The conversation about migrating databases to the cloud tends to focus on the mechanics of getting your data to the cloud. That's jumping the gun. Before talking about how to move the data, you should talk about whether you should move the data to the cloud.
Despite the growth of analytics and the potential value in every byte of data, not all the data stored in your databases is essential. Some is too old and outdated; some duplicates other data; some is invalid or incomplete. Some may be badly structured or might benefit from features available in a different database.
Lifting and shifting a database to the cloud can seem like the most straightforward and efficient method of data migration, but migrating to cloud isn't about short-term benefits. To take full advantage of the cloud, you may want to take this opportunity to restructure your data or switch to a different database. The migration might take longer but you'll end up with a data environment that gives you more flexibility and positions you to take advantage of your data more effectively. Ask yourself these questions to decide whether to invest the time rethinking your database before migrating it:
- Why do you have this data? There's a tendency to want to hold on to all data just in case it's important, but there's a lot of data that's obsolete or has only short-term value. If the data duplicates data stored elsewhere, you may be able to use a single database instance instead of copying each application's version.
- Is the data good data? Incomplete or inconsistent data can be worse than no data. If every application has its own definition of basic terms such as customer, you're limited in applying analytics and discovering insights. You might want to take this opportunity to develop a data model that's consistently applied across applications.
- Is the current data model a good data model? If the database design was created a long time ago, it's probably undergone many changes over the years as applications changed and as the data became better understood. It may be time to redesign the schema and clean up the effect of all those changes.
- Does the current database meet your needs? Much of today's data is unstructured. In the past it was often forced into relational databases, but today's NoSQL databases can provide better support for working with that data. Even if you're happy with your database architecture, it may make sense to upgrade to a later version as part of your migration. You'll need to do that eventually, anyway, and doing it now may take a little extra time up front but save you doing a second validation later.
Have you thought through your database migration to the cloud? Professional services from dcVAST can help you make the necessary decisions. Our expertise with Oracle gives us a deep knowledge base to guide and execute your strategy. If you migrate to Amazon Web Services, managed AWS makes sure your database in the cloud continues to meet your business needs. Contact us to learn more about how to make the best use of your database in the cloud.