The energy consumption of cloud data centers is a critical concern that
could affect both the environment and the availability of energy
resources. For this, the global community and industries are taking
measures to address this issue that is caused by the high electricity
consumption of servers, Heating, Ventilation, and Air Conditioning
(HVAC) in the data centers. With this context, this paper presents a
novel approach for scheduling energy-efficient workflows (EEWS) in cloud
computing using the MaxUtil model. The proposed approach incorporates
the flower pollination algorithm (FPA), a popular meta-heuristic
algorithm inspired by nature. The primary objectives of the proposed
scheduling scheme are to minimize energy consumption and workflow
processing time (makespan). The proposed algorithm involves two key
phases: (i) assigning tasks to available virtual machines (VMs) and (ii)
scheduling the tasks based on optimal criteria. As per our knowledge,
this is the first study that focuses on optimizing energy consumption
and makespan in cloud computing workflow scheduling using FPA. The
proposed approch employs an effective representation of pollen and
dynamic fitness function with multi-objective. The advantage of FPA lies
in its speed of convergence and providing feasible solutions. Extensive
studies have been conducted across five different scientific workflows
from various fields. The proposed algorithm outperforms traditional
workflow scheduling algorithms based on particle swarm optimization
(PSO), gravitational search algorithms (GSA) and genetic algorithm (GA).
The proposed algorithm outperforms GA, PSO, and GSA in the majority of
cases, according to simulation findings. In addition, a well-known
statistical test known as variance analysis (ANOVA) is used to validate
the experimental results of the suggested algorithm. Based on the
result’s of ANOVA test, the article claims that the suggested algorithm
is superior to existing methods.