Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Electronic digital computer (not a continuous role in computer ) > Arithmetic unit and the controller (CPU)

Temperature limits based on multi-core systems saving scheduling algorithm

Author XuXinJian
Tutor HanJianJun
School Huazhong University of Science and Technology
Course Applied Computer Technology
Keywords Real-Time Systems Multi-core processors Scheduling algorithm Dynamic Voltage Scaling Energy
CLC TP332
Type Master's thesis
Year 2011
Downloads 28
Quotes 0
Download Dissertation

In recent years, real-time system response time due to its reliability and strong features , has been widely used as aviation , military, transportation , and industrial and other fields. However, with the LSI process technology continues to evolve , processor power density and energy consumption increased , leading to distribute heat gain and reduce system stability and reliability. Currently, the processor and the temperature on energy issues has become a hot research field of computer . Currently, the system level , there are many problems to solve energy problems and temperature method, the real-time scheduling method and dynamic voltage regulation technology effectively a combination of methods is one of them . To meet the requirements of real-time tasks under the premise of the operating voltage by adjusting the processor so that the processor operates at an appropriate speed , can effectively reduce energy consumption. Processor temperature stabilization voltage related to their work , the use of dynamic voltage scaling processor operating voltage limitation , the processor can effectively alleviate the temperature problem . Real-time on-chip multiprocessor system environment, for tasks that depend on the collection , combined with static scheduling algorithm for optimal energy and temperature optimal static scheduling algorithm, is proposed with a temperature limit of the static energy-efficient scheduling algorithms. Thermal analysis algorithm based on a stable temperature , it would be with dynamic voltage scaling technology, making real-time tasks to meet deadlines at the same time , to reduce the energy consumption of the entire system and the maximum temperature . Respectively, and only consider the experiment temperature and energy consumption algorithm, the proposed algorithm can control the maximum temperature at the same time reducing system power consumption.

Related Dissertations
More Dissertations