Browsing by Author "Kizil, Alper"

Now showing 1 - 2 of 2

Citation - Scopus: 3
Effects of Parameters of an Island Model Parallel Genetic Algorithm for the Quadratic Assignment Problem
(Institute of Electrical and Electronics Engineers Inc., 2019) Alper Kizil; Korhan Karabulut; Kizil, Alper; Karabulut, Korhan
Quadratic Assignment Problem (QAP) is one of the most difficult combinatorial problems in the literature and has a diverse field of applications. This paper presents the results of experiments on the impact of parallelization of a sequential GA using island model. Both of the genetic algorithms are applied to the QAP. For the island model parallel GA we systematically change the number of islands and investigate the effects of dividing the same global population into a number of subpopulations. The number of islands is gradually increased to observe the effects on solution quality and speedup in total execution time using different problem instances. The results clearly indicate that while parallelized version outperforms sequential counterpart in both solution quality and total execution time an increasing number of subpopulations also positively effects the results until a critical point where every subpopulation has a certain number of individuals to be able to evolve independently. Beyond that point the performance of the algorithm begins to decrease. © 2020 Elsevier B.V. All rights reserved.
Makespan and energy based virtual machine scheduling in cloud systems, Bulut sistemlerinde toplam tamamlanma ve enerji tabanli sanal makine çizelgelemesi
(Gazi Universitesi, 2024) Alper Kizil; Korhan Karabulut; Kizil, Alper; Karabulut, Korhan
Cloud computing is one of the newest computing paradigms that emerged after worldwide development of Internet infrastructure. It is a technology that saves both large companies and small and medium scale companies as well as independent developers from the cost of keeping infrastructure hardware up to date and operational while also providing flexibility on resource use as well providing additional opportunity to minimize data losses. While in the future it is evident that demand for cloud computing will be on the rise. These kinds of datacenters due to their nature consume large amounts of energy and even the savings on smallest scales will enable these gigantic centers to save a significant amount of energy in total. If we have a look at the literature we can see green computing is gaining immense popularity over the years. The Cloud Scheduling problem is a proven problem to be NP-Hard aiming to find the best solution for a limited number of cloud resources which could theoretically be serving an unlimited number of user demands. In this study firstly an experimental workload / power consumption model is proposed for a server computer and then two genetic algorithms optimizing makespan and energy consumption are compared on these metrics at different server loads. As a result it has been seen that these two criteria are closely related to each other and it has been determined that optimizing the energy criterion has a more positive effect between 10% and 13% compared to the time criterion optimization at full or near full server loads. In this way it has been shown that significant energy savings can be achieved by using energy optimization as an objective function at high server loads. © 2024 Elsevier B.V. All rights reserved.