Automatic Generation of Efficient Sparse Tensor Format Conversion Routines

@article{Chou2020AutomaticGO,
  title={Automatic Generation of Efficient Sparse Tensor Format Conversion Routines},
  author={Stephen Chou and Fredrik Kjolstad and Saman P. Amarasinghe},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.02609}
}
  • Stephen Chou, Fredrik Kjolstad, Saman P. Amarasinghe
  • Published in ArXiv 2020
  • Computer Science
  • This paper shows how to generate code that efficiently converts sparse tensors between disparate storage formats (data layouts) like CSR, DIA, ELL, and many others. We decompose sparse tensor conversion into three logical phases: coordinate remapping, analysis, and assembly. We then develop a language that precisely describes how different formats group together and order a tensor's nonzeros in memory. This enables a compiler to emit code that performs complex reorderings (remappings) of… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 59 REFERENCES

    Vectorized Sparse Matrix Multiply for Compressed Row Storage Format

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Tensor Algebra Compilation with Workspaces

    VIEW 7 EXCERPTS

    Sparse Tensor Algebra Compilation

    • Fredrik Kjolstad
    • Ph.D. Dissertation. Massachusetts Institute of Technology
    • 2020
    VIEW 1 EXCERPT

    HiCOO: Hierarchical Storage of Sparse Tensors

    VIEW 3 EXCERPTS

    A Unified Optimization Approach for Sparse Tensor Operations on GPUs

    VIEW 1 EXCERPT