The End Of Software - Anibal's Blog

Following the excellent thread of articles from ArsTechnica, now they publish "Cloud platform choices: a developer's-eye view". In the overview, we can find the state of the art for a tech business development in the cloud. As bullet point we can have a tech business running on: A DIY model (buying servers, building a network, etc) Traditional hosting/dedicated services Cloud-based Infrastructure as a Service (IaaS) (e.g. Amazon or Rackspace) Platform as a Service (PaaS) (e.g. Google's App Engine or Saleforce.com's Force.com)

To add a note to the original article, thinking to begin small, in an entrepreneur or SMB context:

A "DIY model" is out of the question, too expensive and complex. There's no way to leverage in time the initial costs. It's better to relay on someone.

"Traditional hosting/dedicated services". It's already commoditized. There's a lot of price choices, satisfied (or not) customers, opinions, etc. A whole market.

"IaaS" (Amazon or Rackspace)... Immature ... Only for experts / early adopters. My advice, begin small... E.g. dumping part of your web server static files... and evaluating the cost. The question is: Which is the benefit between a traditional dedicated server vs. a fully 24x7 deployed instance?

"PaaS"... If IaaS is immature, PaaS is only for hardcode innovators. There's no standard to compare implementations, no previous experience, and you are going be locked in with one vendor.

So, to be practical, go ahead with the traditional services, and keep an eye on the IaaS competitive new offering.

Share this post