Cache leakage power analysis in embedded applications

Power consumption in state-of-the-art embedded systems is determined to a large extent by on-chip caches. In these systems, technology scaling has resulted in an increase of subthreshold and gate leakage currents, which contribute significantly to the total cache power consumption. This underlines a need to analyze leakage power consumption and control techniques at every level of system design in embedded applications. In this paper, we have studied three individual leakage control techniques using Hotleakage as stated in ITRS (2001), applied at architectural level over a range of smaller technology sizes including 70nm. Specifically, evaluation of the effects of varying LI cache sizes (with the size of the other cache constant) demonstrated that the leakage power is directly proportional to the size of the LI cache. With increasing L2 cache sizes the leakage power increased almost to -50/spl deg/ of the total L2-cache power consumption for 70nm technology.

[1]  Trevor Mudge,et al.  Leakage power optimization techniques for ultra deep sub-micron multi-level caches , 2003, ICCAD-2003. International Conference on Computer Aided Design (IEEE Cat. No.03CH37486).

[2]  Trevor Mudge,et al.  MiBench: A free, commercially representative embedded benchmark suite , 2001 .

[3]  Koji Nii,et al.  A low power SRAM using auto-backgate-controlled MT-CMOS , 1998, Proceedings. 1998 International Symposium on Low Power Electronics and Design (IEEE Cat. No.98TH8379).

[4]  Richard T. Witek,et al.  A 160 MHz 32 b 0.5 W CMOS RISC microprocessor , 1996, 1996 IEEE International Solid-State Circuits Conference. Digest of TEchnical Papers, ISSCC.

[5]  David Blaauw,et al.  Drowsy caches: simple techniques for reducing leakage power , 2002, ISCA.

[6]  Todd M. Austin,et al.  The SimpleScalar tool set, version 2.0 , 1997, CARN.

[7]  Margaret Martonosi,et al.  Cache decay: exploiting generational behavior to reduce cache leakage power , 2001, ISCA 2001.

[8]  Dirk Grunwald,et al.  Pipeline gating: speculation control for energy reduction , 1998, ISCA.

[9]  Kevin Skadron,et al.  HotLeakage: A Temperature-Aware Model of Subthreshold and Gate Leakage for Architects , 2003 .