VMotion demo is a sure WOW!
Rapid deployment with VC Templates and or Golden Master.
Portability between dis-similar hardware.
Utilization of Physical Servers
1 person found this helpful
Get the demo and show off Vmotion. It is a big selling point to the guys up stairs.
Another cool factor is once you create your golden images you can show how you can have a server up in minutes not days.
Also figure out what you would save in hardware costs alone. Say you are going to have 100 servers that you will virtualize. Take what it would cost to buy 100 servers and how much it would cost to get say five 4-way servers to run the 100 servers you virtualize.
This can get the thread started and I am sure you will get plenty of ideas here.
1 person found this helpful
On a slightly different tack, mention the potentially large saving in licensing costs when you consider Microsoft's new licensing policy for virtual machines. For example, if you'd like to implement SQL Server and license it on a per-CPU basis, you can save a fortune by virtualising. Also, Windows Server R2 Enterprise has new favourable licensing terms too.
Probably one for the bean-counters in your Company, rather than this guy though!
If you really want to give a great demo, contact a good consultant and have an expert set it up and show it off. I'd suggest going with your preferred HW vendor (assuming they're a VMware reseller) or find a good local consultant who has a solid reputation.
While you can, undoubtedly, do a demo of most of the functionality yourself, someone who does this on a regular basis will be able to talk to both the technical and the business benefits of virtualizing. Even though the technology is "way cool", if there's not a solid business case put forth, you'll get no $$$ to implement with.
I'd definately go for a demo guy, our supplier is IBM, but a rep has as much weight as dust here, they're a no-no.
I can't show off vmotion because it requires ESX correct? I'll be using the beta of virtual server, the vmtn subscription probably won't be purchased in time(there's nothing slower than purchasing through the company).
As to the boss, he generally gets what he wants--so if I sell him on it, I've sold the company. He's not even CIO but CIO listends to him. I picked up a book, I'll do more indepth reading, I hope to be proficient by the time the demo is live. I bought an 8 way old proliant for the demo out of my own pocket, not too expensive, but I can put 16 gigs on it, and load it with 8 700xeons. I know it's not very powerful, but you can put a lot more on it. For a demo it should be ideal.
I really want to get vmware in, and I've got a decent shot. I'm moving over all my servers this weekend...that's my addition to the demo. I can tell him that my servers have been up for a month on it by the time the demo runs.
The paying for it's not the problem, this is a company that spends millions on IT, it's the selling it to them that needs to happen.
Do you think I should go ahead and buy the vmtn out of my pocket as well?I've got the 8 way in process, and I'm sure that's more an ESX than Virtual server item....It's just that I've got a lot invested in hardware at my company already, and another 3 or 400 hundred bucks is starting to get itchy if you know what I mean. My test environment is mostly running on my hardware already.
You have to understand, we just got permission in the last year for projects that had been running for a couple years under the old administration. But it's saved hundreds of man years, taking day long hand installs which were the requirement under the old regime(for 5000 users!) to 20 minutes. We have automated everything to the point that it's as smooth as glass. And it's all on homegrown apps, scripts, and models. We now have standing, because the new boss came in over a year ago and we had something that was faster than altiris and norton. We were doing packaging and everything else just as well as the company he left. And we had no training, no incentive(unless you count negative), and no funds. We've got it all running on old HP netservers.
I want this to succeed, it means more standing for our little two man crew, it means cost savings to the company. I can't afford to fail and lose the hard earned position, but neither can I afford to rest on my laurels. We are expected to turn out "miracles". Even the boss expects it. He gets us what we ask for, but he has to go through the same red tape we do, and it takes a while.
I'd say be very careful investing your own $$$ for company purposes. What happens in this scenario:
\- You migrate your servers over to VMware running on the 8-way servery that you purchased, that's not under warranty, that's not a company resource
\- Three weeks later (one week before your demo), your server dies, hard
\- All the servers that you've migrated to VMware are now dead, with no way to get them back online quickly
Sounds like a great way to ensure that VMware never gets in.
Also, you mentioned that your company spends millions on IT - sounds like they need to spend some more on a decent test environment. You are expected to pull off miracles with no support and you keep doing it (at your own expense!). As long as you keep doing it, they'll keep expecting it.
Just my $0.02 worth
There's several problems.
The original project was black bag, these are recovered servers that were far out of warranty in the first place. This is my baby, and mine alone. It runs only because of my hardware donations, my configurations(and it's been up for months without a blink), using only open source software. Nothing in my projects has company approval or backing. The boss, after seeing this has slated two brand new servers but that takes a while to come through. I have what turns out to be, the power of god over my projects, because I was never assigned projects in the first place. It's sort of a catch 22.
And yes, they do need to spend some money, but our new guy has not been here that long, and testing is something the old regime did not really do, whereas I maintain an extended test phase on everything. I can, if I so desire pull the servers completely off the network and no one can say a thing, because everything on it, from the images to the patches are mine. So I have flexibility here. I'm planning on running this on my backup network anyway, so it won't affect the production network, which is a couple stages behind the test environment anyway. We are divided--we have a set of machines on the live network, and we have a set on a test network. We still use the test network for building the images and testing all of our software before we move it to the network.
You have to realize, we're just pc techs. We're now being pulled to be put over a new team called development and deployment, but that's because we built it. I realize you are looking out for my best interests here, but trust me, I'm the most paranoid and safety concious person on our network. I test a minimum of 2 generations ahead of something moving to the live network.
What problems are your boss(es) trying to solve, that could be solved by implementing a vmware environment?
Find out the answer to the above question, and apply the techonolgies to solving these problems better/faster/cheaper.
Message was edited by:
I wouldn't buy anything yet. I would suggest getting demo licenses from VMware. The software is fully functional so no real down side. If you get the ok to buy just update the keys.
I just downloaded the demo, I'll see what I can do with it.
Okay, the ftp, web, and samba servers are migrated to vm's. So far no performance hit, and plenty of cpu left in just the little duallys that I'm testing. I don't plan on moving the main image server to vm's just because it is VERY file transfer heavy 60 gigs in 20 minutes is not uncommon.
The demo is perfect, it's licensed for two ESX servers, with vmotion. It's like they knew my need! Thanks guys. Now if I can just become an expert in 30 days.
Indeed - the evaluation VIN is just what you need.
Expert in 30-Days, well good luck with that.
We will be here to help you through!
Good Luck and Happy Virtualizations!
Yeah, expert in that timeframe is a little hopeful, but I've picked up harder stuff in less time--so, we'll see how it goes.
I don't think I'll have any problems on the technical side, I'm also going to read all the whitepapers I can find(already 30+) so I can answer the business side questions as well.
I could demo what I've got right now and impress him I think. I'll keep the thread updated with any news.