A Bag of Tricks for Scaling CPU-based Deep FFMs to more than 300m Predictions per Second

Blaž Škrlj, Benjamin Ben-Shalom, Grega Gašperšič, Adi Schwartz, Ramzi Hoseisi +3 more
7/14/2024
cs.LGcs.AIcs.IR

No confident code implementation found for this paper. 1 possible match can be shown below.

Abstract

Field-aware Factorization Machines (FFMs) have emerged as a powerful model for click-through rate prediction, particularly excelling in capturing complex feature interactions. In this work, we present an in-depth analysis of our in-house, Rust-based Deep FFM implementation, and detail its deployment on a CPU-only, multi-data-center scale. We overview key optimizations devised for both training and inference, demonstrated by previously unpublished benchmark results in efficient model search and online training. Further, we detail an in-house weight quantization that resulted in more than an order of magnitude reduction in bandwidth footprint related to weight transfers across data-centres. We disclose the engine and associated techniques under an open-source license to contribute to the broader machine learning community. This paper showcases one of the first successful CPU-only deployments of Deep FFMs at such scale, marking a significant stride in practical, low-footprint click-through rate prediction methodologies.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. 1 weaker match is hidden by default — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Extreme Learning Machines for Fast Training of Click-Through Rate Prediction Models

81% match2024

ML-DCN: Masked Low-Rank Deep Crossing Network Towards Scalable Ads Click-through Rate Prediction at Pinterest

79% match2026

Efficient Fine-Grained GPU Performance Modeling for Distributed Deep Learning of LLM

79% match2025

CubicML: Automated ML for Large ML Systems Co-design with ML Prediction of Performance

78% match2024

BOLT: An Automated Deep Learning Framework for Training and Deploying Large-Scale Search and Recommendation Models on Commodity CPU Hardware

78% match2023

Low Rank Field-Weighted Factorization Machines for Low Latency Item Recommendation

78% match1 repo62024

Cite this paper

@article{krlj2024tricks,
  title  = {A Bag of Tricks for Scaling CPU-based Deep FFMs to more than 300m Predictions per Second},
  author = {Blaž Škrlj and Benjamin Ben-Shalom and Grega Gašperšič and Adi Schwartz and Ramzi Hoseisi and Naama Ziporin and Davorin Kopič and Andraž Tori},
  year   = {2024},
  eprint = {2407.10115},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.10115v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.10115.svg)](https://codeofpaper.com/paper/2407.10115)

Discussion