In search of the Cloud Computing Killer App

  • Article

A Killer Application is a need of the user group that can be filled, and by filling it, create an acceptance of that tool and the supporting methods/results

Let’s not just focus on placing your applications on the Cloud. The just is in effect a whole game changer, but let’s move further away and define some criteria to seek on the cloud killer application. Note that there are persons that consider that the Cloud doesn’t lead to software innovation but just to hardware advances exemplified by NetBooks and Smartphones. For me a Cloud Killer App will have the following attributes:

  • Going beyond limits of current transaction processing. IEEE Spectrum publihed an article proposing that consumers will demand intense user experiences and focuses on a single category: electronic gaming. It makes sense for the scalability of millions of concurrent users. It needs to be demonstratad from the peformance perspective. Load and performance testing have also been nominated as killer apps.
  • Breadth reach. Apps for Democracy list interesting cases that go beyond publishing public data sets on the cloud on government services that are required by all citizens of a given nation. Dan Bricklin envisioned “public IT infrastructures” on May 2004.
  • Benefits from new platforms. Ray Ozzie has been focused on Live mesh. It enables software developers to build applications that sync data between the cloud and all user devices. It’s common to hear 2009 complains on the users that want their home setting to match their works but the software installation are totally independent. A killer app would most likely leverage the new cloud platforms.
  • Green algorithms. The sustainability story is pretty big almost everywhere. But current software design projects do not deeply consider energy consumption. Microsoft is measuring CO2, Water and other energy consumption metrics per applications (e.g. Hotmail) on global datacenters -  the operations perspective is available - but the current engineering and implementation is not explored at all. I envision the possibility of switching to different algorithms given external conditions.
  • Has a real impact. The value of IT is and will always be the way it solves real world problems. I laugh when I hear a new PC Lab has been setup on a school. The real question is how the magic software is really changing the learning capacity of kids. View the first 15 minutes of this video.

There are few Killer applications patterns in existence: ‘mobile marketing’, ‘VoIP’ and other technologies have failed to be considered such. Worst case, the killer application is the existence of a large number of cloud applications helping us solve our personal and business problems. What do you think?