This week I want to take a look at one of the buzz-phrases that’s sometimes bandied about by computer types - utility computing. It’s an idea that’s generated a lot of interest in the past year or so, and I can understand why. The Wikipedia has an item on utility computing, but it’s a bit heavy on tech-speak.
So what is utility computing, and can your company use it? Or do you even need it?
Utility computing is usually done by a third party provider (like a hosting company). So when a potential client comes along 2 things happen: a) the sales team at the hosting company can boast about how much power their computers have and b) the sales team will tell you how they have a “flexible and extensible infrastructure”. Which is true, but it probably doesn’t matter to you if you’re a small or medium business.
So you sign up with the hosting company. Now let’s say for argument’s sake that you keep a corporate blog (like this one) and you post a really funny video that turns into the latest and greatest Internet funny thing. It’s viewed by millions of people worldwide. This is where the utility computing part comes in.
If you had your website at your own office, then your computers probably wouldn’t be able to handle the spike in traffic that comes with getting tens of thousands of hits per day. But in a hosting environment they can handle the spike. So maybe the hosting company will move you to a different part of their office, or they add an extra network cable to the computer your stuff is on. But the important part os that they just make sure that it keeps working.
This is one of the main benefits of keeping your web stuff with a third party hosting company - they have the computing resources. And if you start to use those added resources, they just add it to your bill. Just like the power or gas company will boost your bill if you leave the heat on all night. In essence, computer power becomes like another utility, billed and paid for on a usage basis, the same as gas or electricity.
The disadvantage of this setup is that most hosting companies want you to sign on to a package. This package gives you a certain amount of resources, such as how much network traffic you can have every month. And if you go over that allocation, you start paying. Sometimes very heavily.
So now that you understand it, how does it affect you? Well, for most companies there’s a peace of mind factor, of knowing that they don’t have to worry about power outages, or of knowing that someone else will look after the monitoring and troubleshooting. On the downside though is the cost and lack of control. Over-use gets expensive, and you’re always subject the whims of what someone else is doing. If they declare a maintenance window (to do computer maintenance work) just as you’re launching your international marketing and media campaign, then it’s tough beans for you.
There are alternatives to hosting companies of course, and in my next article I’ll look at commodity computing, and how it can help you address your own utility computing needs.
And as always, a good technical support provider can help you decide which path is the best for you and your business .
Technology In Check :: Commodity Computing | 21-Feb-08 at 9:37 am | Permalink
[…] As promised, I want to talk about commodity computing this week. What it is, how it affects you, and how it relates to utility computing. […]