A generic energy optimization framework for heterogeneous platforms using scaling models

Ujjwal Gupta, Spurthi Korrapati, Navyasree Matturu, Umit Ogras

Research output: Contribution to journalArticle

10 Scopus citations

Abstract

Mobile platforms are becoming highly heterogeneous by combining a powerful multiprocessor system-on-a-chip (MpSoC) with numerous other resources, including display, memory, power management IC, battery and wireless modems into a compact package. Furthermore, the MpSoC itself is a heterogeneous resource that integrates many processing elements such as CPU cores, GPU, video, image, and audio processors. Platform energy consumption and responsiveness are two major considerations for mobile systems, since they determine the battery life and user satisfaction, respectively. As a result, energy minimization approaches targeting mobile computing need to consider the platform at various levels of granularity. In this paper, we first present power consumption, response time, and energy consumption models for mobile platforms. Using these models, we optimize the energy consumption of baseline platforms under power, response time, and thermal constraints with and without introducing new resources. Finally, we validate the proposed framework through experiments on Qualcomm's Snapdragon 800 Mobile Development Platforms.

Original languageEnglish (US)
JournalMicroprocessors and Microsystems
DOIs
StateAccepted/In press - 2015

Keywords

  • Energy optimization
  • Heterogeneous architectures
  • Mobile platforms
  • MpSoC
  • Performance

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'A generic energy optimization framework for heterogeneous platforms using scaling models'. Together they form a unique fingerprint.

  • Cite this