Work

Frugal: Efficient and Economic Embedding Model Training with Commodity GPUs

Year: 2025

Type: conference-paper

Abstract: Embedding models show superiority in learning representations of massive ID-type features in sparse learning scenarios such as recommendation systems (e.g., user/item IDs) and graph learning (e.g., node/edge IDs). Commodity GPUs are highly favored for their cost-efficient computing power, which is ideally suited for the low computing demand of memory-intensive embedding models. However, directly running embedding model training on commodity GPUs yields poor performance because of their deficient communication resources (including low communication bandwidth and no PCIe P2P support).

Source: ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS)

DOI: 10.1145/3669940.3707245

Authors: Minhui Xie, Shaoxun Zeng, Hao Guo, Shiwei Gao, Youyou Lu

Institutions: Renmin University of China, Tsinghua University

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: Universitas Brawijaya

Award: G4625528983