.png){kind=icon}
.png){kind=icon}
Return
Enhanced clustering and filtering algorithm for cloudlet movement and placement in dynamic mobile cloud computing environments
.png){kind=icon}
PRE
AI
DOI：10.1007/s10586-025-05815-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Cloudlet placement in dynamic environments poses significant challenges due to the mobility of users. Effective placement strategies are crucial for enhancing the performance of mobile applications. Recent research has applied clustering algorithms to tackle these challenges, including k-means, mini-batch k-means, and k-medoids. This paper introduces an enhanced clustering and filtering cloudlet movement and placement (e-CFCMP) algorithm designed to identify optimal positions for movable cloudlets in dynamic environments, thereby supporting the efficient operation of mobile applications. The proposed algorithm includes three variants: e-CFkm, e-CFmb, and e-CFkd. Among these, the first variant achieves the optimal position by maximizing the number of covered devices while minimizing assignment costs. It outperforms the other two variants, covering 0.88% and 1.51% more devices, respectively. Additionally, it reduces intra-cluster assignment costs by 1.58% and nearly 99% compared to contemporary algorithms. Furthermore, the algorithm ensures 100% placement of cloudlets with uniformly distributed devices, enhancing resource utilization. Therefore, the e-CFkm variant performs better than its counterparts.
Keywords:
Cloudlet placement
Clustering
k -means
Mini-batch k -means
k -medoids
Mobile cloud computing
Multi-access edge computing
Journal
C
IF:
0
Papers:
691
Citations:
1