Modeling and Performance Analysis of Multi-Server Cloud Database Over Quasi-static Rayleigh Fading Channel

IEEE Internet of Things Journal(2024)

引用 0|浏览4
暂无评分
摘要
With the development of communication in the post-5G era, the combination of communication and cloud computing becomes closer. In order to promote the further development of cloud-network, this paper will study the performance of Multi-server cloud Database under the Communication quality of quasi-static Rayleigh fading channel with Multiple antennas(MC-MD). The CLIENTS with unlimited customers, a COMMUNICATION SYSTEM subject to quasi-static Rayleigh fading, and CLOUD DATABASE with two-phase locking protocol are the three components of the MC-MD model. Transactions are 1)initiated by the CLIENTS, 2)transmitted to the CLOUD DATABASE through the COMMUNICATION SYSTEM for processing, 3)then returned to the CLIENTS. The indicators of the model is mathematically derived by using queuing theory. These include client’s indicators(average concurrent quantity of the system in steady state(CQ), average transactions stay time of the system in steady state(ST), average queue length of the Waiting Area in steady state(QL), and average transactions wait time of the Waiting Area in steady state(WT)) and server’s indicator(average number of service desks in the busy period at steady state(DN)). Under the appropriate conditions, the results indicate that the theoretical value of service performance is basically consistent with the simulation value. Clearly, the high speed improves the service performance of the system and decreases the service pressure. On the basis of this, the optimization strategy is proposed and the simulation indicators Jitter of transactions sojourn time of the system in steady state(STJ) is added. The results show that the transaction scheduling optimization strategy effectively reduce the delay and its jitter.
更多
查看译文
关键词
multi-server cloud database,quasi-static Rayleigh fading channel with multiple antennas,shared lock,exclusive lock,queue system
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要