Competitive advantage is directly tied to how well companies are able to build software around their most important asset: data. Rigid relational schemas often require downtime and significant application code updates in order to make even simple modifications, such as adding a new data attribute. In MongoDB, entities are modeled as documents that map naturally to the same objects that developers are used to working with in their programming languages. Additionally, legacy relational databases were not built to scale horizontally. Sharding data to handle large data volumes and ensure lower latency is typically a significant manual process that requires custom application logic to query across multiple shards and aggregate results.
MongoDB Atlas on Amazon Web Services (AWS)
MongoDB Atlas removes the need for a complex object-relational mapping layer and allows developers to build and release new features quicker. MongoDB Atlas is built to be distributed and to handle sharding transparently to the developer. With Atlas, no application code changes are necessary when an application needs to scale out from a 10MB to a 500TB dataset.
MongoDB Atlas is well integrated into the AWS environment, and the document-based database works seamlessly with AWS products. To learn more about common integration and project requirements, refer to Managed MongoDB on AWS.
Migrate to Atlas on AWS with Relational Migrator
Some customers have successfully migrated their relational workloads to MongoDB Atlas on AWS. One example is Cox Automotive, whose system was hitting limitations on the relational database. The company migrated to Atlas on AWS and leveraged capabilities like Atlas Data Lake (powered by Amazon Simple Storage Service (Amazon S3)) and Atlas App Services. Read our customer case study to learn more about how Cox Automotive uses MongoDB Atlas.
At the same time, other companies have struggled with how to approach this challenge. When considering such a migration, it’s important to think carefully about data modeling. Although it’s possible to naively move a relational schema into MongoDB without any changes, this approach won’t deliver many of MongoDB’s benefits. A better practice is to design a new and better MongoDB schema that’s more denormalized and potentially to take the opportunity to revise the architecture of the application as well.
To make this process easier, we’re developing MongoDB Relational Migrator. Relational Migrator streamlines the process of moving to MongoDB from a relational database and is compatible with Oracle, Microsoft SQL Server, MySQL, and PostgreSQL. MongoDB Relational Migrator connects to a relational database to analyze its existing schema, then helps architects design and map to a new MongoDB schema.
When you’re ready, Relational Migrator will perform the data migration from the source RDBMS to MongoDB. Migration can be a one-shot migration if you’re prepared for a hard cutover. And soon, we will also support a continuous sync in case you need to leave the source system running and continue pushing changes into MongoDB.
With Relational Migrator, you can map your relational schema—or just a piece of it, if needed—to a new MongoDB schema. Relational Migrator helps the design and mapping process with common MongoDB schema design patterns built in. Based on this schema mapping, you can move data into a target MongoDB cluster. Relational Migrator will support both Snapshot (one-time) and continuous data migration.