Roofline-aware DVFS for GPUs

@inproceedings{Nugteren2014RooflineawareDF,
  title={Roofline-aware DVFS for GPUs},
  author={C. Nugteren and Gert-Jan van den Braak and H. Corporaal},
  booktitle={ADAPT '14},
  year={2014}
}
  • C. Nugteren, Gert-Jan van den Braak, H. Corporaal
  • Published in ADAPT '14 2014
  • Computer Science
  • Graphics processing units (GPUs) are becoming increasingly popular for compute workloads, mainly because of their large number of processing elements and high-bandwidth to off-chip memory. The roofline model captures the ratio between the two (the compute-memory ratio), an important architectural parameter. This work proposes to change the compute-memory ratio dynamically, scaling the voltage and frequency (DVFS) of 1) memory for compute-intensive workloads and 2) processing elements for memory… CONTINUE READING
    13 Citations

    Figures and Topics from this paper.

    Exploring GPU performance, power and energy-efficiency bounds with Cache-aware Roofline Modeling
    • 26
    • PDF
    PAQSIM: Fast Performance Model for Graphics Workload on Mobile GPUs
    Power and Energy-efficiency Roofline Model for GPUs
    • 4
    • PDF
    Footprint-Aware Power Capping for Hybrid Memory Based Systems
    Optimizing performance-per-watt on GPUs in high performance computing
    • 12
    • PDF
    Dynamic voltage and frequency scaling for 3D graphics applications on the state-of-the-art mobile GPUs

    References

    Analyzing CUDA workloads using a detailed GPU simulator
    • 1,329
    • Highly Influential
    • PDF