International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-9 Issue-6, April 2020 667 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: E3029039520/2020©BEIESP DOI: 10.35940/ijitee.E3029.049620 High Performance Energy-Aware Cloud Computing: A Scope of Future Computing Shailesh Saxena, Mohammad Zubair Khan, Ravendra Singh Abstract- This publication discusses high-performance energy- aware cloud (HPEAC) computing state-of-the-art strategies to acknowledgement and categorization of systems and devices, optimization methodologies, and energy / power control techniques in particular. System types involve single machines, clusters, networks, and clouds, while CPUs, GPUs, multiprocessors, and hybrid systems are known to be device types. Objective of Optimization incorporates multiple calculation blends, such as “execution time”, “consumption of energy”& “temperature” with the consideration of limiting power/energy consumption. Control measures usually involve scheduling policies, frequency based policies (DVFS, DFS, DCT), programmatic API’s for limiting the power consumptions (such as” Intel- RAPL”,” NVIDIA- NVML”), standardization of applications, and hybrid techniques. We address energy / power management software and APIs as well as methods and conditions in modern HPEACC systems for forecasting and/or simulating power/energy consumption. Eventually, programming examples are discussed, i.e. programs & tests used in specific works. Based on our study, we point out some areas and there significant issues related to tools & technologies, important for handling energy aware computations in HPEAC computing environment. Keywords: Energy, Temperature, Control, Policies, Cloud I. INTRODUCTION Understanding energy and power plays an increasingly important role in modern high-performance cloud computing systems. In near future, such exascale computing systems will develop which consume near about 20 MW of maximum power. On the basis of top performance-oriented ranking (https:/www. top500.org/ lists/top500/), and the ranking of supercomputers per watt efficiency (https:/www.top500.org/green500/), Mass acceptance of GPUs has contributed to expand this balance for programs that can operate on such machines efficiently. Parallelization with programming in these hybrid systems must require achieving high efficiency, but there is an issue while it applies with multiple and many core computing. In place of scheduling policies, scaling policies (like DVFS, DFS & DCT) and power reducing APIs have become acceptable for both-CPU & GPU machines, servers as well as mobile lines as power and energy management approaches. Now we have power capping system like Slrum for job management in cluster, which allows idle nodes to be shut down and rebooted when required for reducing the power consumption through DVFS [29]. Revised Manuscript Received on April 1, 2020. Shailesh Saxena, Research Scholar, MJP Rohilkhand University, Bareilly, India Mohammad Zubair Khan, Department of CS, College of Computer Science and Engg., Taibah University, Medina, KSA Ravendra Singh, Department of CS and IT, MJP Rohilkhand University, Bareilly, India In different applications-total computation time, utilization of resources, energy consumption, and temperature surrounding the machine are define as key parameters including in different contexts and in different patterns. A constant and detailed review of possibilities, processes & techniques are needed to handle the challenges of HPEAC computing for better results, so we review the literature with the same objective and describe our study in further section of the paper. II. EXISTING LITERATURE First of all, in the literature [67], the question of adequate energy and performance metrics was reviewed. Several survey associated with high-performance energy-aware computing are there and based on only the study of related techniques, but the environment and tools are also have the impact on energy aware computation. Which are missing in those surveys? So we also include these missing aspects in our review. In [79] a study of recent data center and cloud were defined which indicate a wide range of energy-aware factors that are associated with computation in cloud environments. The researchers recommended a nomenclature regarding the optimization of power/energy in computing systems including different levels of abstraction and introducing energy-related works including energy model characterizing features of software and hardware both. So we expand this survey with new approaches, tools and represent a more streamlined review of the power/energy issues of today. A segregates energy-aware computational approach for different computing environment (i.e. servers, clusters, data centers, grid and clouds) is address in [80]. But there is a lacks of description regarding the optimization parameters, techniques such as power capping and the analysis of API’s. We therefore include in our classification study of these performance metrics for optimization of techniques regarding the energy-aware regulation and measurements. Work in [68] reviews the strategies based on software & hardware techniques for the energy-aware performance analysis as well as technologies of monitoring the energy of computation in HPEAC systems. However this paper does not review those strategies which evaluate and control the consumption of power/energy. It’s main objective is to accumulate the available monitoring strategies for power/energy. However, the review of existing tools in terms of cost, portability and consumer-friendly criteria is included in the paper. Subsequently, we reviewed the controlling policies of power/energy and include it in our review.