Microsoft Azure as an 'Infrastructure as a Service'
My advice to other executives is with the pace of technology change acclerating, you can’t hold onto any preconceived notions. For instance, years ago, VM environments could not be considered as effective as physical environments. At the time it was not easy to manage VM’s and ensure they managed resources effectively. Today, the majority of our environments are VM’s. They make monitoring and allocating resources to our servers easier and faster. Since the technology continues to evolve, what I have found to be effective is find a trend that you believe can have a major impact to your organization (whether cost savings, feature content, etc) and pilot it within your company. If it seems promising but doesn’t pan out in a pilot don’t completely abandon but revisit it again. Continue to monitor the developments of the trend. If it’s truly a sustainable trend it will evolve. If not it will disappear in favor of a better solution. We have piloted the use of Azure within our company through the stand up of a server that has a limited scope. We test out the ability to stand up the server, manage resources and confirm the cost model. We now have data to base an expansion on and determine our roadmap for the use of cloud services.
A Brief Overview of Microsoft Azure
In my opinion cloud services in its different forms (SaaS, PaaS, IaaS etc) is the trend and Microsoft is establishing itself as a leader in this space. Azure for us is infrastructure as a service (IaaS). We are making decisions based on capital expenditures for end of life servers to see if it makes financial and operational sense to shift some work to the cloud. We believe it will give us the flexibility to scale up and scale out easily without some of the constraints of the CapEx process. If our situation changes we can remove servers, and storage and not have excess capacity we don’t need. It is truly real time infrastructure with no long lead times to obtain hardware, storage, networking. However, there is a learning curve to managing your infrastructure in the cloud and some pitfalls when it comes to sizing and specifications. These pitfalls can lead to your costs escalating quickly without enough oversight.
Since the technology landscape continues to evolve, what I have found to be effective is find a trend that you believe can have a major impact to your organization and pilot it within your company
On the flip side the cost of hosting your own data center gets very expensive especially with the pace of technology change. The ability to bring up an environment and easily remove it are great factors to consider if you are testing new solutions or approaches and don’t want to make a significant long term investment. Over time with additional technology changes and competition the cost curve for IaaS will continue to improve making it more competitive for mid to large organizations.
Challenges in Moving towards Microsoft Services
When considering the challenges of moving to Microsoft Services, I think the shift of cost from a CapEx model to an OpEx model is one to consider. In the past you made an initial investment of X dollars and over time you spend Y to replace failed hardware, created redundancy for DR and other changes. The hardware had a specific life span. Today, based on your initial spend alone, it may appear you’re spending more to get a like environment with Azure. When comparing the two remember to factor other costs like individual component replacement, systems monitoring tools, multiple internet service providers and other factors that you don’t need to account for with a server in the cloud. The full cost may be closer than you think. Don’t forget to consider the soft costs such as personnel, 24 x 7 uptime management, DR, etc. Considering both the hard costs and soft costs beyond initial purchase, they can add up to significant savings moving to the cloud. You need to look at the TCO and not just the monthly fees associated with your environment when you compare.
Needs and Technologies go Hand-in-Hand
When it comes to needs and technologies and the drive to innovate, I am not one that believes in having a solution that is looking for a problem to solve. I start with the challenges or new opportunities we face and then consider the technology landscape to see what may be the best way to solve it immediately and long term. Since, the lifecycle of technology is short; a long term plan of more than three years may be outdated quickly. Finally, if we are considering a shift in technology we consider our plan to sunset current solutions that may be out-dated. Your team’s time may be consumed supporting old technology where it is better spent retiring those solutions. Not only will you gain in efficiency and new capabilities but it will also provide a great source of job satisfaction for your IT staff.
Advice to CIOs
For my peers who may be considering Microsoft Services for their infrastructure, I would say identify an area of your operations that is in need of a hardware upgrade and pilot it on a limited basis. Learn how to manage this new paradigm and consider the total cost of ownership when evaluating the impact on the financials. At that point set a roadmap to migrate the systems that can best be served up through the cloud. If your business is not data center operations, then maybe its best left to the experts whose business it is.