RankMap: Priority-Aware Multi-DNN Manager for Heterogeneous Embedded Devices
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916497376411648 |
|---|---|
| author | Karatzas, Andreas Stamoulis, Dimitrios Anagnostopoulos, Iraklis |
| author_facet | Karatzas, Andreas Stamoulis, Dimitrios Anagnostopoulos, Iraklis |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_17867 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RankMap: Priority-Aware Multi-DNN Manager for Heterogeneous Embedded Devices Karatzas, Andreas Stamoulis, Dimitrios Anagnostopoulos, Iraklis Machine Learning Distributed, Parallel, and Cluster Computing Emerging Technologies 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. |
| title | RankMap: Priority-Aware Multi-DNN Manager for Heterogeneous Embedded Devices |
| topic | Machine Learning Distributed, Parallel, and Cluster Computing Emerging Technologies |
| url | https://arxiv.org/abs/2411.17867 |