June 22, 2023 | Updated: July 30, 2023
Today at MongoDB.local NYC, we unveiled several new capabilities across our developer data platform to help users and customers build, iterate, and scale their applications with MongoDB.
Development teams are being asked to deliver compelling and differentiated user experiences that are faster and smarter than ever before. At the same time, they must do so as quickly and as efficiently as possible.
MongoDB’s developer data platform is essential for teams who strive to innovate quickly and efficiently. It allows developers to support a wide variety of application use cases in their organization through a unified API and eliminates the need to bolt-on, learn, and maintain separate database point solutions.
Expanding the range of modern applications you can build on MongoDB
A new generation of AI-powered experiences are being developed where vectors serve as a foundational element that makes these applications possible. Vectors are mathematical representations of the characteristics of unstructured data — including text, images, videos, audio files, and more — and occupy n-dimensional space where n is the number of characteristics in the dataset. Whether data is similar or not is based on the distance between vectors in this n-dimensional vector space. A vector database allows users to query vectors to determine what’s similar or related, without relying on keyword matching. Announced today, Atlas Vector Search allows you to store, index, and query vectors alongside your operational and transactional data in documents without the overhead of adding, learning, and maintaining another database system. The addition of vector search to MongoDB Atlas enables teams to deliver more relevant, context-aware results to end-users, including the ability to augment applications built on Large Language Models (LLMs) with proprietary data to improve accuracy and performance. Atlas Vector Search is available today in public preview.
Applications today also need to be more real-time than ever, but processing streams of data and bringing them into applications is complex and challenging. Most organizations introduce a stream processing point solution with a different set of APIs, drivers, and tools that create a fragmented developer experience, operational complexity, and additional cost. Coming soon in private preview, Atlas Stream Processing will transform how development teams build event-driven applications. Developers can use the same query language and flexible document data model to work with both streaming data and their data in the database for use cases ranging from monitoring network traffic for intrusions to conducting live route planning based on current road conditions.
Companies such as Albertsons, Glassdoor, and Anywhere Real Estate, the largest franchisor of residential real estate brands in the world, depend on Atlas Search to deliver full-text functionality in their applications without having to deploy and sync data from their database to a separate search engine. With new search query analytics, developers will gain insight into what their end users are searching for, enabling them to better refine and customize their search logic. In addition, Atlas Search indexes can now be created and managed in language drivers (starting with Node.js), MongoDB Compass, and the MongoDB Shell, making it easy for developers who prefer to work with their indexes programmatically. And finally, dedicated search nodes were announced today in private preview, which will enable teams to independently scale and optimize resources for their search workloads for improved performance at scale, higher availability, and faster index builds.
Improving the foundation of performance, scale and security
The work we’re putting into improving the performance, scalability, and security of our developer data platform is just as important as increasing its range of functional capabilities.
Starting with MongoDB 7.0, improvements to query execution will reduce the number of disk reads, compute resources, and memory required to execute certain queries, delivering improved performance and more efficient resource consumption. In particular, the new query execution strategy will speed up grouping and reshaping documents, filtering and sorting documents, and $lookups used for joining data across collections.
Last year we introduced queryable encryption in preview. This industry-first searchable encryption scheme allows users to encrypt sensitive data fields — such as personally identifiable information — client-side, and store that data as fully randomized encrypted data in the database, all while preserving the ability to run expressive queries. With MongoDB 7.0, queryable encryption will support equality search, with support for range, prefix, suffix, and substring queries to follow.
MongoDB 7.0 will be generally available later this summer.
Continued focus on a first-class developer experience
Few things are more critical to software development than ensuring a top-notch developer and operational experience. We remain dedicated to improving the usability of our products and expanding the set of tools available to developers for building with MongoDB.
Kotlin is emerging as a popular language for both mobile development and server-side development. Today we unveiled a new official Kotlin driver, allowing Kotlin developers to build applications on MongoDB with confidence, knowing we are committed to supporting this fast-growing language community. We also released PyMongoArrow, a new library for developers and data analysts to easily export data in MongoDB to Python-based analytics stacks, including Apache Arrow, Pandas, and NumPy.
As more companies optimize their DevOps pipeline, tooling that enables programmatic automation can greatly improve efficiency and productivity. Developers can now provision resources on MongoDB Atlas using the Amazon Web Services Cloud Development Kit (AWS CDK) in C#, Go, Java, and Python, as well as with Node.js and Typescript. Companies who leverage Kubernetes can use the MongoDB Atlas CLI to install the Atlas Kubernetes Operator and export existing deployments for simpler infrastructure management and provisioning.
Path to application modernization
Application modernization continues to be a key investment for many enterprises to unlock rapid software iteration to meet ever-evolving application requirements. MongoDB Relational Migrator, now generally available, helps accelerate and de-risk migrations from common relational databases — including Oracle, SQL Server, MySQL, and PostgreSQL — to MongoDB. Not only does this tool handle the data migration itself, it also allows migration teams to view MongoDB data modeling recommendations, define schema changes with a visual interface, and generate application code in their programming language or framework to get a head start on refactoring their applications to reflect the newly designed MongoDB schema.