CLOUD MIGRATIONS
Embarking on a cloud data migration program?
Don’t forget these 3 key considerations!
In today’s world, migrating your applications & data to the cloud is inevitable
You will have to start this journey — much earlier than you might have planned. This article will help you to draft a high-level strategy when you embark on this journey.
1. What is your “Why”?
There might be multiple reasons for initiating a data migration project
- On-premise software licenses getting expired
- Existing infrastructure not meeting the demands
- Slow procurement process impacting innovations
- Frequent performance issues, missed SLAs, high maintenance
- Coz everyone else is doing it!
Your reason can be anything. It can be just one or all of these.
Data migration programs often get executed in phases. It might not be possible to achieve everything in the first phase. So it is important to discuss, agree & get the buy-in from all the stakeholders on what will be achieved in your phase #1
All the next phases of the project will depend on the success of this phase, so ensure that the success criteria are very well defined & is practically possible to achieve in the agreed timelines
2. What is the Bigger Picture?
As stated above, data migration programs often get executed in phases.
A lot of attention is given to the success of the first phase. People often tend to go for the “low hanging” fruits in this phase to make it a quick success.
As a result, a lot of foundational activities are missed. Architecture blueprints, data migration strategy, reusable frameworks — all these are key for any migration project. Don't lose focus from these.
Your phase #1 should be all about building a strong foundation.
Low hanging fruits can still be part of this. After all, you need to show success to gain trust ( and the budget approvals) from business sponsors.
3. How do you Measure Success?
It’s simple. Just execute the workloads parallelly — on the old, on-prem system as well as on the new, cloud-based system. Compare the two output datasets & prove that all the data is exactly matching!
As easy as it may sound, the actual comparison & reconciliation is not easy. You might have 1000s of tables with 100s of attributes within each of these & millions of records to compare & validate.
And above all, this has to be done for multiple cycles over days, weeks & months. Ensure you have a sound data reconcilliation strategy to do so.
Have the required automation, tools, and expertise — not only to compare the data but to also debug, identify & fix exact issues.
If you are able to reconcile all your critical data after multiple test cycles, it would be a good indication of the migrated data’s accuracy
The key is how quickly you can do it!
Conclusion
Before you start your cloud data migration journey, give a thought to these 3 key considerations.
What is your real “Why”?
What is the Bigger Picture that you envision?
How do you plan to prove & measure the success of the program?
Hope that this article helps you to start your data migration journey. Please comment & let me know your thoughts/questions. Happy Journey!
