On Tuesday, 28th of April 2020, 09:00 AM, M.Sc. student “Noora Nahidh Taher” defended her thesis entitled:

“Energy-Efficient CPU Management for Cloud Data Center”

The discussion committee included:
Prof.Dr. Ali Sadiq Abdul Hadi / College of Information Engineering / Chairman.
Assistant Prof.Dr. Ammar Dawood Jasim / College of Information Engineering / Member.
Assistant Prof.Dr.Bara Munqith Al-Bakr / Al-Iraqia University / College of Engineering. / Member.
Dr. Rabah Aboud Ahmed / College of Information Engineering/member and supervisor.

Abstract :

Cloud computing is a technology used to create a revolution for the next generations within the computing scope. There was an interest in the cloud computing field, This interest raised from the fact that the cloud provides a flexible environment which consequently leads to raising the ability to handle an extended work volume. The main problem that affects this revolution is attributed to the huge amount of energy consumed by the cloud datacenters. In 2020 energy consumption of data centers may reach 112 billion kWh. This issue can be resolved by either increasing the efforts to produce more energy to cover the full need of these datacenters or by finding new techniques to reduce energy consumption and preform new schemes to develop these techniques. For all the reasons mentioned, this work presented a method to manage the virtual central process unit (vCPU) of the cloud data centers to reduce energy consumption. The study comprised of two parts for the experimental work (i) real system hardware work (ii) simulation work.

The hardware work presented a real system and used three different types of workloads (video streaming, file compression process, and video games) to study the influence of vCPU usage percentage on the energy consumed. In The simulation work, the Cloudsim simulator was used to show the influence of five different estimated scenarios on the energy consumption by using the DVFS technique with the proposed approach. The results obtained from the real system hardware work showed an increase in the energy-saving percentage. Also, the result of the Cloudsim simulator shows that about 13% of total energy is saved in the data. This saving percentage result was a better percentage comparing with other results obtained from previous work.

At the end of the discussion, the thesis was accepted with minor corrections within a period of one month.