An increasing number of enterprise systems are moving to as-a-service models, reducing a company’s overhead and turning traditional facets of information technology—those that have used up both real estate and capital expenditures—into outsourced subscriptions that are managed by outside companies. Infrastructure, Networking, and Reporting as a service are already popular. Moving the databases themselves off a company’s property and balance sheet into a cloud architecture entails what is known as Database-as-a-service (DBaaS) (Bonthu, Thammiraju, & Murthy, 2014). There are many factors involved in establishing the DBaaS environment and migrating the data from on-premise boxes to cloud.
There are typically eight steps involved in moving from on-premise to cloud databases:
- Define the scope of migration
- Ensure data security
- Select service provider
- Map the data
- Schedule the migration
- Select tools for migration or develop migration scripts
- Test before (and after) the migration
- Actual data migration
The actual migration, insofar as relational databases are concerned, typically consists of three steps:
- Relational schema migration – it includes the migration of tables, indexes and views.
- Data migration done via tools or migration scripts. The time required for data migration depends on the size of the database.
- Database stored programs migration – the migration of stored procedures and triggers. (Vodomin & Andročec, 2015)
The different types of cloud databases available, relational and non-relational, make for a variety of ways to migrate and a number of considerations for enterprise migration. Regardless, a one-time expenditure on migration can save countless dollars and hours of ballooning infrastructure and database sprawl. It is much easier to handle such sprawl by responding with both storage and virtual machine elasticity as opposed to investing more in onsite resources (Bonthu, Thammiraju, & Murthy, 2014). Further research in this space is warranted as the options for cloud architecture increase and companies have more options for service-based managed IT.
References
Bonthu, S., Thammiraju, S. D. M., & Murthy, Y. S. S. R. (2014). Study on database virtualization for database as a service (dbaas). International Journal of Advanced Research in Computer Science, 5(2), 31-34.
Vodomin, G., & Andročec, D. (2015). Problems during database migration to the cloud. Paper presented at the Central European Conference on Information and Intelligence Systems, Varaždin, Croatia.
Most content also appears on my LinkedIn page.