Low-Precision Hardware Architectures Meet Recommendation Model Inference at Scale

TL;DR

This paper discusses how low-precision hardware architectures can be effectively used for large-scale recommendation model inference, highlighting strategies, optimizations, and tools developed to maintain accuracy and efficiency in production environments.

Contribution

The paper presents practical methods and tools for adapting recommendation models to low-precision hardware, enabling deployment of more complex models with improved efficiency in industry.

Findings

Low-precision architectures can handle complex recommendation models effectively.

Significant hardware and software optimizations are necessary for accuracy preservation.

Deployment of larger models leads to 5X capacity savings in data centers.

Abstract

Tremendous success of machine learning (ML) and the unabated growth in ML model complexity motivated many ML-specific designs in both CPU and accelerator architectures to speed up the model inference. While these architectures are diverse, highly optimized low-precision arithmetic is a component shared by most. Impressive compute throughputs are indeed often exhibited by these architectures on benchmark ML models. Nevertheless, production models such as recommendation systems important to Facebook's personalization services are demanding and complex: These systems must serve billions of users per month responsively with low latency while maintaining high prediction accuracy, notwithstanding computations with many tens of billions parameters per inference. Do these low-precision architectures work well with our production recommendation systems? They do. But not without significant effort. We share in this paper our search strategies to adapt reference recommendation models to low-precision hardware, our optimization of low-precision compute kernels, and the design and development of tool chain so as to maintain our models' accuracy throughout their lifespan during which topic trends and users' interests inevitably evolve. Practicing these low-precision technologies helped us save datacenter capacities while deploying models with up to 5X complexity that would otherwise not be deployed on traditional general-purpose CPUs. We believe these lessons from the trenches promote better co-design between hardware architecture and software engineering and advance the state of the art of ML in industry.