RankMap: Priority-Aware Multi-DNN Manager for Heterogeneous Embedded Devices
This addresses workload management challenges for edge data centers handling multi-DNN tasks, representing a strong domain-specific improvement.
The paper tackles the problem of managing multiple Deep Neural Networks (DNNs) on heterogeneous embedded devices by introducing RankMap, a priority-aware manager that uses stochastic space exploration and a performance estimator. The results show RankMap achieves 3.6x higher average throughput and improves prioritization of specified DNNs by 57.5x compared to existing methods.
Modern edge data centers simultaneously handle multiple Deep Neural Networks (DNNs), leading to significant challenges in workload management. Thus, current management systems must leverage the architectural heterogeneity of new embedded systems to efficiently handle multi-DNN workloads. This paper introduces RankMap, a priority-aware manager specifically designed for multi-DNN tasks on heterogeneous embedded devices. RankMap addresses the extensive solution space of multi-DNN mapping through stochastic space exploration combined with a performance estimator. Experimental results show that RankMap achieves x3.6 higher average throughput compared to existing methods, while preventing DNN starvation under heavy workloads and improving the prioritization of specified DNNs by x57.5.