506 with a Sup IV(4215), a 24-port 10/100/1000 card

(WS-X4424-GB-RJ45) and two 100BaseX (WS-X4306-GB) line cards. I have been told the 4424

line card is designed for 4 to 1 over subscription and not true gig on all 24 ports. How

can I tell the utilization in regards to this over subscription? Can I look at the 4 to 1

group or do I look at the individual ports in groups of four? Is there a way I can look at

the load on the back plane? The resaon for the question is that I almost have all 24 ports

connected to things like my routers, PIX, VPN concentrator, etc and are about to add a

SAN. I want to make sure I still have enough available bandwisth to support and see if I

need to re-arrange the way things are connected so as to not over subscribe anything.

Prashanth Krishnappa · ‎09-10-2008

http://www.cisco.com/en/US/docs/switches/lan/catalyst4500/hardware/module/guide/03instal.html#wp1037598

To check to see if the ports are being oversubscribed, you can do show interface and see if you are seeing input and output drops. You can also do show interface counters detail and see if flow-control is being asserted.

The oversusbscribed cards are designed for Gigabit to desktop deployments and not for interconnecting networking equipment. If I were you I would connect, your SAN device to 4306 which is non-blocking card.

Joseph W. Doherty · ‎09-10-2008

The non "-E" 4500 chassis is limited to 6 GB per slot. If your line card's ports provide more than 6 GB, you're oversubscribed. How line card ports share the 6 GB, depends on the card. You can often control individual bandwidth subscription by how you allocate line card ports, including leaving some unused.

See table 2, and oversubscription footnote, in: http://www.cisco.com/en/US/prod/collateral/modules/ps2710/ps5494/product_data_sheet0900aecd802109ea_ps4324_Products_Data_Sheet.html

For something like SAN, if it provides a 10 gig connection, you might consider upgrading your supervisor to a sup V-10GE.