mlsanityData to 7 Oct 2026

FusedMM: A Unified SDDMM-SpMM Kernel for Graph Embedding and Graph Neural Networks

arXiv 2011.06391 ↗First posted 7 Nov 2020cs.LG cs.DC cs.SIcited by 26

Not on the map. Why?

Abstract (arXiv)

We develop a fused matrix multiplication kernel that unifies sampled dense-dense matrix multiplication and sparse-dense matrix multiplication under a single operation called FusedMM. By using user-defined functions, FusedMM can capture almost all computational patterns needed by popular graph embedding and GNN approaches. FusedMM is an order of magnitude faster than its equivalent kernels in Deep Graph Library. The superior performance of FusedMM comes from the low-level vectorized kernels, a suitable load balancing scheme and an efficient utilization of the memory bandwidth. FusedMM can tune its performance using a code generator and perform equally well on Intel, AMD and ARM processors. FusedMM speeds up an end-to-end graph embedding algorithm by up to 28x on different processors.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). First posted on arXiv 7 Nov 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 7; 2023: 5; 2024: 4; 2025: 7; 2026: 2.

26 papers.

Sort:

By area:

  1. 2025FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Servingcited by 248
  2. 2022SparseTIR: Composable Abstractions for Sparse Compilation in Deep Learningcited by 47
  3. 2022Distributed Graph Neural Network Training: A Surveycited by 32
  4. 2022Parallel and Distributed Graph Neural Networks: An In-Depth Concurrency Analysiscited by 29
  5. 2023A Survey on Graph Neural Network Acceleration: Algorithms, Systems, and Customized Hardwarecited by 18
  6. 2022RSC: Accelerating Graph Neural Networks Training via Randomized Sparse Computationscited by 12
  7. 2023PhysGraph: Physics-Based Integration Using Graph Neural Networkscited by 5
  8. 2023BitGNN: Unleashing the Performance Potential of Binary Graph Neural Networks on GPUscited by 4
  9. 2022Distributed-Memory Sparse Kernels for Machine Learningcited by 3
  10. 2021A Comprehensive Analytical Survey on Unsupervised and Semi-Supervised Graph Representation Learning Methodscited by 2
  11. 2022Scalable Graph Convolutional Network Training on Distributed-Memory Systemscited by 2
  12. 2024Edge Graph Intelligence: Reciprocally Empowering Edge Networks with Graph Intelligencecited by 2
14 more
  1. 2024DF-GNN: Dynamic Fusion Framework for Attention Graph Neural Networks on GPUscited by 2
  2. 2025Fused3S: Fast Sparse Attention on Tensor Corescited by 2
  3. 2023SENSEi: Input-Sensitive Compilation for Accelerating GNNscited by 1
  4. 2023Performance Optimization of Deep Learning Sparse Matrix Kernels on Intel Max Series GPUcited by 1
  5. 2024iSpLib: A Library for Accelerating Graph Neural Networks using Auto-tuned Sparse Operationscited by 1
  6. 2022Architectural Implications of Embedding Dimension during GCN on CPU and GPUcited by 0
  7. 2024GeoT: Tensor Centric Library for Graph Neural Network via Efficient Segment Reduction on GPUcited by 0
  8. 2025MixGCN: Scalable GCN Training by Mixture of Parallelism and Mixture of Acceleratorscited by 0
  9. 2025SparseTransX: Efficient Training of Translation-Based Knowledge Graph Embeddings Using Sparse Matrix Operationscited by 0
  10. 2025Ember: A Compiler for Efficient Embedding Operations on Decoupled Access-Execute Architecturescited by 0
  11. 2025FuseFlow: A Fusion-Centric Compilation Framework for Sparse Deep Learning on Streaming Dataflowcited by 0
  12. 2025AutoSAGE: Input-Aware CUDA Scheduling for Sparse GNN Aggregation (SpMM/SDDMM) and CSR Attentioncited by 0
  13. 2026FlashSchNet: Fast and Accurate Coarse-Grained Neural Network Molecular Dynamicscited by 0
  14. 2026Scalable and Adaptive Parallel Training of Graph Transformer on Large Graphscited by 0