Durreesamin Journal (ISSN: 2204-9827) July Vol 4 Issue 2, Year 2018 1 TOWARDS A MACHINE LEARNING BASED ARTIFICIALLY INTELLIGENT SYSTEM FOR ENERGY EFFICIENCY IN A CLOUD COMPUTING ENVIRONMENT Yao Francois Michael Kra, 1840192845@qq.com, School of Computer Science and Engineering, Southeast University, Nanjing, China Noah Kwaku Baah, baah.noah@gmail.com, School of Computer Science and Engineering, Southeast University, Nanjing, China. Imran Memon, imranmemon52@zju.edu.cn, College of Computer Science, Zhejiang University, Hangzhou 310027, China William Gyasi-Mensah, kyfm349349@gmail.com, School of Finance and Economics Jiangsu University ABSTRACT Cloud computing has become the mainstream of the emerging technologies for information interchange and accessibility. With such systems, the information accessed from any geographic location on this planet with some decent kind of internet connection. Applying machine learning together with artificial intelligence in dealing with the problem of energy reduction in cloud data center is an innovative idea. A large combination of Artificial intelligence is playing a significant role in cloud environment. For that matter, the Big organization providers like Amazon have taken steps to ensure that they can continue to expand their fast-growing cloud services to commensurate with the fast growth of population. These companies have built large data centers in remote parts of the world to overcome a shortage of information. These centers consume significant amounts of electrical energy. There is often a lot of energy wastage. According to IDC white paper, data centers have tremendously wasted billions of energy regarding billing and cash. Additionally, researchers have argued that by the year 2020 the energy consumption rate would have doubled. Research in this area is still a hot topic. This paper seeks to address the energy efficiency issue at a Cloud Data Center using machine learning methodologies, principles, and practices. This article also aims to bring out possible future implementation methods for artificially intelligent agents that would help reduce energy wastage at a Cloud data center and thus help ameliorate the great big energy problem at hand. Keywords: Cloud Computing; PUE; Energy Efficiency, Machine Learning, Artificial Intelligence, Cloud Service Provider (CSP) Virtualization I. INTRODUCTION Recent years, cloud computing has demonstrated, established and founded itself as one of the brains and drivers in modern technology. As a process paradigm faculty economy of scale, when organized and used effectively, the cloud computing presents significant advantages relating to computation power whereas reducing expenditures and saving energy. Massive data centers are in places wherever the concept of cloud computing involves life. Through virtualization technology, data center resources and services became substantially potential for several users to share, and to avoid having to line up their infrastructure to try and do things that have been completed within the cloud. Efficient use of energy in cloud computing has been receiving attention by researchers over the past decade. Some studies have suggested various optimization approaches to the challenge of minimizing the expenditure of energy within cloud computing setting [36],[37],[25],[20],[22]. Several scenarios also exist for using machine instruction strategies to material supplies and management within the cloud, with several goals. (The study will provide a survey towards a machine learning based artificially intelligent system for the efficient use of energy in a cloud computing setting). (The aim of this study area is to analyze and delve into energy efficiency, and carry up to the machine learning research, as well as support their invention in innovative ways capable of producing preferred outcomes. As computing has become very vast and sophisticated engine worldwide, cloud computing as a traditional model delivers, computing resources on cloud computing uses pay as you use method. The public IT corporations like Microsoft, Google, Amazon, and IBM have a unit of measurement running expansive data knowledge Centres worldwide to handle their always-rising requests. Notably, the rising demands for cloud computing facilities have considerably multiplied the power usage of knowledge centers, thereby making it an important issue). The third drop in energy charge for an outsized associate company like Google will reach over 1,000,000 dollars in value savings [35]. High power consumption does not only interpret to the great value but to boot leads to high carbon emissions that do not appear to