Corpus ID: 168169756

Predict-and-recompute conjugate gradient variants

@article{Chen2019PredictandrecomputeCG,
  title={Predict-and-recompute conjugate gradient variants},
  author={Tyler Chen},
  journal={ArXiv},
  year={2019},
  volume={abs/1905.01549}
}
  • Tyler Chen
  • Published 2019
  • Mathematics, Computer Science
  • ArXiv
  • The standard implementation of the conjugate gradient algorithm suffers from communication bottlenecks on parallel architectures, due primarily to the two global reductions required every iteration. In this paper, we study conjugate gradient variants which decrease the runtime per iteration by overlapping global synchronizations, and in the case of pipelined variants, matrix-vector products. Through the use of a predict-and-recompute scheme, whereby recursively-updated quantities are first used… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.

    Acceleration of Structural Analysis Simulations using CNN-based Auto-Tuning of Solver Tolerance

    VIEW 1 EXCERPT
    CITES METHODS

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 55 REFERENCES

    Improving Performance of GMRES by Reducing Communication and Pipelining Global Collectives

    VIEW 1 EXCERPT

    Scalable Non-blocking Preconditioned Conjugate Gradient Methods

    • Paul R. Eller, William Gropp
    • Computer Science
    • SC16: International Conference for High Performance Computing, Networking, Storage and Analysis
    • 2016