Server wannabe Cisco Systems is trying to drive server and storage network convergence and benefit from the inevitable shift to virtualised servers, storage and networks. And the best way to sell a new architecture is to use it. It was this shift towards virtualisation that compelled Cisco to take on its partners and enter the …
With 10Gig networking, and vmware and RedHat, it should take less than 2 minutes.
Hell, with 1Gig, ESXi and CentOS, i can do it in 70 minutes.
Re: two weeks?
2-70 minutes? Wow, that's quick.
How much of that is spent determining the performance requirements for the system/application/network?
What's that you say? You only deply servers and worry about things like the end-user experience and support latter in an adhoc fashion. That explains it....
re: That explains it...
I was looking at the statement from the other side. That being: here's an organization that uses structured vm's. Sureley they have an archive of pre-built/tweeked server images with descriptions like:
1) vanilla server
2) low-volume web server
3) medium-volume web server
4) high-volume web server
Personally, I would expect that the 'what server should we deploy' part of the process would take the same amount of time regardless of the final decision.
I've been network support for a world-wide supplier, and I know that we had (I'm now with a different firm) contracts with customers that specified 24 hour delivery for specified products. So if Cisco (not a small business) orders a server from thier supplier (?Cisco), I would expect that the server would be delivered the next day. In the world of virtual machines, I see no reason why the time from order-the-server to server-is-installed-and-ready-to-use should be more than 10 minutes. If 90% of the time is spent determining 'which' server should be deployed, then I can understand it; however, the article lumped all of the separate, and some being very application/user base specific, tasks into one lump time.
I do appreciate; howver, that you left the pin in the grenade.
Marketing disguised as achievement
These stories are borderline press releases.
We are actually getting rid of networking by virtualizing and using hypersockets in the box.
So while Oracle/Cisco/HP wants everyone to go to small SMP with lots of software and networking we are going to a few large SMP's and saving $10 of software and personnel costs for every extra dollar we spend on better hardware.
Why no mention of Dell?
This is a borderline Cisco press release, with nods to Oracle/Sun, IBM and HP, but no discussion or comparison of Cisco's UCS to Dell's VIS/AIM, which is the sophisticated, less-complex and leading solution.
By comparison, AIM works not only on Dell gear, but also HP and IBM's, virtualized and non-virtualized, with just about anyone's networking (Juniper, Brocade, Cisco, etc.).
Cisco's by comparison only works on Cisco server hardware with Cisco Nexus networking.
Dell is after all the x86 market leader in the US, #2 in the world, whereas Cisco has significantly <1% market share.
Yes, I work for Dell.