So I read on TechRepublic how the “cloud” is going to take away jobs but open others. This is full of bull corn and probably just a bunch of hype that TechRepublic uses to get people interested in reading their articles. Kudos on the tactic; however, I find it despicable because you’ll have up-and-comers into the IT field who will get discouraged. I was almost one of them in the early 2000’s…
Lets take a moment and think about what the “cloud” really is. The “cloud” is nothing new and is just some marketing jargon to generate profits to the hosted industry. Perhaps the one difference is the abstraction of finite system administration details being replaced with easy-to-use GUIs that seemingly remove the need for education of system administration. This is a major misunderstanding but we’re not going to get into that debate here.
Everything seems to be labeled “cloud” these days. In fact, hosted Microsoft Exchange is considered “cloud” now. Lets just wake up and face the facts that all that Hosted Exchange is just another server hosting several domains, just like everyone else has been doing for now almost 20+ years. Hosted applications aren’t a new thing either. In fact, I remember back in 2003 I used to say that platform specific implementations were stupid because it “vendor locked” people into specific platforms and made administration a nightmare. I was also pro for building everything as a standards based web application with a back end database of the organizations choice. It would offer the greatest flexibility because all you need is a server than can run HTML/Java/JScript/PHP and then pick a database of your choice to run the organization. Sure, the vendor could offer support on a few databases; however, extend the choice to the customer and give them the rope to hang themselves. Or, quite simply, hire better programmers to build the code to operate on different databases since they are all SQL based and the more finite details are implementation specific and you can charge for support based on complexity. I also said that with a setup like that an organizations infrastructure could be shrunken and remote connectivity could be simplified.
So, now it looks like that is the path that most are taking and I really wish I would have got off my arse and built something and stuck to it because I’d be rich now! However, I don’t see “the cloud” as a viable alternative to infrastructure in house. You’re data and applications are no longer governed by you entirely and your policies are restricted to the policies that they implement for you. For some small organizations this is not an issue because there isn’t a lot of IT expertise to handle complex configurations; however, extremely simple designs tend to lack redundancy, security and robustness.
Things in the cloud seem easier, yes. However, take into consideration your SLA and how much control you lose over your infrastructure. Thanks to VMware we’re able to consolidate many isolated hardware machines into one machine. Because software is no longer evolving as fast as hardware you don’t need the biggest and baddest server on the planet to run your infrastructure on VMware. In reality you wouldn’t even need iSCSI or super fast disks. Just get some redundancy built into a modest priced server with enough RAM and HDD space with enough CPU power and you’ll be able to run numerous machines on that VMware device effortlessly. Very little infrastructure costs and space/cooling/power savings all rolled into one. The investment will pay for itself within a year versus a monthly subscription cost to a cloud provider.