Interactive presentation: PowerQuest: trace driven data mining for power optimization

We introduce a general framework, called PowerQuest, with the primary goal of extracting "interesting" dynamic invariants from a given simulation-trace database, and applying it to the power-reduction problem through detection of gating conditions. PowerQuest adopts machine-learning techniques for data mining. The advantages of PowerQuest in comparison with other state-of-the-art Dynamic Power Management (DPM) techniques are: 1) Quality of ODC conditions for gating 2) Minimization of extra logic added for gating. We demonstrate the validity of our approach in reducing power through experimental results using ITC99 benchmarks and real-life microprocessor test cases. We present up to 22.7 % power reduction in comparison with other DPM techniques.

[1]  Jaime G. Carbonell,et al.  Machine learning: paradigms and methods , 1990 .

[2]  Mani B. Srivastava,et al.  Predictive system shutdown and other architectural techniques for energy efficient programmable computation , 1996, IEEE Trans. Very Large Scale Integr. Syst..

[3]  Giovanni De Micheli,et al.  Don't care set specifications in combinational and synchronous logic circuits , 1993, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[4]  Norbert Wehn,et al.  Automating RT-level operand isolation to minimize power consumption in datapaths , 2000, DATE '00.

[5]  Luca Benini,et al.  Dynamic power management - design techniques and CAD tools , 1997 .

[6]  José C. Monteiro,et al.  Sequential logic optimization for low power using input-disabling precomputation architectures , 1998, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[7]  Sridhar Narayanan,et al.  IODINE: a tool to automatically infer dynamic invariants for hardware designs , 2005, Proceedings. 42nd Design Automation Conference, 2005..

[8]  Jiarong Hong,et al.  AE1: An extension matrix approximate method for the general covering problem , 1985, International Journal of Computer & Information Sciences.

[9]  Xindong Wu.  Rule induction with extension matrices , 1998, KDD 1998.

[10]  Marios C. Papaefthymiou,et al.  Precomputation-based sequential logic optimization for low power , 1994, IEEE Trans. Very Large Scale Integr. Syst..

[11]  Luca Benini,et al.  A scalable algorithm for RTL insertion of gated clocks based on ODCs computation , 2005 .

[12]  William G. Griswold,et al.  Dynamically discovering likely program invariants to support program evolution , 1999, Proceedings of the 1999 International Conference on Software Engineering (IEEE Cat. No.99CB37002).