期刊
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
卷 40, 期 2, 页码 626-640出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSAC.2021.3118403
关键词
Computational modeling; 5G mobile communication; Servers; Load modeling; Data models; Adaptation models; Training; Edge computing; machine learning (ML); neural networks; 5G mobile communication
资金
- Faculty Innovation Award of Sony Research Award Program
This paper introduces HiveMind, a practical multi-split machine learning system designed for 5G cellular networks. HiveMind optimizes the complex split assignment problems and achieves optimal split decisions with low overhead. The system is also adaptable to various ML frameworks and tasks, demonstrating wide applicability.
The increasing processing load of today's mobile machine learning (ML) application challenges the stringent computation budget of mobile user equipment (UE). With the wide deployment of 5G edge-cloud, a new ML offloading scheme called split ML is provisioned to enable computation-intensive mobile ML applications by splitting an ML model across mobile UE, edge, and cloud. However, the complex split assignment problems pose new challenges for split ML system design. In this paper, we introduce HiveMind, the first practical multi-split ML system tailored for 5G cellular networks. HiveMind reformulates the complicated multi-split problem to a min-cost graph search and optimizes the distributed algorithm to drastically reduce the signaling overhead. Benefit from its low overhead property, HiveMind makes the optimal split decision on multiple computing nodes in real-time and adapts the split decisions to the instantaneous network dynamics. HiveMind also incorporates a multi-objective mechanism that accommodates heterogeneous objectives for a single ML task. HiveMind adapts to a wide range of ML frameworks, including non-linear models like Recurrent Neural Network (RNN), Federated Learning (FL), and Multi-agent Reinforcement Learning (MARL). We evaluate HiveMind on 5G MEC network simulators with realistic traffic patterns and real-life MEC computation/communication profiles. Our experiments demonstrate that HiveMind achieves the optimal efficiency comparing to state-of-art split ML designs.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据