Treffer: Efficient Study on Evaluating Processor's Affinity in Multi-Core Architecture and Multi-Processor Systems.

Processor's technologies are advancing at a fast pace, introducing higher computational power. However, this requires the presence of data closer to the processing unit in order to utilize the introduced computational power. Allowing processes to execute in any processor, results in exploiting the potential of the computational power of the processors, However, this diminishes the locality factor 1, in other words, every time a process is scheduled to execute, it is required to migrate its data to the new processor. This has two consequences: Firstly, the data should reside in the main memory all the time. Second, when the process finishes execution, it is required to update the copy of the data in the cache to the main memory, and to load the data in the next execution period. The significance of this paper comes from the fact that new systems are equipped with more processors and more cores per processor. In the near future, processors with many cores are going to be available commercially, hence, it is important to provide an insight on how to execute the processes on such processors. In this paper the processor Affinity helps in achieving better performance, as the process's data is expected to reside in the processors cache rather than the main memory. However, this might affect the performance, as limiting the number of processors that processes allowed to execute in, which might degrade the performance of the application. This paper evaluates the processor affinity (or core affinity) in order to measure the factors of locality and computational power on the processes execution. [ABSTRACT FROM AUTHOR]

Copyright of International Journal of Advances in Soft Computing & Its Applications is the property of Al-Zaytoonah University of Jordan, Faculty of Science & Information Technology and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)