The Materials Project Scales to 616,000+ users with MongoDB and AWS

OUR SOLUTION

A scalable cloud architecture for scientific innovation

The Materials Project team released a major architectural overhaul in 2022, modernizing its platform to support growing global demand. This new framework included updates to data pipelines, cloud resources, and web applications. MP transitioned to a microservices-based cloud architecture and adopted containerized environments to ensure uninterrupted service. To support this modernization, the MP team deployed MongoDB Atlas on AWS for a fully managed, cloud-based document database service that combines MongoDB’s flexible architecture with the global scale and reliability of AWS. “MongoDB Atlas on AWS helps us build the platform we need to handle data and traffic during exponential growth, so we can focus on delivering value to our scientific community," says Patrick Huck, Ph.D., Senior Computing Engineer at the Lawrence Berkeley National Laboratory and Technical Lead of the Materials Project.

MongoDB Atlas on AWS eliminated the need for manual database operations and gave the team confidence that its platform could scale with rising demand without compromising performance. Designed for distributed environments, MongoDB Atlas on AWS offers workload isolation, dynamic resource allocation, and global data distribution with low latency and high availability. It also supports structured and unstructured data, making it ideal for MP’s collaborative model. “When you’re distributing hundreds of thousands of data points to a community around the world, data management can be difficult and time-consuming,” says Andrew Davidson, Senior Vice President of Products at MongoDB. “MongoDB Atlas on AWS is designed so you can stop worrying about the complexity and pains of handling and serving data.”

OUTCOME

Exponential growth accelerates innovation

Since modernizing with MongoDB Atlas on AWS, the MP platform has become a global gold standard for open-access scientific data repositories. As of April 2025, the Materials Project supports a rapidly growing community from 178 countries with more than 616,000 registered users—a 130% increase since the launch of the new framework in May 2022. In the past two years, the platform has managed over 400 terabytes of data transferred in more than 500 million API and Amazon S3 requests, all with 99.98% uptime and without an increase in staff. The streamlined infrastructure enables MP to maintain 24/7 availability while serving the needs of scientists, professors, students, and companies of all sizes.

The system’s flexibility and performance have also enabled deeper collaboration across the global materials science community. Researchers can access, analyze, and contribute to MP’s data through intuitive tools and open APIs, and MongoDB Atlas on AWS simplifies the integration of that user-contributed information. A wide variety of properties for more than 200,000 materials are now available on MP. Partnerships—including the Open Catalysts Project at Meta—have used the platform’s data and infrastructure to drive large-scale, AI-powered discoveries. At Berkeley Lab’s autonomous materials research laboratory, A-Lab, MP data powered an AI-driven synthesis process that successfully produced 41 novel compounds in just 17 days. This outcome would have taken more than a year—or might not have been possible at all—using traditional methods.

Looking ahead, MP is emerging as a key catalyst for innovation and aims to be the starting point for materials researchers. With a scalable, cost-efficient foundation in place, the team can focus on building the future of materials research, supporting faster innovation in energy technology, consumer electronics, and more. “We aim to be the premier platform that scientists turn to for materials research,” says Dr. Huck. “MongoDB Atlas on AWS plays an important role in helping us deliver on that vision.”

To learn how you can build modern, cloud-based solutions, visit MongoDB Atlas on AWS.