MAGMA: An Optimization Framework for Mapping Multiple DNNs on Multiple Accelerator Cores
Authors:
- 1 April 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 814-830
Abstract
As Deep Learning continues to drive a variety of applications in edge and cloud data centers, there is a growing trend towards building large accelerators with several sub-accelerator cores/chiplets. This work looks at the problem of supporting multi-tenancy on such accelerators. In particular, we focus on the problem of mapping jobs from several DNNs simultaneously on an accelerator. Given the extremely large search space, we formulate the search as an optimization problem and develop an optimization framework called M3E. In addition, we develop a specialized optimization algorithm called MAGMA with custom operators to enable structured sampleefficient exploration. We quantitatively compare MAGMA with several state-of-the-art methods, black-box optimization, and reinforcement learning methods across different accelerator settings (large/small accelerators) and different sub-accelerator configurations (homogeneous/heterogeneous), and observe MAGMA can consistently find better mappings.Keywords
All Related Versions
This publication has 58 references indexed in Scilit:
- Optimizing FPGA-based Accelerator Design for Deep Convolutional Neural Networks2015
- Particle swarm optimization2014
- Resource-constrained spatial multi-tasking for embedded GPU2014
- Differential Evolution2013
- The CMA Evolution Strategy: A Comparing Review2007
- Libra: a computational economy‐based job scheduling system for clustersSoftware: Practice and Experience, 2004
- Scheduling multiprocessor tasks with genetic algorithmsIEEE Transactions on Parallel and Distributed Systems, 1999
- A genetic algorithm for multiprocessor schedulingIEEE Transactions on Parallel and Distributed Systems, 1994
- Genetic AlgorithmsScientific American, 1992
- LIII. On lines and planes of closest fit to systems of points in spaceJournal of Computers in Education, 1901