We just built out our new data center and used the 4948's for server aggregation. They are solid, dependable and fast switches. We are currently using 12 of them.
We designed the switches to use 10G, Layer 3 uplinks to a dual core setup. The cores are 6509's with single sup 720's each.
We run the 4948's in an HSRP pair between two cabinets. Each server has dual NICs, with one NIC attached to one switch, and the other NIC to the other switch. The 4948's have a 10 gig link between them and a 10 gig uplink to one of the cores. This will result is a highly fault tolerant server aggregation layer. Any single device or link from the server to the core can fail, and you will not lose connectivity.
I do not see a creative way, even with redundant power supplies, to make a single switch any more fault tolerant. If that switch fails, then you are down.
You can look at using 2 switches to serve 4 cabinets (assuming 12 servers per rack). There will be some long patch cables from servers to switch, but in practice these will be less than 50 feet. I would recomend using a lesser switch for ILO (say a 2950 or what ever you have on hand - we use 3550-EMIs). This will off load the port count to a lower capacity and speed switch. Think of the vlaue of a 4948 port compared to the value of a 3550 port. Roughly 1/4 the cost per port.
Our business is 24/7/365. Each hour of down time can cost us from 100K to 500K in lost revenues, blown SLAs, product spoilage, trucker delays, overtime, etc, etc, etc. It's expensive not to be able to move Cherrios! So when you try to sell this type of fault tolerance to a business, you gotta ask what's it worth. I can buy 4948-10G's for about 12K each. Cheap insurance if you ask me...
Question We run asr9001 with XR 6.1.3, and we have a very long delay to
login w/ SSH 1 or 2 to the device compare to IOS device. After
investigation, the there is 1s delay between the client KEXDH_INIT and
the server (XR) KEXDH_REPLY. After debug ssh serv...
Introduction The purpose of this document is to demonstrate the Open
Shortest Path First (OSPF) behavior when the V-bit (Virtual-link bit) is
present in a non-backbone area. The V-bit is signaled in Type-1 LSA only
if the router is the endpoint of one or ...
Hi, I am seeing quite a few issues with patch install and wanted to
share my experience and workaround to this. Login to admin via CLI, then
access root with the “shell” command Issue “df –h” and you’ll probably
see the following directory full or nearly ...