Last year, I created a version of DotNetNuke called the open web portal edition. This version provides a first time visitor of a site with a DotNetNukeb child portal (called portlet) in administration mode where the user can immediately start playing around with DotNetNuke and get a feel without the need for them to sign up first. The portal is intelligent enough to remember all the changes made when the user comes back to the site, if they left without signing up. When the user finally decides to sign up, they simply create a friendlier URL for their portlet. They can then use this URL to sign in from any PC and make changes and they can also send it to friends, customers, family, etc depending on what they created the portlet for. Please note that there is no fee involved, so users can use it as they please.

 

This implies that the open web portal will have users who create their portals and those users will have users of their portals. Now you can imagine how resource intensive the open web portal edition is.

 

I am generally happy with the functionality of the initiative, which is basically DNN but in a web 2.0 open web portal style. What I am not happy about is the performance of the site which is quite understandable since a brand new portal is created on every first time visit. Also, we all know that activities taken on the workflow for creating a new portal are much more intensive than when you are just visiting a page of an already created portal. What also compounds this performance issue even more is that currently I am hosting on a share environment where I am sharing resources with 250+ websites. Try it here www.iflaker.com. Now that I have over 3,000 portlets, it will take you over a minute on a first time hit to load. The decision for hosting on shared environment for such a resourceful intensive initiative was purely financially based. I just wanted the initiative to go live so that I can showcase the concept to a few people. Now I believe it’s a right time to invest on performance and roll it out at a larger scale.

 

Cloud computing is the current web 2.0 hype when it comes to web hosting. I believe that the open web portal edition as well as cloud computing might prove to be a perfect marriage. I have looked at a couple of cloud computing vendors and I personally believe GoGrid is at the top of my list based on the fact that they bill per RAM per server per hour. This implies that at any given point in time you can adjust the resources you need based on whether it is pick time or whether there is an unexpected influx of visitors, etc. They also claim that making such changes is just a self help configuration so you don’t have to wait for your hoster to do it while your site is down. I am more likely to go with them unless you have someone to suggest.

 

If you are interested in the developments on this project please follow me on twitter.com/gabrielnkuna. I will reveal it all every step of the way.