There are continual, iterative conversations taking place around moving to a hyper-scale cloud platform like Amazon Web Services and Microsoft Azure. There is significant innovation taking place, and more and more enterprises are taking the leap. What is influencing that decision? Here are a few key actions that are being taken by these forward-thinking organizations:
The path to hyper-scale platforms used to require a complete re-architecting of your applications to take advantage of the economies of these solutions. The evolution of cloud platforms now enables a phased approach to migrating your systems. This method minimizes the work required to migrate applications or workloads to a cloud design as the first step in ultimately positioning your offerings to take full advantage of flexible architectures.
Brown Field Design
“Brown field” is a term that AWS coined and has really embraced. Economies have shifted so you can migrate your systems in more of an “as is” state into AWS without having to re-architect it on day one. This was once a limitation, as the costs would outweigh the benefits, but today there are situations where that can be accomplished in a cost-neutral manner, and in some cases, lead to savings.
Considerations with brown field design include:
- Optimization – The key step after migrating systems to a brown field design is to optimize your resources:
- CPU - In the case of moving to AWS, you may find that you do not require the same number of CPUs you once leveraged in your on-premise solution. Most firms that manage applications on premise have over-purchased so that they can grow, scale and protect against unforeseen performance issues. When you adopt a brown field approach in AWS, you will want to run your environment with the proper number of resources to assure scalability. The same story is often true with storage usage.
- Storage - Much like CPU, on-premise solutions typically mean there is a lot of headroom purchased for scale. It is also possible that the storage units leveraged are designed to account for an imbalance in terms of performance. When you move to a brown field, you can not only right-size to consumption levels, you can leverage different classes of storage like EBS, EFS, Glacier and more to make sure you are paying the right amount for the need.
- Other – There are countless processes and procedures that will require inspection as you make these migrations. Processes are easily the most important aspect of the administration of any workload, so you will really want to evaluate those and make sure you know how you intend to manage expectations and keep everything performing and protected.
- Monitor and Manage – You can certainly leverage many tools you have already invested in when moving to the hyper-cloud. However, you should also dig into the built-in offerings that can provide great information and automation capabilities. As an example, in AWS there is an Advisor feature that will not only monitor the performance of your systems, but also give you quantifiable advice regarding changes you may be able to make in order to save money. If you want to look at log information and you want that to feed into other systems you leverage for security/compliance, there are built-in tools that can be utilized to compile that information.
- Services – The one thing to really keep in mind is that with the hyper-providers, services from the manufacturer are limited. You have to make a conscious decision whether you want to manage these designs internally, which requires comprehensive training and diligent practice, or if you prefer to have someone help you. Having someone to call that knows your environment, knows the platform and can bring people to bear that can address your needs is something that you will likely want to consider.
We are in an exciting time in this industry and are inspired daily by the innovative forces at work. Don’t fear the cloud¬¬—embrace it—but make sure you plan accordingly so your experience is exponentially more productive than your current state.