Top 5 Cloud Migration Tips
Migrating to the Cloud can be daunting, complex and time consuming. The first few migrations may be tough. However, If you plan correctly and have prepared adequately then you will likely have a successful migration. This post is for anyone considering migrating or implementing cloud technology.
I’m Gus Segura, Director – Principal Data Science Engineer at Blueskymetrics. We have been helping customers migrate to/from the Cloud for years. I’m ex-Amazon and ex-Google and have built world scale architectures in Marketing, HealthCare, Banking and Industrial IOT / Controls. I learn allot working in those environments and the following is list of the Top 5 Tips for a successful migration.
If you’re starting a new project or need help with your current migration – Please Contact US.
Email: email@example.com – Phone: 765.325.8373 ( call / text )
1) Understand Goals – Success Metrics.
You’re migrating to the cloud. Know the goals. I.E. Saving money, reduced complexity in the workflow, implementation of new and advanced technology, etc.. Also, Know how to measure success. Data mine you’re trouble tickets, issues tracking, customer support tickets, resource allocation, sprint planning / burn-down. Show how you have decreased resource cost and improved customer experience. Be introspective and celebrate success along the way.
2) Understand Project Constraints, Timeline, Resources.
Project Management 101. Know your timeline, cost and quality constraints. If your running DevOps while your migrating – that’s fine. Be prepared for a bit of extra workflow to update your team on resources (like a database endpoint or new s3 bucket location). Know the elements and critical path items that “must be” completed by a specific time or event and have a backup plan.
Don’t be afraid to “leap” to a migrated cloud resource before your ready “if its really needed” – Do you have good controls, testing, qa, etc.; Have confidence that things are going to work and you can always recover to your legacy systems before final cut-over.
Remember, Cloud Technology is very dynamic and we are our best resource.
3) Understand the Data, Schema, Embedded Logic, etc.
It very important to understand “what” you are migrating. Please see my previous article on Data Warehouse Migration Considerations as an example. [ size metrics, database objects, program units and more. ]
4) Understand the limits of the Technology Stack.
Know, There are limits to the services and technologies you may be considering. Carefully, research these during your architectural process. Ask for white-papers, case-studies and reference sites from you’re vendor. Know the limits on size, speed, location, latency, scale and more.
I.E. Should I build a high speed pipeline using a relational database or should I be using a data pipeline and messaging service? Why? Remember, there are “So, many services to choose from now and more getting added regularly.” It’s over-whelming at times but it is manageable.
5) Be Flexible, Bend but don’t break.
You will most likely be over-run with all the things that have to get done and all at the same time. Back in the “rack and stack” days, we could plan for more time waiting for facilities to wire our new machines. In the cloud, everything “that’s not in beta” is fairly available. Literally, when implementing cloud orchestration, one could do everything at once in parallel.
Don’t do that … until your ready and have evolved through a few migrations. You will be pushed hard enough on your first few migrations. Be flexible in what gets done first and move away from older methodologies as you migrate to the cloud – your mind set may change along with the way you solve problems.
Stay informed, click subscribe below to get a copy of the Blueskymetrics newsletter.
Please Contact US or Subscribe to our Blog or would like more information.