Windows Azure: Learnings from SaaS Data Center Migration
July 01, 2014

Windows Azure: Learnings from SaaS Data Center Migration

Byron Mathews, MBA | TrustRadius Reviewer
Score 8 out of 10
Vetted Review
Verified User

Overall Satisfaction with Windows Azure

Looking at lower costs alternatives to maintaining our own data center assets for our commercial buisness applications.
  • PAAS: we are a Microsoft shop so there were some benefits for us in the offering from a platform standpoint
  • Reduction in CapEx expenditures for servers, data center environments
  • Reduction in Expenses for warranty, maintenance, etc.
  • Greatly reduced provisiing cycle for server assets
  • Built in disaster recovery
  • Lays the foundation for continuous deployment of software changes without bring down te application service (SaaS)
  • Bandwidth. It's internet based access . . . . .
  • Can't really have dedicated links to Azure so that is a little unsettling from a reliability and QoS (Quality of Service) standpoint.
  • No a silver bullet. You will need to really analyze your applications and determine which ones are best suited for this offering. Still plan on hybrid cloud architecture.
  • Data costs: when you really crunch the numbers you can spend a lot for data retrieval/storage depending on the scenario. Pay attention to your data flows and retention.
  • Faster development and release cycle. This was a huge opportunity for the organization to be able to continuously deployment applications without having any customer downtime. No more maintenance windows. Position yourself for being always on to the customer. Improves customer satisfaction.
  • Pay only for what you need - shutdown what you don't need/use
  • Requires some organization changes to fully develop a DevOps function. Also requires lots of scripting but we found a good 3rd party vendor for that. The point is that it's just not a technical exercise -- takes planning and commitment.
Azure PaaS platform was better suited for us over AWS (we are a hard core Microsoft shop)
The offering wil keep getting better.
It is not suitable for big data applications as the pricing model would be prohibitive for the data movement, etc. It is suitable web tier applications - no hesitation.