Explore the latest research from Facebook

All Publications

November 1, 2021 Saeed Rashidi, Pallavi Shurpali, Srinivas Sridharan, Naader Hassani, Dheevatsa Mudigere, Krishnakumar Nair, Misha Smelyanskiy, Tushar Krishna
Paper

Scalable Distributed Training of Recommendation Models: An ASTRA-SIM + NS3 case-study with TCP/IP transport

In this paper, we port the previously proposed ASTRASIM simulation platform on top of the versatile NS3 network simulator by introducing a portable network interface for ASTRA-SIM.
Paper
August 23, 2021 Satyajeet Singh Ahuja, Varun Gupta, Vinayak Dangui, Soshant Bali, Abishek Gopalan, Hao Zhong, Petr Lapukhov, Yiting Xia, Ying Zhang
Paper

Capacity-Efficient and Uncertainty-Resilient Backbone Network Planning with Hose

Our core design involves heuristic algorithms to select Hose-compliant traffic matrices and cross-layer optimization between the optical and IP networks. We evaluate the system performance in production and share insights from years of production experience.
Paper
August 23, 2021 Hang Zhu, Varun Gupta, Satyajeet Singh Ahuja, Yuandong Tian, Ying Zhang, Xin Jin
Paper

Network Planning with Deep Reinforcement Learning

In this paper, we propose NeuroPlan, a deep reinforcement learning (RL) approach to solve the network planning problem. This problem involves multi-step decision making and cost minimization, which can be naturally cast as a deep RL problem. We develop two important domain-specific techniques.
Paper
July 30, 2021 Zhizhen Zhong, Manya Ghobadi, Alaa Khaddaj, Jonathan Leach, Yiting Xia, Ying Zhang
Paper

ARROW: Restoration-Aware Traffic Engineering

This poses a cross-layer challenge from the Traffic Engineering (TE) perspective that has not been considered before: “Which IP links should be restored and by how much to best match the TE objective?”
Paper
July 9, 2021 Zhizhen Zhong, Manya Ghobadi, Maximilian Balandat, Sanjeevkumar Katti, Abbas Kazerouni, Jonathan Leach, Mark McKillop, Ying Zhang
Paper

BOW: First Real-World Demonstration of a Bayesian Optimization System for Wavelength Reconfiguration

We demonstrate a practical Bayesian Optimization system for wavelength reconfiguration at Facebook backbone. Our system uses a firewall for safe deployment.
Paper
June 16, 2021 Erik Boch, Julius Kusuma
Paper

Combining CLOS and NLOS Microwave Backhaul to Help Solve the Rural Connectivity Challenge

The hybrid combination of clear-line-of-sight (CLOS) and non-line-of-sight (NLOS) techniques can often be employed to create a practical and cost-effective networking solution for rural and deep-rural connectivity challenges.
Paper
June 14, 2021 Anton Shkel, Alireza Mehrabani, Julius Kusuma
Paper

A Configurable 60GHz Phased Array Platform for Multi-Link mmWave Channel Characterization

We present the Terragraph Sounder platform, developed to promote and enable research in mmWave propagation, systems, and networks, and is actively employed in research studies in over 13 institutions.
Paper
June 14, 2021 Saeed Rashidi, Matthew Denton, Srinivas Sridharan, Sudarshan Srinivasan, Amoghavarsha Suresh, Jade Nie, Tushar Krishna
Paper

Enabling Compute-Communication Overlap in Distributed Deep Learning Training Platforms

This work makes two key contributions. First, via real system measurements and detailed modeling, we provide an understanding of compute and memory bandwidth demands for DL compute and comms. Second, we propose a novel DL collective communication accelerator called Accelerator Collectives Engine (ACE) that sits alongside the compute and networking engines at the accelerator endpoint.
Paper
June 11, 2021 Ryan M. Dreifuerst, Samuel Daulton, Yuchen Qian, Paul Parayil Varkey, Maximilian Balandat, Sanjay Kasturia, Anoop Tomar, Ali Yazdan Panah, Vish Ponnampalam, Robert W. Heath Jr
Paper

Optimizing Coverage and Capacity in Cellular Networks using Machine Learning

In this paper, we develop and compare two approaches for maximizing coverage and minimizing interference by jointly optimizing the transmit power and downtilt (elevation tilt) settings across sectors. To evaluate different parameter configurations offline, we construct a realistic simulation model that captures geographic correlations.
Paper
April 12, 2021 Yiting Xia, Ying Zhang, Zhizhen Zhong, Guanqing Yan, Chiun Lin Lim, Satyajeet Singh Ahuja, Soshant Bali, Alexander Nikolaidis, Kimia Ghobadi, Manya Ghobadi
Paper

A Social Network Under Social Distancing: Risk-Driven Backbone Management During COVID-19 and Beyond

This paper presents Facebook’s risk-driven backbone management strategy to ensure high service performance throughout the COVID-19 pandemic.
Paper