New technology is fun, productive and always gets a lot of attention. The newest shiny objects and magic bullets attract us. But with every shift in technology there are things we lose, as well.
One of the big technology shifts that’s been under way for the past few years is "the cloud." The idea behind cloud computing is basically this: Instead of having a large pile of processing power and hard drive space (what information technology wonks like to call "compute assets") of your own, you lease these things from a vendor.
The big advantage of cloud computing is that it can reduce the amount of money spent through the IT department: The lump cost of hardware gets spread over time in the lease of the cloud resources.
Cloud computing is a technology that reduces costs, not a technology that increases opportunity (someone will raise the issue of mobile access of cloud resources as an example of increasing opportunity, but that’s all about mobile access — not cloud computing — increasing the opportunity).
But we also lose things. Change often works that way: Get one thing and lose some others. I don’t want anyone to misconstrue what I’m going to lay out in this column as "Gahlord hates cloud computing."
Nothing could be further from the truth. I just want to make sure we all know that there are trade-offs being made as we switch from local compute assets to remote compute assets.
Constrained by a third party
The first and probably most important thing that’s lost when an organization goes whole hog for cloud computing is local control of technology. Sometimes this is a good thing, like when your IT department is also known as the "sales prevention unit."
But sometimes it isn’t a good thing. Once our data and applications rely on a third party in order to be useful, then we can innovate or grow only at the same pace as that third party.
If the cloud vendor needs to pursue a certain development path in order to make its business model work, and if that development path doesn’t support your business needs, then there’s heartache in the future.
This is important to watch for as data and information become competitive assets. If our ability to store and process data is constrained by a third party — which is by definition beyond our control — then we are exposed to risk.
Constrained by connectivity
The hard drives and servers that make up cloud-computing centers are typically not directly connected to our own organizations. This means that our access to those assets can be interrupted by something going haywire in between.
The connectivity issue can be looked at as similar to putting all the eggs in one basket. If the primary business-computing activities are taking place somewhere else, and your access to that computing center is interrupted, then your primary business activities are also interrupted.
One of the common concerns about cloud computing has to do with security risk. If your business data resides on a server that is outside your control then you may have less control of the data itself.
The additional security risks associated with storing your data elsewhere range from accidental fire, earthquake, etc., at the cloud facility to physical theft of hard drives.
These calamities could, of course, occur at your own facility, but you have less control over the various factors involved when outsourcing to a cloud facility.
Some software services that operate cloud-computing facilities do so in order to increase "switching costs" among their customers. If all of your data is in a special format or is exceedingly large, then switching from one software service to another is difficult, for example.
History has more than one example of an organization using this lock-in strategy to stymie innovation and competition. If your organization requires innovation in order to compete in the marketplace, lock-in can tie you to an anchor.
But it isn’t all bad
Again, I don’t want anyone to think that cloud computing is a bad idea. For organizations where the IT budget or experience is thin, cloud computing offers the ability to get stuff done at a lower cost.
If your organization isn’t adept at handling the risks outlined above, then cloud computing is unlikely to make it any worse.
When examining what functions or data types are good candidates for switching over to the cloud, remember to review what you’re losing.
In particular, consider how your organization will handle additional constraints. Clearly define how much innovation and scalability you will require.
Clearly define how much lock-in you can tolerate (and have a plan in case things go sour).