Header menu link for other important links
X
A review on evaluating the usage of multiple cores effectively in personal computer based applications
Published in International Journal of Pharmacy and Technology
2016
Volume: 8
   
Issue: 2
Pages: 3916 - 3927
Abstract
In the current Era all personally used computers (desktop, Laptop etc.,) are coming with multiple cores. But Current processors exploiting parallelism in a static manner and they are not assuring complete Parallelization on application running in it. This Paper Analyze the usage of multiple cores by various personal computer based Applications with respect to operating Hardware, software and middleware technologies. For the Effective use of available resources for computations, process allocation and de-allocation, each core and various types of parallelization techniques used with scalar applications running in Personal computing devices. A detailed analysis followed by a Evaluation proposal on the comparative Vectorization and optimal parallelization methodologies is needed. By creating a middleware between hardware and source code, Middle-ware also acts as a parallel program execution platform. A review section presented here gives the detailed Analysis on various parallel platforms, Software and algorithms used. OpenCL based implementation is used here because it uses kernel code execution and acts as User Interface. In general, OpenCL is used in GPGPU for high level processing, which executes multiple threads at GPU (Graphical Processing Unit). In this paper, a Comparative framework for ILP (Instruction Level Parallelism) which is not only proposing execution of code, but also proposes some techniques for parallelization of sequential code. To test this methodology, large size matrix based operations are used. After detailed analysis and observations, using chosen methodologies, vectorization of sequential code is executed in different parallel environments like OpenMP, OpenCL, and MATLAB environment. Finally program execution with and without Vectorization, using OpenMP, OpenCL and MATLAB is analysed. © 2016, International Journal of Pharmacy and Technology. All rights reserved.
About the journal
JournalInternational Journal of Pharmacy and Technology
PublisherInternational Journal of Pharmacy and Technology
ISSN0975766X