.png){kind=icon}
.png){kind=icon}
Return
Simulating Load Sharing for Resource Constrained Devices
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3475996.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recently, we have witnessed unprecedented communication across heterogeneous devices. These devices exploit their shared environment to harvest, process, and utilize data gathered to improve the consumer's experience and overall quality of life. However, most of these multi-purpose devices have limited computing resources, such as storage, battery, and processing capabilities. Typically, requests for tasks requiring complex computation are routed across the Internet, leading to delays, congestion, and battery drain. This paper presents the why and how to move these computations closer to the edge devices minimizing network latency and improving the overall network quality. The division of computational tasks across multiple devices, known as load sharing, is not new. However, there is a disconnect regarding the choice and implementation of simulators for resource-constrained devices. Thus, we propose a standard resource-constrained device-focused load-sharing simulation framework that is modular, extendable and integrates with existing general-purpose simulation tools. To demonstrate the effectiveness of our method, we implement the framework in OMNeT++ and investigate the performance of several traditional load-sharing algorithms. We also propose a novel heuristics-based space-efficient load-sharing algorithm and find that this approach improves the fairness in task distribution across the network.
Keywords:
Heuristic algorithms
Performance evaluation
Internet of Things
Cloud computing
Biological system modeling
Ecosystems
Computational modeling
Real-time systems
Data processing
Load management
Edge computing
Load sharing
edge computing
resource constrained devices
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W
View more organizations