Software Power Analysis And Optimization For Power-Aware Multicore Systems
نویسنده
چکیده
SOFTWARE POWER ANALYSIS AND OPTIMIZATION FOR POWER-AWAREMULTICORE SYSTEMSbySHINANWANGMay 2014 Advisor: Dr. Weisong ShiMajor: Computer ScienceDegree: Doctor of Philosophy Among all the factors in sustainable computing, power dissipation and energy consump-tion, arguably speaking, are fundamental aspects of modern computer systems. Different fromperformance metric, power dissipation is not easy to measure because hardware instrumenta-tion is usually required. Yet as an indispensable component of a computer system, softwarebecomes a major factor affecting power dissipation besides hardware energy-efficiency andpower states. With detailed information on resource usage and power dissipation of an appli-cation/software, software developers will be able to leverage algorithms and implementationsin order to produce power-efficient solutions. Hardware instrumentation, despite its accuracy,is costly and complicated to set up. A general solution to connect software with hardware alongwith detailed power and system information will improve the system overall efficiency.In this work, we design and implement a general solution to analyze and model softwarepower dissipation. Based on the analysis, we propose a combined solution to optimize the en-ergy efficiency of parallel workload. Starting from the hands-on power measurement methodin detail, we provide a fine-grain power profile of two computer systems using hardware in-strumentation. Being focusing on dynamic power dissipation analysis, we propose a two-levelpower model for power-aware multicore computer systems. Based on the model, we designand implement SPAN to relate power dissipation to the different portions of an application
منابع مشابه
Design of a novel congestion-aware communication mechanism for wireless NoC architecture in multicore systems
Hybrid Wireless Network-on-Chip (WNoC) architecture is emerged as a scalable communication structure to mitigate the deficits of traditional NOC architecture for the future Multi-core systems. The hybrid WNoC architecture provides energy efficient, high data rate and flexible communications for NoC architectures. In these architectures, each wireless router is shared by a set of processing core...
متن کاملDesign and analysis of hybrid systems solar, wind, osmotic for green plants using ant colony optimization algorithm
Nature has always proven that it is able to overcome its problems. However, human manipulation has led to environmental degradations. The dryness of a thousand-year Urmia Lake (a brinewater lake in Iran) is an example of environmental degradation that happened due to successive droughts and construction of dams on the basin of this lake. This study examines methods for the revival of Urmia Lake...
متن کاملEcoMobile: Energy-aware Real-time Framework for Multicore Mobile Systems
In this paper, we propose a design and development plan to build an energy-aware framework for multicore mobile systems which extends Android with two features it lacks: fine-grain power management functionality and real-time support. We also provide three novel energy-aware real-time multicore scheduling algorithms to reduce both dynamic and leakage power consumptions. These algorithms address...
متن کاملCPPM: a Comprehensive Power-aware Processor Manager for a Multicore System
The growing functionality of mobile devices explains increasing system performance requirements and the subsequent wide adoption of multicore processors. As mobile systems are battery powered, battery life largely limits these high performing multicore mobile devices. Developing an efficient power-aware processor manager for mobile multicore systems has received considerable attention. The conv...
متن کاملA Survey on Low-Power Techniques for Single and Multicore Systems
This paper surveys state of the art low-power techniques for both single and multicore systems. Based on our proposed power management model for multicore systems, we present a classification of total power reduction techniques including both leakage and active power. According to this classification, three main classes are discussed: power optimization techniques within the cores, techniques f...
متن کامل