Research Area
Year Published

125 Results

February 16, 2019

Machine Learning at Facebook: Understanding Inference at the Edge

IEEE International Symposium on High-Performance Computer Architecture (HPCA)

This paper takes a data-driven approach to present the opportunities and design challenges faced by Facebook in order to enable machine learning inference locally on smartphones and other edge platforms.

By: Carole-Jean Wu, David Brooks, Kevin Chen, Douglas Chen, Sy Choudhury, Marat Dukhan, Kim Hazelwood, Eldad Isaac, Yangqing Jia, Bill Jia, Tommer Leyvand, Hao Lu, Yang Lu, Lin Qiao, Brandon Reagen, Joe Spisak, Fei Sun, Andrew Tulloch, Peter Vajda, Xiaodong Wang, Yanghan Wang, Bram Wasti, Yiming Wu, Ran Xian, Sungjoo Yoo, Peizhao Zhang
December 7, 2018

Rethinking floating point for deep learning

Systems for Machine Learning Workshop at NeurIPS 2018

We improve floating point to be more energy efficient than equivalent bit width integer hardware on a 28 nm ASIC process while retaining accuracy in 8 bits with a novel hybrid log multiply/linear add, Kulisch accumulation and tapered encodings from Gustafson’s posit format.

By: Jeff Johnson
December 3, 2018

Training with Low-precision Embedding Tables

Systems for Machine Learning Workshop at NeurIPS 2018

In this work, we focus on building a system to train continuous embeddings in low precision floating point representation. Specifically, our system performs SGD-style model updates in single precision arithmetics, casts the updated parameters using stochastic rounding and stores the parameters in half-precision floating point.

By: Jian Zhang, Jiyan Yang, Hector Yuen
November 24, 2018

Deep Learning Inference in Facebook Data Centers: Characterization, Performance Optimizations and Hardware Implications


The application of deep learning techniques resulted in remarkable improvement of machine learning models. In this paper we provide detailed characterizations of deep learning models used in many Facebook social network services.

By: Jongsoo Park, Maxim Naumov, Protonu Basu, Summer Deng, Aravind Kalaiah, Daya Khudia, James Law, Parth Malani, Andrey Malevich, Satish Nadathur, Juan Pino, Martin Schatz, Alexander Sidorov, Viswanath Sivakumar, Andrew Tulloch, Xiaodong Wang, Yiming Wu, Hector Yuen, Utku Diril, Dmytro Dzhulgakov, Kim Hazelwood, Bill Jia, Yangqing Jia, Lin Qiao, Vijay Rao, Nadav Rotem, Sungjoo Yoo, Mikhail Smelyanskiy
November 3, 2018

RacerD: Compositional Static Race Detection

Conference on Object-Oriented Programming, Systems, Languages & Applications (OOPSLA)

Automatic static detection of data races is one of the most basic problems in reasoning about concurrency. We present RacerD—a static program analysis for detecting data races in Java programs which is fast, can scale to large code, and has proven effective in an industrial software engineering scenario.

By: Sam Blackshear, Nikos Gorogiannis, Peter O'Hearn, Ilya Sergey
October 8, 2018

Sharding the Shards: Managing Datastore Locality at Scale with Akkio

USENIX Symposium on Operating Systems Design and Implementation (OSDI)

Akkio is a locality management service layered between client applications and distributed datastore systems. It determines how and when to migrate data to reduce response times and resource usage. Akkio primarily targets multi-datacenter geo-distributed datastore systems.

By: Muthukaruppan Annamalai, Kaushik Ravichandran, Harish Srinivas, Igor Zinkovsky, Luning Pan, Tony Savor, David Nagle, Michael Stumm
September 23, 2018

From Start-ups to Scale-ups: Opportunities and Open Problems for Static and Dynamic Program Analysis

IEEE International Working Conference on Source Code Analysis and Manipulation (SCAM)

This paper describes some of the challenges and opportunities when deploying static and dynamic analysis at scale, drawing on the authors’ experience with the Infer and Sapienz Technologies at Facebook, each of which started life as a research-led start-up that was subsequently deployed at scale, impacting billions of people worldwide.

By: Mark Harman, Peter O'Hearn
August 27, 2018

Providing Streaming Joins as a Service at Facebook

International Conference on Very Large Data Bases (VLDB)

This paper describes an end-to-end streaming join service that addresses the challenges above through a streaming join operator that uses an adaptive stream synchronization algorithm that is able to handle the different distributions we observe in real-world streams regarding their event times.

By: Gabriela Jacques da Silva, Ran Lei, Luwei Cheng, Guoqiang Jerry Chen, Kuen Ching, Tanji Hu, Yuan Mei, Kevin Wilfong, Rithin Shetty, Serhat Yilmaz, Anirban Banerjee, Benjamin Heintz, Shridhar Iyer, Anshul Jaiswal