Edge computing has emerged as a popular paradigm for supporting mobile and IoT applications with low latency or high bandwidth needs. The attractiveness of edge computing has been further enhanced due to the recent availability of special-purpose hardware to accelerate specific compute tasks, such as deep learning inference, on edge nodes. In this paper, we experimentally compare the benefits and limitations of using specialized edge systems, built using edge accelerators, to more traditional forms of edge and cloud computing. Our experimental study using edge-based AI workloads shows that today's edge accelerators can provide comparable, and in many cases better, performance, when normalized for power or cost, than traditional edge and cloud servers. They also provide latency and bandwidth benefits for split processing, across and within tiers, when using model compression or model splitting, but require dynamic methods to determine the optimal split across tiers. We find that edge accelerators can support varying degrees of concurrency for multi-tenant inference applications, but lack isolation mechanisms necessary for edge cloud multi-tenant hosting.
This content will become publicly available on October 1, 2023
Optimal Task Allocation and Coding Design for Secure Edge Computing With Heterogeneous Edge Devices
- Award ID(s):
- 1730325
- Publication Date:
- NSF-PAR ID:
- 10387993
- Journal Name:
- IEEE Transactions on Cloud Computing
- Volume:
- 10
- Issue:
- 4
- Page Range or eLocation-ID:
- 2817 to 2833
- ISSN:
- 2372-0018
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The envisioned capabilities of mobile edge computing are predicated on a delivery infrastructure with capacity, ubiquity, robustness, and capabilities to serve a country-wide user base. In this paper, we present an empirical study of key aspects of mobile edge infrastructure toward the goal of understanding their current characteristics and identifying future deployments. We start by analyzing a dataset of over 4M cell tower locations in the US. We evaluate the geographic characteristics of deployments and highlight how locations correspond to population density in major metropolitan areas and in rural areas. We also show how deployments have been arranged along highways throughout the US. Our analysis highlight areas where new deployments would be warranted. Finally, we analyze how cell tower deployments correspond to current major data center locations and assess how micro servers might be deployed to improve response times and to better serve customers.
-
Localization in urban environments is becoming increasingly important and used in tools such as ARCore [11], ARKit [27] and others. One popular mechanism to achieve accurate indoor localization as well as a map of the space is using Visual Simultaneous Localization and Mapping (Visual-SLAM). However, Visual-SLAM is known to be resource-intensive in memory and processing time. Further, some of the operations grow in complexity over time, making it challenging to run on mobile devices continuously. Edge computing provides additional compute and memory resources to mobile devices to allow offloading of some tasks without the large latencies seen when offloading to the cloud. In this paper, we present Edge-SLAM, a system that uses edge computing resources to offload parts of Visual-SLAM. We use ORB-SLAM2 as a prototypical Visual-SLAM system and modify it to a split architecture between the edge and the mobile device. We keep the tracking computation on the mobile device and move the rest of the computation, i.e., local mapping and loop closure, to the edge. We describe the design choices in this effort and implement them in our prototype. Our results show that our split architecture can allow the functioning of the Visual-SLAM system long-term with limited resourcesmore »
