Abstract ▼ AI-Helper

The spread of mobile devices in people's lives has brought many changes, these changes in the demand for the mobile content market is also gradually spread. Using a mobile device is a mobile content service, that is to say the mobile application. People demand mobile applications by creating a docum...

The spread of mobile devices in people's lives has brought many changes, these changes in the demand for the mobile content market is also gradually spread. Using a mobile device is a mobile content service, that is to say the mobile application. People demand mobile applications by creating a document, games, photos, banking, movies, ringtones such as in various aspects of services. With the increasing demand for those HD video playback, AV streaming service, multi-tasking, web browsing, 3D gaming and other applications that demand high-performance hardware is emerged. In these applications, depending on the supply of the mobile device is currently in the performance of the core having reached octa-core device is being deployed. However, the effectiveness of multicores depends on how well a scheduler can assign tasks onto the cores. In a heterogeneous multicore platform, the execution time of an application depends on which core it executes on. This work proposes a task scheduling technique that analyzes execution time of each task by profiling. The profiling provides a basic information to predict which task-to-core mapping is likely to provide the best performance. By using such information, the proposed technique is efficiently able to get the best performance and to keep load balancing among processing cores.

Keyword