12-29-2010 06:42 PM
While researching comparisons between Cisco and Juniper and specifically looking from ways Cisco is better than Juniper (Haven't found any valid claims, yet) I ran across some questionable videos on YouTube.
A few things I noticed about the videos is that the user that posted them has only posted one (1) video per username and that the usernames are generally based on fairy-tales (maybe a hint as to what they are telling). Some seem to be tied to a test Cisco paid for as well.
The point of this post is to hear what people think, backed by technical references (not flaming). I don't believe the claims being made, but I myself haven't had the opportunity to perform tests on the mentioned in the videos.
What do you think. I would also like to hear what a Juniper representative has to say about it and whether any claims are true.
Juniper MX960 3D Falls Short..Yet Again!
JUNOS: Reality vs Myth!!!!!
Reality of Juniper
MX Routers - Demand More but Get Less? [Same as "Reality of Juniper"]
Will the Cisco ASR 9000 kill the Juniper MX960?
Miercom Test Results
12-29-2010 09:36 PM
Someone (I wonder who...) spent an awful lot of money to produce some of those. They are very dated at this point -- for example the claims about IPv6 support on SRX -- and just plain wrong on others (the exact same binary runs on M-Series, T-Series, etc). And the MX 3D (based on the Trio chipset) didn't even exist when those were made http://www.juniper.net/us/en/products-services/routing/mx-series/
An interesting question about the One Junos strategy that the posts fail to address is: What is the important part of that strategy? Is it that the exact same binary run on every single device? Probably not. We think the important part is that total cost of ownership is reduced because the user-experience (from the CLI, to diagnostics, management, output formats, etc) is as similar as possible given the intrinsic differences in the platforms.
Being able to flip your KVM across consoles attached to a router, a switch, and a firewall, and pretty much already knowing how to operate all 3 (by knowing one) - is IMHO pretty cool.
Sponsored performance tests?...meh