Deep Learning Inference in Facebook Data Centers: Characterization, Performance Optimizations and Hardware Implications
This work addresses optimization challenges for deep learning inference in data centers, but it is incremental as it builds on existing systems and focuses on specific applications.
The paper characterizes deep learning models used in Facebook's social network services, identifies performance optimizations and hardware limitations, and suggests future improvements for inference hardware and co-design.
The application of deep learning techniques resulted in remarkable improvement of machine learning models. In this paper provides detailed characterizations of deep learning models used in many Facebook social network services. We present computational characteristics of our models, describe high performance optimizations targeting existing systems, point out their limitations and make suggestions for the future general-purpose/accelerated inference hardware. Also, we highlight the need for better co-design of algorithms, numerics and computing platforms to address the challenges of workloads often run in data centers.