Elham Khodabakhsh - Student of Department of Computer Engineering, Islamic Azad University, UAE Branch, Dubai, UAE,
Hamid Reza Naji - Associate Professor of Department of Computer Engineering, Islamic Azad University, UAE Branch, Dubai, UAE
Mohammad Malakoti - Associate Professor of Department of Computer Engineering, Islamic Azad University, UAE Branch,Dubai, UAE,

Cloud Computing is an emerging concept combining many fields of computing. It provides services,software and processing capacity over the internet. Cloud computing is abundantly accepted by the business markets and organizations. Data centers host hundreds or thousands of servers comprised of software and hardware to respond to the clients' requests. A large amount of energy is consumed by data centers to provide promised services. The power consumption in data centers raises many serious issues including excessive carbon dioxide emission; therefore, many solutions have been offered and much research has been done to reduce energy consumption in data centers. Task scheduling has effect on energy consumed by datacenters. One of the important factors for task scheduling is how the tasks can access their required data. Certainly, delay in accessing data by tasks can impress on the total completiontime. This paper presents a multi- agent system which is used for task scheduling in cloud to reduce the energy consumption of data centers.

cloud computing, task scheduling, energy consumption, multi-agent