Meta Research

Meta Core Systems researchers and engineers design and build the distributed systems that power Meta’s infrastructure. Our work spans across the engineering spectrum of research, development, deployment, and production as we ensure that our systems run efficiently, reliably, and securely across millions of machines in tens of geo-replicated data center regions.

Core Systems performs forward-looking research in the area of distributed systems and architecture at a global scale. Billions of people rely on the services we build and manage to connect and communicate. Throughout the lifecycle of these distributed services, we encounter fundamental research challenges in multiple areas, including capacity management, configuration management, cluster management, deployment, distributed tracing, efficiency, fault tolerance, monitoring, performance, power management, reliability, routing, scalability, service discovery, and storage systems.

We build a strong collaboration pipeline with key experts in academia through Distributed Systems PhD fellowships, requests for proposals, faculty summits, as well as internships and visiting researcher programs.

In recent years, we’ve published work on cluster management (Twine, OSDI 2020), configuration management (Configerator, SOSP 2015), fault tolerance (Kraken, OSDI 2016; Maelstrom, OSDI 2018; Taiji, SOSP 2019), tracing (Canopy, SOSP 2017), data center power management (Dynamo, ISCA 2016), and consensus protocol (Delos, OSDI 2020). View our Publications for a list of all our published research.