What's Happening

Azure – Where Do I Start? (Part 1)

Once your business has decided to go to the cloud, there are quite a few things that need to be done to create a path for success. The articles in this series give a high-level explanation of the questions that have to be answered prior to deployment.

What are you trying to achieve?

One of the common problems I’ve seen companies have while moving to the cloud is a lack of understanding what goals are driving their business towards the cloud. My suggestion for anyone thinking about moving to the cloud is to reach a consensus within your business about what goals you are targeting before getting started. Some common examples of goals driving companies to the cloud are:

  • A desire to close their data center or colocated hosting
  • Scale on demand
  • Lower Costs
  • Reliability
  • Geo-redundancy
  • PAAS tooling only available in the cloud

What does success look like?

Once the business goals are agreed upon, the next step is to understand what success looks like. This should be something measurable. Some examples based on the previous list might be:

  • Percent of software systems running in data centers vs the cloud
  • A difference in time to execute a process under a consistent small load vs consistent large load
  • Current Cost of Ownership of current solution vs cloud-based solution
  • Measurement of cost to the business during an outage multiplied by the average outage length in a year
  • Time to rebuild and get back online after the entire data center is wiped out multiplied by the cost of an outage

Once these measurements are understood the team should measure current state and define the desired state that is being targeted.

Should we proceed?

Now that we have established what the business is trying to achieve and how that can be measured, the next step is to calculate the monetary value. For example, if the reason to move to the cloud is to improve SLA, we can look at the total cost of outages in the current state subtracted from the total cost of outages in the desired. Once this has been calculated, the decision to proceed is a matter of looking at the value of the project versus the cost to do the project. In some circumstances, this calculation is much more difficult. In the example of moving from a private data center to the cloud, this may not be a monetary decision, but could instead be required due to a data center closing. (Something we are seeing more of as the cloud become prevalent) In this case, we need to understand what the alternative is since staying in the current state is not an option, and compare the cost of moving to the cloud with the cost of the alternative.

In reality, this is an oversimplification. There are many potential benefits and all of them need to be factored in. For example, if the primary driving factor was to improve SLA, but we find that the value is about the same as the cost, the other factors may tip the balance in favor of still moving to the cloud.

We’re ready, now what?

After determining what you are trying to accomplish, why you are trying to do it, and if it brings value to your business, you are ready to get started right? No, you are ready to start planning your cloud deployment. This planning consists of technical work balanced with business needs which includes:

  • IAAS or PAAS?
  • Hybrid or Cloud Only?
  • What architecture should I use?
  • How should I store data?
  • What are my Security and SLA requirements?
  • How do I optimize for cost?

I will cover these topics in part 2!

Share