By Ben Stephenson, Journey to the Cloud
As you probably know, Jason Segal and Cameron Diaz recently came out with a new movie called “Sex Tape.” In the movie, Segel and Diaz play a couple who decide to make an adult home movie, and it accidently gets released online.
I saw the trailer a couple of weeks ago and one clip from it grabbed my attention. After Jason and Cameron realize the tape has been released, they start to panic. The clip shows the two driving in a frenzy talking about how the tape got released (you can watch the clip here).
Cameron: “How do you forget to delete your sex tape?”
Jason: “It kept slipping my mind, and then the next thing I knew it went up…it went up to the cloud.”
Cameron: “And you can’t get it down from the cloud?”
Jason: “Nobody understands the cloud. It’s a mystery”
This got me thinking about some companies’ abruptness to go to the cloud without fully understanding the consequences. IT decision makers are under increasing pressure from CEOs & CFOs to utilize the benefits of cloud. If your IT department doesn’t have a well thought out strategy, however, chances are you’re not going to be successful. You don’t want to move to the cloud for the sake of moving to the cloud and then be in a frenzy if something goes wrong. I spoke with one of our bloggers, and cloud expert, John Dixon to get his take on what organizations need to consider before deciding to move to the cloud. Here’s what John had to say:
Not everything is a great fit for cloud
Have some high-end CAD desktops? Not a great fit for cloud (at least right now). Just setup a new ERP system on new hardware? Not a great fit for cloud at the moment. Testing some new functionality on your website or a new brand entirely? Now we’re talking — setup the infrastructure, test the market, scale up if needed, take everything down if needed.
Benefits of cloud are potentially huge, but hard to measure
Back in the datacenter consolidation days, the ROI of doing a virtualization project was dead easy. Consolidate at least 10 physical servers down to 1, and you have instant savings in power, cooling, floor space, administrative burden, etc. Some nice features came from having virtualized infrastructure, like the ability to provision servers from templates, easier DR, etc. Infrastructure teams were the big winners. In short, you could easily calculate the financial benefit of virtualizing servers. With cloud, this is not the case. If you look closely, there is limited benefit in just “moving” servers to the cloud. In fact, it may cost you more to host servers in the cloud than it does in your physical datacenter. However, IaaS clouds allow you to do things that you couldn’t do on your own. A pharmaceutical company can “rent” 10,000 servers to run a 2 day simulation; an online school can build infrastructure in a cloud datacenter in Australia to test a new market in APAC; a startup can use cloud to run ALL of its technical services (with no capital investment). In short, you’ve got to understand your existing costs, your use cases, and the benefits you are seeking. Jumping in to cloud without this mindset may comprise the benefits of doing so.
Portability or optimization? Especially in Amazon Web Services
As of now, you can’t have both. Choose wisely. Optimizing your application for cloud (for example, in AWS) by making use of RDS, SNS, SQS, Cloudformation, Autoscaling, Cloudwatch, etc. can deliver some amazing benefits in terms of scalability, supportability, and reliability. However, doing so destroys portability and any hope of brining that application back in house. On the flip side, VMware vCHS offers awesome portability, but the opportunities for optimization are fewer.
So, lessons for the kids out there? Don’t upload certain home videos online and have a well thought out cloud strategy before jumping into anything you may regret.
Oh, and Jason…borderline offensive when you say “no one gets the cloud”…because GreenPages gets it…
If you would like to hear more from John, download his ebook on the evolution of the corporate IT department.