Wednesday, May 9, 2012

The Cloudcast - Eps.36 - New, Open, Simplified Cloud Architectures


Guest: Randy Bias (@randybias, Cloudscaling Blog) - CTO/Founder of Cloudscaling

Topic 1 - Going back a couple years ago, you were one of the most vocal proponents of AWS, using Public Cloud and the AWS architecture. You also were very vocal about your belief that “enterprise models” (often calling out VCE Vblock) would fail. Did you find that it was necessary to use both examples to explain your belief in the AWS model?  

Topic 2 - Recently you seem to have shifted your focus (or at least the clarify of focus) on the idea that it’s “Legacy vs. Commodity” to “Old Apps vs. New Apps (or Web Apps)”, essentially putting a line in the sand for either an Enterprise or SP offering services. Is this a refocus for Cloudscaling, or just adjusting the lens for the conversation?

Topic 3 - Can these “new apps” be classified as mostly focused on a mobile strategy, or is there still a lot of room for apps that blur the line between consumer functions (Gaming, Social Media, Entertainment) and people wanting this as part of their work (and hence could also be used with browsers)?

Topic 4 - You talk about leveraging the concepts of AWS for the clouds that Cloudscaling builds for your customers. Beyond the idea of using commodity hardware or open-source software, can you touch on some of the technical aspects that are different from what you’d consider a Virtualization 2.0 architecture?

Topic 5 - Cloudscaling is a big supporter of OpenStack and that’s a key component of the clouds that Cloudscaling are building. How well equipped are you finding your customers to be able to operate OpenStack-based clouds? How well equiped are companies to build “design for failure” architectures or applications?

Topic 6 - Even though a few major companies have been leaders in Cloud for a while (AWS, Salesforce, Google, etc.), I think it’s reasonable to say we’re still in the early days of this industry trends. Should we just accept that it’s early days and winners will eventually emerge, but lots of learning is still needed?