Deep Learning Recommendation Model (DLRM)

DLRM Training

• Heterogeneous Acceleration Pipeline for Recommendation System Training (ISCA 2024) [arXiv]

  • UBC & GaTech

  • Hotline: a runtime framework.

  • Utilize CPU main memory for non-popular embeddings and GPUs’ HBM for popular embeddings.

  • Fragment a mini-batch into popular and non-popular micro-batches (μ-batches).

• Accelerating Neural Recommendation Training with Embedding Scheduling (NSDI 2024) [Paper] [Slides] [Code]

  • HKUST

  • Herald: an adaptive location-aware inputs allocator to determine where embeddings should be trained and an optimal communication plan generator to determine which embeddings should be synchronized.

• Bagpipe: Accelerating Deep Recommendation Model Training (SOSP 2023) [Paper]

  • UW-Madison & UChicago

DLRM Inference

• DisaggRec: Architecting Disaggregated Systems for Large-Scale Personalized Recommendation (arXiv 2212.00939) [Personal Notes] [Paper]

  • Meta AI & WashU & UPenn & Cornell & Intel

  • Disaggregated system; decouple CPUs and memory resources; partition embedding tables.

Pruning

• AdaEmbed: Adaptive Embedding for Large-Scale Recommendation Models (OSDI 2023) [Paper]

  • UMich SymbioticLab & Meta

  • In-training pruning.

GPU Cache

• UGache: A Unified GPU Cache for Embedding-based Deep Learning (SOSP 2023) [Personal Notes] [Paper]

  • SJTU

  • A unified multi-GPU cache system.

  • Used for GNN training and DLR inference.

• EVStore: Storage and Caching Capabilities for Scaling Embedding Tables in Deep Recommendation Systems (ASPLOS 2023) [Personal Notes] [Paper] [Code]

  • UChicago & Beijing University of Technology & Bandung Institute of Technology, Indonesia & Seagate Technology & Emory

  • A caching layer optimized for embedding access patterns.

Model Update

• Ekko: A Large-Scale Deep Learning Recommender System with Low-Latency Model Update (OSDI 2022) [Paper]

  • Tencent & Edinburgh

  • P2P model update dissemination.

Acronyms

• DLRM: Deep Learning Recommendation Model