Service composition considering energy consumption of users and transferring files in a multicloud environment

In the internet and cloud environment, service composition is always used to enhance the function and processing ability of clouds. Those clouds work together for a user and provide different functions. A service request may involve multiple clouds. The past work focuses on the method of service composition and ignores the energy composition when files are transferred between clouds, including the energy consumption for transferring files (sending files from the user to the cloud and receiving files from the cloud to the user) of the user. The paper models the service composition in a multicloud environment. Based on those models, we use the GA (genetic algorithm) algorithm (GA-C) to solve the service composition problem with multiple targets in a multicloud environment. Simulation results show that the GA-C can: (1) reduce the average number of involved clouds and the energy consumption between clouds, and (2) reduce the energy consumption of the user and the failure rate of service composition.

Automated summary

This paper investigates the service composition problem in a multicloud environment and finds that previous work has overlooked the energy consumption during file transfer. The author proposes the use of a genetic algorithm to solve this issue and simulation results confirm that it can reduce the energy consumption between clouds and for the user, as well as decrease the failure rate of service composition.