paper-with-me

홈 › Papers

Accelerating Communication in Deep Learning Recommendation Model Training with Dual-Level Adaptive Lossy Compression

2024-07-05 · Hao Feng, Boyuan Zhang, Fanjiang Ye, Min Si, Ching-Hsiang Chu, Jiannan Tian, Chunxing Yin, Summer Deng, Yuchen Hao, Pavan Balaji, Tong Geng, Dingwen Tao

DLRM is a state-of-the-art recommendation system model that has gained widespread adoption across various industry applications. The large size of DLRM models, however, necessitates the use of multiple devices/GPUs for efficient training. A significant bottleneck in this process is the time-consuming all-to-all communication required to collect embedding data from all devices. To mitigate this, we introduce a method that employs error-bounded lossy compression to reduce the communication data size and accelerate DLRM training. We develop a novel error-bounded lossy compression algorithm, informed by an in-depth analysis of embedding data features, to achieve high compression ratios. Moreover, we introduce a dual-level adaptive strategy for error-bound adjustment, spanning both table-wise and iteration-wise aspects, to balance the compression benefits with the potential impacts on accuracy. We further optimize our compressor for PyTorch tensors on GPUs, minimizing compression overhead. Evaluation shows that our method achieves a 1.38$\times$ training speedup with a minimal accuracy impact.

📄 PDF Abstract BibTeX arXiv:2407.04272

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

ProactivePIM: Accelerating Weight-Sharing Embedding Layer with PIM for Scalable Recommendation System

2024-02-06 · Youngsuk Kim, Junghwan Lim, Hyuk-Jae Lee, Chae Eun Rhee

The model size growth of personalized recommendation systems poses new challenges for inference. Weight-sharing algorithms have been proposed for size reduction, but they increase memory access. Recent advancements in pr…

CPURecommendation Systems

NestPipe: Large-Scale Recommendation Training on 1,500+ Accelerators via Nested Pipelining

2026-04-08 · Zhida Jiang, Zhaolong Xing, Huichao Chai, Tianxing Sun 외 arxiv

Modern recommendation models have increased to trillions of parameters. As cluster scales expand to O(1k), distributed training bottlenecks shift from computation and memory to data movement, especially lookup and commun…

FeLoG: Scalable and Efficient Distributed Graph Embedding with Feedback Loop Mechanism

2026-06-20 · Peng Fang, Arijit Khan, Ziqiang Wu, Zhenli Li 외 arxiv

Graph embedding maps graph nodes into low-dimensional vectors to support applications such as recommendation, fraud detection, and graph-based retrieval-augmented generation (GraphRAG). As graphs scale to billions of edg…

Fraud DetectionGraph Embedding

Communication optimization strategies for distributed deep neural network training: A survey

2020-03-06 · Shuo Ouyang, Dezun Dong, Yemao Xu, Liquan Xiao

Recent trends in high-performance computing and deep learning have led to the proliferation of studies on large-scale deep neural network training. However, the frequent communication requirements among computation nodes…

Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping

2025-01-11 · Muru Zhang, Mayank Mishra, Zhongzhu Zhou, William Brandon 외

Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are used in multi-gpu training and inference t…

GPULarge Language Model