PreSto: An In-Storage Data Preprocessing System for Training Recommendation Models

TL;DR

PreSto is a storage-centric data preprocessing system for recommendation models that significantly speeds up preprocessing, reduces costs, and improves energy efficiency by offloading operations to in-storage processing units.

Contribution

PreSto introduces a novel in-storage processing approach to overcome CPU bottlenecks in RecSys data preprocessing, enhancing speed and efficiency.

Findings

9.6× faster preprocessing time

4.3× better cost-efficiency

11.3× improved energy efficiency

Abstract

Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and feed them to the GPU for training in a seamless manner. To sustain high training throughput, state-of-the-art solutions reserve a large fleet of CPU servers for preprocessing which incurs substantial deployment cost and power consumption. Our characterization reveals that prior CPU-centric preprocessing is bottlenecked on feature generation and feature normalization operations as it fails to reap out the abundant inter-/intra-feature parallelism in RecSys preprocessing. PreSto is a storage-centric preprocessing system leveraging In-Storage Processing (ISP), which offloads the bottlenecked preprocessing operations to our ISP units. We show that PreSto outperforms the baseline CPU-centric system with a $9.6\times$ speedup in end-to-end preprocessing time, $4.3\times$ enhancement in cost-efficiency, and $11.3\times$ improvement in energyefficiency on average for production-scale RecSys preprocessing.