Task processing optimization using cuckoo particle swarm (CPS) algorithm in cloud computing infrastructure

• Published in: Cluster computing Vol. 26; no. 1; pp. 745 - 769
• Main Authors: Zavieh, Hadi, Javadpour, Amir, Li, Yuan, Ja’fari, Forough, Nasseri, Seyed Hadi, Rostami, Ali Shokouhi
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.02.2023; Springer Nature B.V
• Subjects: Algorithms; Cloud computing; Computer centers; Computer Communication Networks; Computer Science; Energy consumption; Fault tolerance; Infrastructure; Operating costs; Operating Systems; Optimization; Particle swarm optimization; Processor Architectures; Quality of service; Virtual environments; Cloud computing; Fault tolerance; Virtual machine migration; Task processing; Cuckoo algorithm; PSO
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-022-03796-9

Recently, cloud computing infrastructure (CCI) models have received much attention for their exceptional scalability, dependability, Data Information Sharing (DIS), and low cost rate. There are many hardware and software elements that are accessed over the internet by cloud data centers. Modern data centers utilize Virtualization Techniques (VT) to offer a dispersed CI that employs Virtual Machines (VM) based on Physical Hosts (PH). With the increasing number of centers, optimizing energy consumption has become vital to saving costs due to DCC's high energy consumption. In our CPS algorithm, we combine the Cuckoo algorithm and the particle swarm optimization (PSO). It is determined which virtual machine can be assigned to each host, thus choosing the best virtual machine. As a result, if the selected host is overloaded, it is determined which virtual machines are generating high loads and migrated to another host, which is determined based on the cuckoo algorithm and PSO. In testing each algorithm separately, the combination method proved to consume less energy and execute faster than the other methods in the CloudSim simulation environment. Fault tolerance for our network and evaluation of VMs have also been emphasized in vSphereTM.