Utilizing Predictors for Efficient Thermal Management in Multiprocessor SoCs

Abstract

Conventional thermal management techniques are reactive, as they take action after temperature reaches a threshold. Such approaches do not always minimize and balance the temperature, and they control temperature at a noticeable performance cost. This paper investigates how to use predictors for forecasting temperature and workload dynamics, and proposes proactive thermal management techniques for multiprocessor system-on-chips. The predictors we study include autoregressive moving average modeling and lookup tables. We evaluate several reactive and predictive techniques on an UltraSPARC T1 processor and an architecture-level simulator. Proactive methods achieve significantly better thermal profiles and performance in comparison to reactive policies.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2009
Accession Number
ADA542649

Entities

People

  • Ayse K. Coskun
  • Kenny C. Gross
  • Tajana S. Rosing

Organizations

  • University of California, San Diego

Tags

Communities of Interest

  • Advanced Electronics
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Computational Science
  • Computer Programs
  • Computers
  • Data Science
  • Databases
  • Energy Consumption
  • Information Science
  • Measurement
  • Network Science
  • Operating Systems
  • Reliability
  • Simulations
  • Simulators
  • Stationary Processes
  • Temperature Control
  • Workload

Readers

  • Combustion science or combustion engineering.
  • Parallel and Distributed Computing.
  • Regression Analysis.