Reducing the leakage of memory blocks aggressively

Excessive leakage of cache blocks encourages forcing the cache block into low-leakage state all the time except during read/write operations. By retaining the values of the cache cells during low leakage state, the penalty for re-activating the sleepy cache cells is also reduced. This is achieved using aggressive policies with simple control strategies using one of two methods. Either by adding a low voltage supply, operating the cache at low voltage all the time and enabling normal voltage only during a read/write operation to the cache line or by adding a ground gating NMOS transistor, which cuts the path from the supply voltage to ground all the time except during a read/write operation to the cache line. Both methods succeed to reduce leakage power during idle state by around 90% using minimum hardware control overhead, but suffer from process variation vulnerability and lower SNM.

[1]  E. Seevinck,et al.  Static-noise margin analysis of MOS SRAM cells , 1987 .

[2]  Mahmut T. Kandemir,et al.  Leakage Current: Moore's Law Meets Static Power , 2003, Computer.

[3]  Dalia A. El-Dib Leakage power considerations in actively running blocks , 2010, 2010 53rd IEEE International Midwest Symposium on Circuits and Systems.

[4]  T. N. Vijaykumar,et al.  Gated-V/sub dd/: a circuit technique to reduce leakage in deep-submicron cache memories , 2000, ISLPED'00: Proceedings of the 2000 International Symposium on Low Power Electronics and Design (Cat. No.00TH8514).

[5]  W. Dehaene,et al.  Read Stability and Write-Ability Analysis of SRAM Cells for Nanometer Technologies , 2006, IEEE Journal of Solid-State Circuits.

[6]  Kaushik Roy,et al.  A single-Vt low-leakage gated-ground cache for deep submicron , 2003, IEEE J. Solid State Circuits.

[7]  Z. Abid,et al.  Investigating an aggressive mode for drowsy cache cells , 2008, 2008 Canadian Conference on Electrical and Computer Engineering.

[8]  T. Mudge,et al.  Drowsy caches: simple techniques for reducing leakage power , 2002, Proceedings 29th Annual International Symposium on Computer Architecture.

[9]  Narayanan Vijaykrishnan,et al.  On load latency in low-power caches , 2003, ISLPED '03.