Optimizing scheduling in cloud computing using the Cuckoo optimization algorithm | ||
| International Journal of Nonlinear Analysis and Applications | ||
| مقالات آماده انتشار، اصلاح شده برای چاپ، انتشار آنلاین از 28 تیر 1405 اصل مقاله (1.13 M) | ||
| نوع مقاله: Research Paper | ||
| شناسه دیجیتال (DOI): 10.22075/ijnaa.2024.35006.5225 | ||
| نویسنده | ||
| Kasra Rashidi* | ||
| Department of Computer, Isfahan University, Isfahan, Iran | ||
| چکیده | ||
| The cuckoo optimisation algorithm is a meta-exploratory optimization algorithm used to solve nonlinear continuous optimization problems. Optimization is to find a way of fulfilling a task in the best manner. Principally, optimization refers to changing a series of primary information and using problem information to achieve more appropriate responses. Recently, graphics processors have been proposed as a multi-purpose computational device due to low cost, parallel architecture and improved access provided by programming environments such as the CUDA framework. The Master-Slave Model is one of the parallelizing models of optimization algorithms. In the present paper, shared memory, reduction operations and other factors affecting GPUS have been employed using CUDA and the Master-Slave model with a fine-grain technique to increase the efficiency of the parallel algorithm. In this work, the implementation time of parallel and serial algorithms has been evaluated using a benchmark function. The experiments' results revealed an increase in speed-up of the parallel algorithm compared to the serial algorithm. | ||
| کلیدواژهها | ||
| optimizer algorithm؛ graphic processing units (GPU)؛ parallel and series algorithms؛ cloud computing | ||
| مراجع | ||
|
| ||
|
آمار تعداد مشاهده مقاله: 55 تعداد دریافت فایل اصل مقاله: 42 |
||
| تعداد نشریات | 22 |
| تعداد شمارهها | 723 |
| تعداد مقالات | 10,394 |
| تعداد مشاهده مقاله | 72,856,244 |
| تعداد دریافت فایل اصل مقاله | 64,553,645 |