Networking is important to today’s large-scale distributed systems that power Facebook’s family of apps, used by billions of people. At Facebook, we focus on making networking technology more flexible and scalable by rethinking traditional approaches and designs, and collaborating with research and open source communities to develop an ecosystem of open hardware and software.

To foster further innovation in networking and deepen our collaboration with academia, we launched the Networking Systems request for proposals (RFP) in April and invited faculty and graduate students to respond.

Networking systems research covers a wide range of topics, and this year we focused our interest on proposals that fell into these two categories:

• Improving network efficiency with intelligent control, such as traffic engineering, routing, cache management, and congestion control
• Programmable switches and their applications, such as troubleshooting, hardware acceleration, and new routing protocols

We received 71 proposals from 17 countries. The six research award winners of the Networking Systems RFP, listed below, will receive an unrestricted gift of $50,000 each. These proposals could have great potential in improving global backbone, data centers, and edge networks. Facebook is committed to advancing computer networking research, and we are excited to see the outcomes of these research proposals.

Research award winners

Principal investigators are listed first unless otherwise noted.

Automating the Analysis of Transport and Application Protocols
Italo Fernando Scota Cunha and Elverton Carvalho Fazzion (Universidade Federal de Minas Gerais)

Making In-Network Compute Systems Tolerant of Switch Failures
Srinivasan Seshan (Carnegie Mellon University)

Safe and Efficient Packet Processing with Verification of eBPF Programs
Srinivas Narayana Ganapathy and Santosh Nagarakatte (Rutgers University)

Semantic Management of Packet Drops
Radhika Mittal (University of Illinois at Urbana-Champaign)

Record & Replay: Framework for Network-Wide Monitoring and Debugging
Mun Choon Chan, Nishant Shyamal Budhdev, Pravein Govindan Kannan, and Raj Joshi (National University of Singapore)

Toward Highly Reliable, Programmable, and Efficient Network Control
Y. Richard Yang, Geng Li, and Qiao Xiang (Yale University)

Runners-up

In addition to the research award winners, we would like to acknowledge the following ten runners-up and share their exciting work.

Data-Driven Cache Management Algorithms
Mohammad Alizadeh (Massachusetts Institute of Technology)

Accelerating Database Queries with Programmable Switches
Minlan Yu (Harvard University)

A High-Speed Stateless Load-Balancer with Perfect Connection-Affinity
Marco Chiesa and Tom Barbette (KTH Royal Institute of Technology)

LFC: Learning-Based Functional Caching in Data Processing Networks
Tian Lan and Vaneet Aggarwal (George Washington University)

Real-Time Telemetry and Inference with Programmable Networks
Nick Feamster (University of Chicago)

Application-Specific Browser Network Stacks
Ravi Netravali (University of California, Los Angeles)

NetLock: Fast, Centralized Lock Management with Programmable Switches
Xin Jin (Johns Hopkins University)

Data-Driven Web Optimizations for the Next Billion Users
Theophilus Benson (Brown University)

Efficient and Flexible Virtualization of Programmable Data Planes
Fernando Manuel Valente Ramos and Salvatore Signorello (Faculdade de Ciências, Universidade de Lisboa)

Taurus: An Intelligent Switch Data Plane
Oyekunle A. Olukotun and Muhammad Shahbaz (Stanford University)

—

Thank you to everyone who submitted a proposal, and congratulations to the winners. To view our currently open research awards, and to subscribe to our RFP newsletter, visit our research awards page.