Enabling Near-Threshold Voltage(NTV) operation in Multi-VDD cache for power reduction

Power constraint for modern processors becomes a very serious problem with the increasing core counts and cache capacity in multi/many core processors. Compared with processing cores with mature techniques like DVFS to alleviate the situation, last level cache which consume largest portion of processor chip needs effective power management strategy. In this paper, we explore the feasibility of Near-Threshold Voltage(NTV) SRAM and Multi-Voltage Domain (Multi-VDD) for power reduction in large capacity cache. To prevent data corruption in cache, we propose redundancy-based data salvaging technique for fault recovery. To solve the dilemma of power reduction and reliability guarantee, we try to match vulnerable/invulnerable data sets to high/low voltage domains. Different from previous work, we take into consideration multi-bit errors and redundancy masking effects in Multi-VDD cache. Experimental results show that our Multi-VDD cache achieves considerable improvements in energy efficiency.

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

[2]  Yu Hu,et al.  Theoretic analysis and enhanced X-tolerance of test response compact based on convolutional code , 2005, ASP-DAC.

[3]  Jan M. Rabaey,et al.  SRAM supply voltage scaling: A reliability perspective , 2009, 2009 10th International Symposium on Quality Electronic Design.

[4]  Milo M. K. Martin,et al.  Multifacet's general execution-driven multiprocessor simulator (GEMS) toolset , 2005, CARN.

[5]  Yu Hu,et al.  Embedded Test Decompressor to Reduce the Required Channels and Vector Memory of Tester for Complex Processor Circuit , 2007, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[6]  Xiaowei Li,et al.  A unified online Fault Detection scheme via checking of Stability Violation , 2009, 2009 Design, Automation & Test in Europe Conference & Exhibition.

[7]  Takayasu Sakurai,et al.  24% Power reduction by post-fabrication dual supply voltage control of 64 voltage domains in VDDmin limited ultra low voltage logic circuits , 2012, Thirteenth International Symposium on Quality Electronic Design (ISQED).

[8]  H.S. Kim,et al.  Device-Orientation Effects on Multiple-Bit Upset in 65 nm SRAMs , 2008, IEEE Transactions on Nuclear Science.

[9]  P. Reviriego,et al.  Reliability Analysis of Memories Suffering Multiple Bit Upsets , 2007, IEEE Transactions on Device and Materials Reliability.

[10]  Rajiv K. Kalia,et al.  Large-scale atomistic modeling of nanoelectronic structures , 2000 .

[11]  R. Harboe-Sorensen,et al.  Multiple-Bit Upset Analysis in 90 nm SRAMs: Heavy Ions Testing and 3D Simulations , 2007, IEEE Transactions on Nuclear Science.

[12]  Arijit Biswas,et al.  Computing architectural vulnerability factors for address-based structures , 2005, 32nd International Symposium on Computer Architecture (ISCA'05).

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

[14]  S. Nassif,et al.  Analytical Modeling of SRAM Dynamic Stability , 2006, 2006 IEEE/ACM International Conference on Computer Aided Design.

[15]  J. Draper,et al.  Parallel double error correcting code design to mitigate multi-bit upsets in SRAMs , 2008, ESSCIRC 2008 - 34th European Solid-State Circuits Conference.

[16]  P. Roche,et al.  Heavy Ion Testing and 3-D Simulations of Multiple Cell Upset in 65 nm Standard SRAMs , 2008, IEEE Transactions on Nuclear Science.

[17]  Joel Emer,et al.  Computing Architectural Vulnerability Factors for Address-Based Structures , 2005, ISCA 2005.