AWS Database Blog

Amazon DocumentDB (with MongoDB compatibility) re:Invent 2019 recap

The Amazon DocumentDB team had a great time meeting with many of you at AWS re:Invent 2019, listening to your feedback and use cases, and understanding what you want us to build next. For me, the highlights of re:Invent were the two great customer use cases presented by FINRA and Fulfillment by Amazon (FBA). Both customers talk about their respective journeys migrating to Amazon DocumentDB and how they were able to scale their applications faster while enabling developers to be more efficient.

This blog post rounds up the Amazon DocumentDB recorded sessions from re:Invent 2019.

DAT372 – Migrating your databases to Amazon DocumentDB

Jeff Duffy, the senior specialist solutions architect for Amazon DocumentDB, provides an overview of customers’ strategies and considerations when migrating to Amazon DocumentDB. The session includes discussions about relational and nonrelational sources, migration phases of discovery/planning/testing/execution, cluster sizing, and migration tooling. The highlight of the session is FINRA’s story of how they migrated from a relational database using XML to Amazon DocumentDB. With Amazon DocumentDB, FINRA can move faster using JSON natively in their database because there is no transformation layer needed from the application. FINRA also discusses how Amazon DocumentDB met their high SLA and security requirements.

DAT326 – Amazon DocumentDB Deep Dive

In this talk, I give an overview of why developers choose the document model, common customer use cases, examples of when you should consider an alternative data store, and how Amazon DocumentDB solves specific customer problems related to managing and scaling document databases. The highlight of the session is how Fulfillment by Amazon (FBA) reduced their database infrastructure footprint from 96 instances of their previous solution to just two Amazon DocumentDB instances. Migrating to Amazon DocumentDB simultaneously increased performance by 66 percent and reduced costs by 45 percent. FBA’s example shows the result of using the right database for the job. FBA also provides a great set of lessons learned for scaling reads and writes, and indexing strategies with Amazon DocumentDB. The talk also includes three live demos where I create a snapshot backup of a 1.5 TB cluster in less than one minute, add a new instance to the same 1.5 TB cluster in five minutes, and query data in Amazon DocumentDB with a SQL interface using the new federated query capabilities in Amazon Athena.

DAT338 – Hands-on workshop: How to migrate to Amazon DocumentDB

In this workshop, Daniel Bento and Hugo Rozestraten, AWS Solution Architects, walk you through what you need to think about before, during, and after migrating to Amazon DocumentDB. This workshop teaches you how to migrate workloads to Amazon DocumentDB, and also demonstrates using AWS Database Migration Service (DMS). Although this session was not recorded, you can find the step-by-step workshop content below.

DAT209-L – Leadership session: AWS purpose-built databases

Shawn Bice, AWS VP of Databases, provides an overview of AWS purpose-built databases and how customers use these databases to simplify their existing applications and to build net-new applications. In this talk, Tobias Ternström, Director of Product Management for Amazon RDS and Aurora, and I also discuss the data models, access patterns, and customer use cases for relational, document, key-value, graph, ledger, and in-memory databases. I discuss why Liberty Mutual chose to model their application data using JSON, and why using Amazon DocumentDB is a natural choice to be able to store, query, and index JSON data, enabling Liberty Mutual’s developers to move faster.

DAT357-R – Chalk talk: Build flexible data architecture with Amazon DocumentDB change streams

In this chalk talk, Jeff Duffy and Andrew Whitaker, AWS principal engineer, walk you through how to use Amazon DocumentDB change streams to stream data to Amazon S3 for analytics, to Amazon Elasticsearch Service for enhanced search capabilities, and to Amazon ElastiCache to pre-warm a Redis cache. Though this chalk talk was not recorded on video, Jeff recorded a similar session on Twitch a few weeks before re:Invent 2019.

DAT212-L – Leadership session: Database and analytics

Raju Gulabani, AWS VP of Databases, Analytics, and Machine Learning, provides a concise overview of the entire AWS database and analytics landscape that includes new announcements, how AWS thinks about building services, and prescriptive guidance about when to use each service. Raju has led this space at AWS for nearly 10 years, and he offers a rare perspective born from building many of these services. In this talk, Raju discusses why AWS built Amazon DocumentDB.

To learn more about Amazon DocumentDB, see the product page and documentation.

We look forward to seeing you at AWS re:Invent 2020!


About the Author


Joseph Idziorek is a Principal Product Manager at Amazon Web Services.