There are different public cloud use cases. Here at The Virtualization Practice we moved our datacenter from the north to the south part of the country and utilized the cloud to host the workloads during the transition. Edward Haletky, yesterday posted about Evaluating the Cloud: Keeping your Cloud Presence and presented the question and his thoughts of is it worth staying in the cloud or bringing the data home.
Technology for the cloud and cloud services is currently in explosion mode with the big players like Amazon, Rackspace, Microsoft, VMware and now we can add IBM to the list with their recent purchase of Softlayer. This acquisition will become the foundation for a new IBM Cloud Service that will another division of IBM Global Technology Services and setting the stage for IBM’s managed service expansion. These platforms and services being, for the most part, stable and available, but one of my biggest concerns with where the technology is currently is the difficulty getting from point A to point B in a smooth non-disruptive way. It is like we have several superhighways available but the entrance and exit ramps are still be developed. I believe this is part of the growing pains of the technology and believe more and more companies with technologies like Hotlink to build and provide the transition between highways.
In my humble opinion, I see the current cloud as the 21st century managed services or in a way datacenter co-location with an ability to “burst” your resources to keep up with any sudden and sustained demand. A good number of people that I have spent time talking with seemed to be more inclined to keep things in house and private as much as possible. Security, Control and Migration being the concern terms that I seem to hear the most from my peers in the industry.
The push to a Software Defined Data Center may help to really tie the public cloud as an extension of the private infrastructure which could help elevate some of the security and control issues. More time is going to be needed for the migration part but I have confidence that solution is not going to be that much further in the future.
The possibilities for the cloud seem almost endless and once all the pieces get into the place to really tie things together, the cloud boundaries can really be expanded and isolated. There is a tried and true rule that has been in place since the start in that just because you are able to do something does not mean you should. With that said, will tomorrows cloud be a tightly knitted extension of the private corporate resources or will things and prices change enough that it would get cost effective to completely, in part co-locate, and outsource to a managed service provider?