Insider Inc. went to MongoDB World — here’s what we learned

We recently attended MongoDB World in New York City, where we had a great time and enjoyed our conversations with the Mongo team as well as industry peers. We also left excited about all the new announcements we heard at the event.

Here are some of the most useful things we learned: (We’re excited to begin implementing them ourselves in the coming months.)

• A $merge pipeline operator so views can be materialized on demand with far less load on the database
• Major enhancements to the MongoDB query language, including more expressive updates and wildcard indexes
• Client-side field level encryption, which gives customers complete control over who can see what data is in their clusters
• Retryable reads which, along with retryable, reduces the complexity of writing code to handle transient cluster failures
• Atlas Full-Text Search: rich text search capabilities against fully managed databases in MongoDB Atlas, with no additional systems or infrastructure to provision, learn, or manage
• Atlas Data Lake: Allowing customers to quickly and easily query data in many common formats on Amazon S3 using the MongoDB Query Language (MQL)

We attended some great sessions and learned a lot about improvements in 4.2 Pipeline, including materialized views, Tips and Tricks for Querying and Indexing, Comparing performance between RDBMS (MySQL) and MongoDB Aggregation.

The Methodology to Data Modeling for MongoDB talk was especially insightful. We learned that modeling data by first understanding use cases for reading and writing can greatly help performance, including making decisions on when to link vs embed sub documents. As we continue to improve our Product Library, Full Text Search should be a great help for our editors and users.

We saw great demos like a Machine Learning project using Google’s Vision API to store insights on photos without metadata assigned to them. The demo identified a wide range of images, from icons like the Statue of Liberty to car crashes. This has great potential for our editors.

Developers at the event were happy to answer our questions, but we were also excited to also contribute, for example looking at this repository we can add Go Lang for exporting our aggregations pipelines in compass.

For anyone interested in what we’re working on — or who has an interest in joining our team — please reach out to us!