Public Cloud Computing Shows Its Limits
Posted by Michael Bullock on Tue, Dec 15, 2009
Welcome to the hybrid world.
I've hardly been alone in my skepticism about cloud computing and my belief that public cloud computing for the enterprise is not ready for prime time. But I’m sensing a general global warming on the subject. And if applied properly, I'm one of them.
Here’s what I said back in January 2009 in my top 5 predictions for 2009
(Hype Overload): "There will be a well publicized cloud computing disaster in 2009 – probably linked to security, reliability or service restoration delays."
Was I right? You be the judge.
• Google alone had 7 significant outages. Google outages damage cloud credibility (PC World).
• Amazon had at least 3 outages including taking down the Amazon retail site. Outage for Amazon web service (Data Center Knowledge).
• A Salesforce.com outage lasting nearly a day cut off access to critical business data for many of the company's customers on Tuesday in what appears to be Salesforce's most severe service disruption to date. Salesforce outage angers customers (cnet news).
• And of course security has been a concern. See The Cloud Security Survival Guide.
So there were over a dozen significant outages at Google, Amazon and Salesforce alone. The only surprising thing here is the widespread numbness to these failures that helps keep the hype coming.
And why do people gloss over these disasters? It's because, in truth, very little serious, mission critical, life-threatening, or deep financial work is going on in public clouds. So as long as nobody dies, loses millions, or breaches thousands or millions of personal records, these failures may be acceptable. The key is using the public option for a suitable purpose.
Some things will continue to be best left outside the cloud. The security issues alone are huge and may block deployment: PCI (payment card industry), HIPAA, information privacy laws like 201 CMR 17.00 in Massachusetts (and similar laws in other states) make it almost impossible to insure adequate security in the cloud.
These security concerns mean that it will be a long, long time before you see companies making a full transition to public cloud services. However, in the near term, these companies will continue to leverage stand-alone cloud applications where it makes sense (like Salesforce.com). They will continue to develop their internal virtualized infrastructure to look more and more like a private cloud. But this approach is evolutionary, not revolutionary.
Some vendors seem to get this evolutionary approach. They understand that the old must work with the new. They understand that applications and infrastructure are equally important to adopting cloud computing in the enterprise. They also understand that with the cloud’s increased infrastructure flexibility comes complexity that must be managed and monitored in order to assure the highest availability of applications to the users. After all, that's the only true measure of success.
So the companies to watch in 2010 that can enable these hybrid cloud/legacy applications include:
• Cisco, EMC and VMware which, through a joint venture called Acadia, have partnered to accelerate internal cloud development with unified management of the data center.
• Nimsoft, which has one of the most advanced platforms for monitoring and measuring end-to-end applications in a hybrid internal/public cloud environment. By tracking end-to-end application performance in addition to physical infrastructure Nimsoft enables application monitoring independent of the underlying infrastructure.
• IBM (with a good hybrid cloud vision) and Cordys, the latest Jan Baan company with a focus on enterprise cloud enablement.
As always, I welcome feedback, questions and comments. And if you know of other companies effectively enabling cloud computing with an impact on the enterprise you believe similar to those listed above, I’d be interested in learning more. You may reach me at cioblog@transitionaldata.com.