Guyue Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Guyue Huang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Yufei Ding
University of California San Diego
4 shared publications - Yuan Xie
The Hong Kong University of Science and Technology, Hong Kong, Hong Kong
3 shared publications - Guohao Dai
Shanghai Jiao Tong University, Shanghai, China
2 shared publications - Boyuan Feng
UC Santa Barbara, Santa Barbara, United States of America
1 shared publication - Chen Zhang
Shanghai Jiao Tong University, Shanghai, Shanghai, China
1 shared publication - Enshu Liu
Tsinghua University
1 shared publication
Latest Publications
- Open Access
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1880-1893
TRACI: Network Acceleration of Input-Dynamic Communication for Large-Scale Deep Learning Recommendation Model
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1880-1893
Open AccessTRACI: Network Acceleration of Input-Dynamic Communication for Large-Scale Deep Learning Recommendation Model
- 5 December 2024
- Published by Springer Nature
- p. 285-302
MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
- 5 December 2024
- Published by Springer Nature
- p. 285-302
MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
- Open Access
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 338-352
RM-STC: Row-Merge Dataflow Inspired GPU Sparse Tensor Core for Energy-Efficient Sparse Acceleration
This paper proposes RM-STC, a novel GPU tensor core architecture designed for sparse Deep Neural Networks (DNNs) with two key innovations: (1) native support for both training and inference and (2) high efficiency for all sparsity degrees. To achieve the first goal, RM-STC employs a uniform sparse encoding scheme that natively supports all operations holistically in forward and backward passes, thereby eliminating the need for costly sparse encoding transformation in between. For the second goal, RM-STC takes inspiration from the row-merge dataflow and combines the input-gathering and output-scattering hardware features to minimize the energy overhead. Experiments show that RM-STC achieves significant speedups and energy efficiency improvements over dense tensor cores and previous sparse tensor cores.
- 28 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 338-352
Open AccessRM-STC: Row-Merge Dataflow Inspired GPU Sparse Tensor Core for Energy-Efficient Sparse Acceleration
This paper proposes RM-STC, a novel GPU tensor core architecture designed for sparse Deep Neural Networks (DNNs) with two key innovations: (1) native support for both training and inference and (2) high efficiency for all sparsity degrees. To achieve the first goal, RM-STC employs a uniform sparse encoding scheme that natively supports all operations holistically in forward and backward passes, thereby eliminating the need for costly sparse encoding transformation in between. For the second goal, RM-STC takes inspiration from the row-merge dataflow and combines the input-gathering and output-scattering hardware features to minimize the energy overhead. Experiments show that RM-STC achieves significant speedups and energy efficiency improvements over dense tensor cores and previous sparse tensor cores.
- Open Access
TC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs
Recently, graph neural networks (GNNs), as the backbone of graph-based machine learning, demonstrate great success in various domains (e.g., e-commerce). However, the performance of GNNs is usually unsatisfactory due to the highly sparse and irregular graph-based operations. To this end, we propose TC-GNN, the first GNN acceleration framework based on GPU Tensor Core Units (TCUs). The core idea is to reconcile the "Sparse" GNN computation with the high-performance "Dense" TCUs. Specifically, we conduct an in-depth analysis of the sparse operations in mainstream GNN computing frameworks. We introduce a novel sparse graph translation technique to facilitate TCU processing of the sparse GNN workload. We implement an effective CUDA core and TCU collaboration design to fully utilize GPU resources. We integrate TC-GNN with the PyTorch framework for high programmability. Rigorous experiments show an average of 1.70X speedup over the state-of-the-art DGL framework across various models and datasets.
Open AccessTC-GNN: Bridging Sparse GNN Computation and Dense Tensor Cores on GPUs
Recently, graph neural networks (GNNs), as the backbone of graph-based machine learning, demonstrate great success in various domains (e.g., e-commerce). However, the performance of GNNs is usually unsatisfactory due to the highly sparse and irregular graph-based operations. To this end, we propose TC-GNN, the first GNN acceleration framework based on GPU Tensor Core Units (TCUs). The core idea is to reconcile the "Sparse" GNN computation with the high-performance "Dense" TCUs. Specifically, we conduct an in-depth analysis of the sparse operations in mainstream GNN computing frameworks. We introduce a novel sparse graph translation technique to facilitate TCU processing of the sparse GNN workload. We implement an effective CUDA core and TCU collaboration design to fully utilize GPU resources. We integrate TC-GNN with the PyTorch framework for high programmability. Rigorous experiments show an average of 1.70X speedup over the state-of-the-art DGL framework across various models and datasets.
- Open Access
Efficient Sparse Matrix Kernels based on Adaptive Workload-Balancing and Parallel-Reduction
Sparse matrix-vector and matrix-matrix multiplication (SpMV and SpMM) are fundamental in both conventional (graph analytics, scientific computing) and emerging (sparse DNN, GNN) domains. Workload-balancing and parallel-reduction are widely-used design principles for efficient SpMV. However, prior work fails to resolve how to implement and adaptively use the two principles for SpMV/MM. To overcome this obstacle, we first complete the implementation space with optimizations by filling three missing pieces in prior work, including: (1) We show that workload-balancing and parallel-reduction can be combined through a segment-reduction algorithm implemented with SIMD-shuffle primitives. (2) We show that parallel-reduction can be implemented in SpMM through loading the dense-matrix rows with vector memory operations. (3) We show that vectorized loading of sparse rows, being a part of the benefit of parallel-reduction, can co-exist with sequential-reduction in SpMM through temporally caching sparse-matrix elements in the shared memory. In terms of adaptive use, we analyze how the benefit of two principles change with two characteristics from the input data space: the diverse sparsity pattern and dense-matrix width. We find the benefit of the two principles fades along with the increased total workload, i.e. the increased dense-matrix width. We also identify, for SpMV and SpMM, different sparse-matrix features that impact workload-balancing effectiveness. Our design consistently exceeds cuSPARSE by 1.07-1.57x on different GPUs and dense matrix width, and the kernel selection rules involve 5-12% performance loss compared with optimal choices. Our kernel is being integrated into popular graph learning frameworks to accelerate GNN training.
Open AccessEfficient Sparse Matrix Kernels based on Adaptive Workload-Balancing and Parallel-Reduction
Sparse matrix-vector and matrix-matrix multiplication (SpMV and SpMM) are fundamental in both conventional (graph analytics, scientific computing) and emerging (sparse DNN, GNN) domains. Workload-balancing and parallel-reduction are widely-used design principles for efficient SpMV. However, prior work fails to resolve how to implement and adaptively use the two principles for SpMV/MM. To overcome this obstacle, we first complete the implementation space with optimizations by filling three missing pieces in prior work, including: (1) We show that workload-balancing and parallel-reduction can be combined through a segment-reduction algorithm implemented with SIMD-shuffle primitives. (2) We show that parallel-reduction can be implemented in SpMM through loading the dense-matrix rows with vector memory operations. (3) We show that vectorized loading of sparse rows, being a part of the benefit of parallel-reduction, can co-exist with sequential-reduction in SpMM through temporally caching sparse-matrix elements in the shared memory. In terms of adaptive use, we analyze how the benefit of two principles change with two characteristics from the input data space: the diverse sparsity pattern and dense-matrix width. We find the benefit of the two principles fades along with the increased total workload, i.e. the increased dense-matrix width. We also identify, for SpMV and SpMM, different sparse-matrix features that impact workload-balancing effectiveness. Our design consistently exceeds cuSPARSE by 1.07-1.57x on different GPUs and dense matrix width, and the kernel selection rules involve 5-12% performance loss compared with optimal choices. Our kernel is being integrated into popular graph learning frameworks to accelerate GNN training.