7 Cloud Computing Myths Busted
Amazon, Google, Microsoft and others are investing aggressively in the cloud, even as critics point to security, reliability, and compatibility issues. We cut through the fog.
What is it about "the cloud" that has people, well, getting their heads up in the clouds over it? Almost no other IT innovation in recent memory has engendered this much enthusiasm -- and furor, and confusion, and outright misunderstanding.
The cloud isn't exclusively a cure-all or a calamity in progress; neither is it a savior or sinner. It's a new tool for solving emergent problems, and like every new hammer in someone's hands it can make everything look like a nail.
More Cloud Insights
- IDC Analysis: Secure, Compliant Collaboration in the Cloud
- Combining Cloud-Based DDoS Protection and Managed DNS Services to Thwart Large Attacks.
In this piece we'll examine many of the current myths -- good, bad, and bogus -- about cloud computing. Many are borne by simple ignorance or inexperience. Others are legitimate criticisms in the guise of gripes. And some are entirely too on target, and need to be nipped in the bud by prospective cloud-creators before they get bitten by them.
1. Compatibility Issues
Myth: Cloud computing is too proprietary.
At present, no two clouds are alike -- both in nature and in IT. Amazon's cloud platform is nothing like Google's, which is nothing like Microsoft's, which is nothing like and you can insert the name of any other up-and-coming cloud provider here.
And yet "proprietary" has not proved to mean "useless" -- not by a long shot.
Think back to the early days of the personal computer. The first wave of PCs were all from different makers, used different hardware, and weren't remotely cross-compatible. Programs written for the Apple II weren't assumed to have any interchangeability with the Atari, the Amiga, or even the IBM PC itself.
What few common platforms that existed -- e.g., CP/M -- were largely for the sake of porting and running existing applications to those platforms, rather than for creating a crossbar of compatibility among them. None of this stopped a remarkable amount of development from taking place -- and the various platforms were able to compete heavily based on their differences.
Granted, the situation today is totally unlike that. People expect a great deal more cross-compatibility as a matter of course -- between devices, between applications, between platforms and environments. What's most proprietary about the platforms isn't so much the way they work on the inside as the fact that talking to each cloud, getting data into and out of each cloud, and managing functionality within each cloud are all done differently.
The proprietary nature of the first wave of cloud computing platforms is, for lack of a better way to put it, a necessary evil. And maybe even not all that evil in the first place, when it grants you access to platforms like Linux (Amazon.com) and languages like Python (Google), which on their own terms are as open as they get. Things could be made less proprietary outside clouds and among clouds, although odds are the standards that will exist between clouds will develop more as a consequence of what people are actually using (e.g., EC2) rather than something drafted in the abstract.