Performance Analysis of Greedy and Auction-Based Resource Allocation Algorithms in Ubiquitous Computing Environments

In the era of ubiquitous computing, efficient resource allocation is critical to managing the diverse and dynamic environments created by interconnected devices. This paper presents a comprehensive comparative analysis of greedy and auction-based resource allocation algorithms within ubiquitous computing systems. The study examines the performance of these algorithms under three implementation approaches: centralized, decentralized, and hierarchical. We use a process-based discrete-event simulation model to evaluate the algorithms based on key performance metrics, including throughput, average response time, and energy utilization. Our results demonstrate that auction-based algorithms, particularly in a hierarchical implementation, consistently outperform greedy algorithms in both throughput and energy efficiency while maintaining competitive response times. The findings highlight the advantages of auction-based strategies in dynamically adapting to changing conditions and optimizing resource use in complex, decentralized environments. This research contributes valuable insights into optimal resource management strategies for ubiquitous computing, guiding system designers and researchers toward more efficient and scalable solutions. Future work should address the limitations of simplified network and energy models to enhance the applicability of these findings in real-world scenarios.