All around the world, businesses and organizations are becoming increasingly data driven, products and services are built more and more around intelligence derived from data, and the need for reliable and efficient data storage and processing at a global scale is becoming even more critical. Modern data infrastructure architectures have emerged from years of evolution in analytical and transactional data systems, along with a continuous infusion of capabilities stemming from new use cases and new data processing paradigms. Tightly coupled data warehouses are being replaced by more flexible ecosystems built around low-cost globally available storage and open file formats; data science and machine learning workloads are increasingly sharing the same infrastructure as analytical workloads; transactional systems and key-value stores are exploring ways to preserve consistency, reliability, and performance while operating efficiently at global scale. Yet, despite all these efforts and progress, many challenges still remain as the data management community is seeking out the defining characteristics of next-generation data infrastructure.
Facebook has had a long history of making contributions to the data management space – Hive, Presto, RocksDB, MyRocks all being examples of innovative work that started within the company. The scale at which we run and the unique constraints of our workloads make many existing solutions infeasible and provide a perspective that leads to new ideas. As we continue to build and evolve our data infrastructure, we are focused on a number of problems. These range from techniques to optimize CPU usage (and thus power consumption during large scale query processing) to strategies to optimize physical layouts and data transfer bandwidth, and from techniques to address the challenges rising from data storage and processing across widely separated data centers to novel approaches in converging data wrangling, machine learning, and analytics. Since guaranteeing correctness is a key requirement for our data storage and processing systems, we also remain focused in systems for testing and verification. Despite the unique constraints of our workloads, a lot of these problems are common in the industry and we believe that there is a lot to be gained by collaborating with academia in this area.
To foster further innovation in this area, and to deepen our collaboration with academia, Facebook is pleased to invite faculty to respond to this call for research proposals pertaining to the aforementioned topics. We anticipate awarding a total of 10 awards, each in the $50,000 range. Payment will be made to the proposer’s host university as an unrestricted gift. In addition, PIs and Co-PIs on the winning proposals will be automatically granted access to CrowdTangle, a public insights tool from Facebook that makes it easy to follow, analyze, and report on what’s happening with public content on social media. Learn more about CrowdTangle here.