21296

PDAWL: Profile-based Iterative Dynamic Adaptive WorkLoad Balance on Heterogeneous Architectures

Tongsheng Geng, Marcos Amaris, Stephane Zuckerman, Alfredo Goldman, Guang R. Gao, Jean-Luc Gaudiot
University of California, Irvine, CA, USA
23rd Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP’20), 2020

@inproceedings{geng2020pdawl,

   title={PDAWL: Profile-based Iterative Dynamic Adaptive WorkLoad Balance on Heterogeneous Architectures},

   author={Geng, Tongsheng and Amaris, Marcos and Zuckerman, St{‘e}phane and Goldman, Alfredo and Gao, G and Gaudiot, Jean-Luc},

   booktitle={23rd Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP 2020)},

   year={2020}

}

Download Download (PDF)   View View   Source Source   

363

views

While High Performance Computing systems are increasingly based on heterogeneous cores, their effectiveness depends on how well the scheduler can allocate workloads onto appropriate computing devices and how communication and computation can be overlapped. With different types of resources integrated into one system, the complexity of the scheduler correspondingly increases. Moreover, for applications with varying problem sizes on different heterogeneous resources, the optimal scheduling approach may vary accordingly. We thus present PDAWL, an event-driven profile-based Iterative Dynamic Adaptive Work-Load balance scheduling approach to dynamically and adaptively adjust workload to efficiently utilize heterogeneous resources. It combines online scheduling (DAWL), which can adaptively adjust workload based on available real time heterogeneous resources, with an offline machine learning (profilebased estimation model) which can build a device-specific communication computation estimation model. Our scheduling approach is tested on control-regular applications, Stencil kernel (based on a Jacobi Algorithm) and Sparse Matrix-Vector Multiplication (SpMV) in an event-driven runtime system. Experimental results show that PDAWL is either on-par or far outperforms whichever yields the best results (CPU or GPU).
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2020 hgpu.org

All rights belong to the respective authors

Contact us: