November 13th, 2008
Making clouds a move forward
One of the big IT challenges of 2009 will have to be assuring that cloud computing really does represent a move forward.
While there has been a lot of talk about clouds over the last year, about Amazon, about Google and about Microsoft, in fact Amazon’s is the only one open for business. Google has a consumer focus, and Microsoft’s Azure is still vapor. (Yes, funny.)
Cloud computing is an outgrowth of virtualization, separating hardware identity from the software operating system. That’s the big difference with regular hosting — you don’t know exactly where your application is running and don’t care.
But for clouds to be really useful we need visibility into what they are doing. Announcements like Hyperic’s management system for Amazon’s cloud, or rPath’s extending its services to clouds, are a big part of that evolution.
Because clouds are sold as SaaS this lack of visibility could mean very, very bad things for open source. When operations and costs are both invisible, it’s easy for someone to siphon your money out of their cloud without your knowledge.
This may be why Gartner Group is telling big enterprises to build their own clouds. Having your own operation provides visibility into what vendors are doing and lets you make apples-to-apples comparisons of price and value.
But when people talk about getting started with cloud computing, what exactly are they talking about? Right now it’s mainly choosing vendors and building relationships.
I would hope that clouds will end the license wars and cause computing to evolve. The question is how you can help.
The best way may be to stand up for your rights. Maintain your own applications and use clouds to scale them. Demand proof of cost savings by cloud vendors. Hold tight to your money and your really important data.
For a decade I’ve been writing about Web sites, about e-commerce companies, and about Web 2.0 outfits that hit a wall of scaling. When users “rush to the rail” even systems like Twitter can slow down or crash.
The real promise of clouds is to end this fear. You build your thing, you pass management to a cloud vendor, then you concentrate on marketing and cash flow.
So let me end with a poll question. When do you think this vision for clouds may be realized? And for the comments, what do you think are the key questions in getting there?
Loading ...
Dana Blankenhorn has been a business journalist for 30 years, a tech freelancer since 1983. You can follow Dana on Twitter. See his full profile and disclosure of his industry affiliations.
Subscribe to Linux and Open Source via Email alerts or RSS.
| 11/13/08
