Alright, so we're putting together a design for a smaller, green field datacenter and I wanted to reach out to this community for feedback because I've gotten mixed messages about the 4900M platform.
A pair of 4900Ms could work as distribution (we're ok density-wise) for our datacenter but we've had a couple engineers say that it might struggle under load if there are ACLs, VACLS, or PBR, or anything that might require processor intervention. Most of the engineers give the 4900M a thumbs up, with an understanding of the limitations of the box (density, flexibility, service modules, etc.)
We're not moving a huge amount of traffic through these devices and probably won't be for some time, but has anyone seen the 4900M struggle in the distribution role processor-wise?
You didn't mention - are you using the 4900M's as strictly L2 switches? If so, you shouldn't have any issues pumping them at or near wire speed across the board. Even with L3 services running , it should be fine for most purposes.
These will be L3 switches used to aggregate L2 DC access switches (low density). I totally agree with all the comments as far as horsepower - even if there were L3 convergence events, I can't imagine the switches grinding to a halt due to lack of processor resources - especially in a network of less than 500 users... Of course, it's almost impossible to know for sure, except via thorough load testing, etc.
The Achilles' heel of most L3 devices would be traffic that is not handled by hardware, i.e. forwarding drops back to main processor. It can be difficult to ferret out exactly all the conditions or limitations, via documentation, that would trigger such situations. I.e., without load testing using your expected configurations, with expected traffic, might be the only "easy" way to know.
From a raw performance perspective, on paper, the 4900M is very impressive. Basically, it offers half the fabric bandwidth of a sup720 and enough Mpps to guarantee wire-rate for the fabric's bandwidth (the latter difficult to accomplish on a 6500).
For only 500 users, unless you're aggregating 10 gig uplinks, the 4900M might be more than you need. If you are aggregating many 10 gig uplinks, the 4900M might be an ideal choice.
Other than performance, the choice between a 4500/4900 series or 6500 series switch, might take into consideration other features of the two platforms.
Another "inexpensive" switch you might consider, if not using many 10 uplinks, would be the 3750-E, stacked.
Depending on your need for 10G ports, Leo's suggestion could be better than mine suggestion of the 3750-E for a less expensive switch than the 4900M. Downside includes you can't stack it, but also true of the 4900M, and this particular model's fabric and PPS, are only about 50% of ports' bandwidth, but 4900M is also oversubscribed if you go more than 16 10G ports.
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 ...