Cloud computing has fundamentally changed the way that organisations can do business. Bringing benefits, that were previously reserved for those that could spend the cash to build capabilities and capacity, to anybody with a credit card.
Digital disruption means that to remain competitive organisations, increasingly have little choice but to utilise cloud technologies.
Migration to the cloud should be about adding scale, being efficient, enabling innovation and providing great value. It should not result in head scratching and introduced problems! Before pulling the trigger on cloud projects, pause to make sure that you are not about to make an easily avoidable mistake.
Capability Based Planning
So, you’ve got engagement from the business, worked through your goals, principals and you’ve decided on your target cloud architecture. Before you move to implement, don’t forget the people who are going to be implementing and supporting the cloud platform.
What capabilities do your existing teams have? Will they be able to support the platform without additional training? do they already possess the required skills? Are those skills formalised through certification?
Migrating to the slickest cloud platform imaginable isn’t going to add value if you forget to bring existing teams with you.
As part of your planning and architectural work to define the cloud target architecture, don’t forget to analyse existing team capabilities and factor that into your decision making. This might not impact the target architecture. However, it will help to identify gaps, new hire or training costs.
Would you deploy a new data centre without having a clear idea of how you are going to manage every aspect of it? Would you not insist upon detailed plans for monitoring, backup and protection of the data centre? Why would should that be any different for cloud services?
This isn’t to say that you need to design and deploy a monolithic application stack to cover each of those disciplines, but the need for those disciplines isn’t removed by the cloud.
As part of your planning and architectural work, make sure you have plans for each element that makes up your management stack in your current data centre. Trying to retro fit management technology, is a position that nobody wants to be in.
Target the Right Services
I know that the reason you are looking at cloud services in the first place is the key monolithic application, that hasn’t been updated since 1990 and if it fails half of the organisation will be sat twiddling their thumbs.
I know that the reason you are looking at cloud services in the first place are those legacy applications in the corner, that seem to work even if you’re not sure how and that nobody has touched since ‘Dave’ retired.
But let me ask you this, are those services really the best services to pick to start your organisations cloud journey? What will be the damage caused by encountering problems trying to migrate those services? Is there an outside chance that it could irrevocably damage your cloud aspirations long into the future?
When considering applications to migrate into the cloud ask yourself these questions;
- Do you consider the application to be well designed? If it runs badly now, then chances are it is going to run badly in the cloud. Think about re-architecting the services for the public cloud rather than migrating something sub-optimal.
- Does the application rely on legacy security frameworks? Migrating an application that is utilising an outdated security model to the cloud, is not going to change that fact or fix associated vulnerabilities.
- Where does the application process data? Is the application closely coupled to its data? Consider decoupling the application and its data before attempting to migrate. Moving a tightly bound application to a platform designed for distributed, scaled and reliable workloads, is to ignore key benefits of cloud computing.
These are a few mistakes to avoid when working to create a cloud strategy, hopefully now you are aware of them, you can successfully avoid them!