I’m often taken back by the number of enterprises moving the cloud that have not considered data modernization as part of their cloud migration strategy. I know why: the money. However, not doing data modernization as part of moving to the cloud will cost you much more than you think you are saving.
Data should be a first-class citizen when it comes to your cloud effort. While the number of things to think for cloud migration about gets longer, don’t let data fall of your radar. The ROI for any improvements you make there is 20-fold—trust me.
Here are three things to consider.
1. Look at the efficiency of special-purpose databases
No matter if you’re talking about a blockchain database, an in-memory database, a distributed database, or a data lake, there are special-purpose databases that are built for a particular purpose that may be better fits for your applications.
At least ask the question before your migration. With many enterprises using the same old enterprise relational databases that are demanding higher and higher license fees, moving to a special-purpose database not only will be an increase in functionality and optimization, but it is likely to be much cheaper in the cloud.
2. Consider revising both models and structures
Most databases are not designed well, and many organizations are just picking up those bad designs and relocating them to the cloud. Moreover, they are not considering other databases models, such as object databases and graph databases. Moving to the cloud is a great time to look at other database models.
However, what is pretty much mandatory is that you revise any deficiencies in the existing structures. This means revising your databases so they best resemble the business. Most databases don’t do that today because the business changed over the years but the databases did not.
3. Find and remove redundancy
How many versions of customer data do you have in your databases? How about inventory data? More than one? More than 20? This is sadly pretty normal, so moving to the cloud is the time to remove database redundancy and come up with single sources of truth for the data.