Now LinkedIn and Microsoft are getting into the cloud computing game (by cloud computing I mean providing a platform for you to run your own application without a data center of some type). LinkedIn, Microsoft (Azure), SalesForce.com, Amazon, Google, you name it. Cloud computing is the new black.
It is a great concept. The costs are less and you can spread the cost of your hardward across a VERRRRRRRY long period of time, which is great in times when you can't get large amounts of credit (like 2008).
However, there's a huge difference in the way some of these vendors are implementing their cloud solution: some are locking you into their platform, some are trying not to.
By "trying not to lock you in" I mean that Amazon allows you to run a generic Linux image, while all of the others require you to write your software against their proprietary API's and platforms. You can take a Linux image which is running in your own data center in the next room, and run it on Amazon EC2. Perhaps in the future other vendors will adopt the EC2 image format which will make these images even more portable.
If you write an application on any of the other platforms, you're stuck on that platform.
However, there's a caveat here: even Amazon gives you other services to use which help you adapt your application to the cloud, such as a message queue service, a storage service, and so on. Use those, and you're stuck.
So today you're really deciding on your LEVEL of vendor lock-in. Least (Amazon) to most (everyone else).
To change vendors, you're looking at at a lot of development work. Choose your vendor wisely.
If I were choosing a vendor today (which I'm not), I would base the decision on platform maturity and vendor stability. You can't get much more mature than Linux at your core, so Amazon wins that one in a landslide. SalesForce.com is pretty mature as well, but their configurabliity and flexibility sucks. By vendor stability, I mean how likely the vendor is to be around twenty years from now. You don't want to put your revenue stream in jeopordy or be forced into a platform conversion.
In terms of vendor stability and liklihood to be around in 20 years, here's my ranked list:
- Microsoft (billions in cash, proven revenue streams)
- Amazon.com (lots of cash, proven revenue streams)
- Google (lots of cash, proven revenue streams)
- SalesForce.com (proven revenue stream)
- LinkedIn (still running on borrowed money, use at your own risk)
It'll be interesting to see how Microsoft's offering develops over time. For large companies looking to run in the cloud, I think the choice will boil down to Microsoft and Amazon for the forseeable future.
