View From Above

Apr 25 2011   4:53AM GMT

Cloud Haters Jump At Amazon EC2 Fiasco



Posted by: Ron Miller
Tags:
Amazon Web Services
enterprise IT
Outages
Cloud haters had a field day last week when Amazon EC2 was pretty much hosed for the better part of two days. By now you know the story. It took down major swaths of online service real estate including Reddit, Quora and Foursquare.

But as Randy Moss, once said after a particularly testy exchange with the media, “The New England Patriots are 2-0. We got one in the division. So all you haters keep hating. We’re coming!”

Moss was right about one thing, if you want to go negative, that’s your choice, but neither the Pats nor the cloud are going anywhere, no matter how big the disaster. It’s not. So just take a deep breath and let’s see what we can learn from what happened last week.

Here’s what what we do know. Services like Amazon EC2 are a good thing. They give small companies like Reddit, Quora and Foursquare incredible scale and elasticity (the ability to scale up and back as traffic requires). And they can do it at a price that would be out of reach without scores of venture capital to build their own data centers.

Here’s another thing: What happened to Amazon can happen to anyone. As I wrote the other day on the Business Service Management Hub, “If it can happen to Amazon, it can happen to you because at its heart what is Amazon but a giant data center, whose core business is keeping other businesses going.”

That’s right, the idea that you are somehow protected from disasters like this because you keep everything behind your firewall is simply naive. A data center is a data center is a data center, and if it could happen to Amazon with all of its disaster recovery plans, it could certainly happen to you too, whether you believe it or not.

Perhaps, the best lesson to come out of this was what Patrick Corrigan wrote on Storage Bytes Now, and that’s an age-old one, don’t put all your eggs in one basket. Just as I’m sure, you spread your in-house data around various resources, don’t hitch your cloud wagon to a single vendor because when disasters strike like happened last week, you want a way to keep it going.

One way to do that is to make deals with more than one vendor. Corrigan was speaking in the context of storage and backup, of course, but it’s a lesson that could apply to any online infrastructure service. Just as your Exchange server goes down from time to time, taking the company email with it, so will your online infrastructure provider, as we learned last week in a spectacular fashion.

The take-away here isn’t to stay away from the cloud because bad stuff happens. It’s to realize that bad stuff will happen and try to find ways to mitigate that, just as you do with your in-house systems.

I’m probably being hopelessly naive here, but I’m hoping we can get past the point of a rash of this-is-the-end-of-the-cloud blogs and articles every time an outage happens in the cloud. It’s going to happen again, and it’s probably going to give you all cloud haters a chance to keep on hating — whether it makes sense or not.

Photo by Keith Allison on Flickr. Used under Creative Commons License.

3  Comments on this Post

 
There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when other members comment.

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
  • Rawler
    The problem with the "cloud" is that it's vaporware. (Pun intended). It's a marketing-term, slightly hyped, and really doesn't say anything about what is actually offered, much like "Web 2.0". The term itself is of no value, since it does not represent anything concrete other than possibly being equivalent of "outsourced". Instead, one SHOULD be talking of exactly what is offered: * A remote-application, and over what interface? (Web, Remote Desktop...) * Remote storage, and on what level (SQL, NoSQL, Block Device, Filesystem ...) * Virtual machine (With what capabilities?) * Cycles on compute-clusters (I.E. distributed.net) * And of course all services should come with a clear Service Level Agreement. On the other hand, the REAL decision-makers in the large Enterprises may not understand these terms, which is why "cost reducing, web 2.0-enabling cloud-services" is a better label. Just hire consultants afterwards and let them figure it out.
    0 pointsBadges:
    report
  • Ron Miller
    Rawler: I have to disagree that it's vaporware. The cloud is very real. What Amazon is doing is one part of the cloud. There are three parts: * Infrastructure as a Service (what Amazon is doing) * Platform as a Service (what VMware and Salesforce are doing) * Software as a Service (what Google and many others are doing). You are also confusing Web 2.0 and the cloud. Web 2.0 was a term coined by Tim O'Reilly to encompass services that enabled end users to create content very easily. You are right that both terms are overused and abused, but that's not to say that these terms don't refer to a valid set of services. Thanks for the comment. Ron
    500 pointsBadges:
    report
  • Rathweg
    The companies that are looking at cloud computing need to hire a pro to look hard at their design- some techies who have been arounf the block a few time. what is the cloud anyway? a big enterprise server - like a mainframe.
    0 pointsBadges:
    report

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to: