Low Vccmin fault-tolerant cache with highly predictable performance

Transistors per area unit double in every new technology node. However, the electric field density and power demand grow if Vcc is not scaled. Therefore, Vcc must be scaled in pace with new technology nodes to prevent excessive degradation and keep power demand within reasonable limits. Unfortunately, low Vcc operation exacerbates the effect of variations and decreases noise and stability margins, increasing the likelihood of errors in SRAM memories such as caches. Those errors translate into performance loss and performance variation across different cores, which is especially undesirable in a multi-core processor. This paper presents (i) a novel scheme to tolerate high faulty bit rates in caches by disabling only faulty subblocks, (ii) a dynamic address remapping scheme to reduce performance variation across different cores, which is key for performance predictability, and (iii) a comparison with state-of-the-art techniques for faulty bit tolerance in caches. Results for some typical first level data cache configurations show 15% average performance increase and standard deviation reduction from 3.13% down to 0.55% when compared to cache line disabling schemes.

[1]  Israel Koren,et al.  Defect tolerance in VLSI circuits: techniques and yield analysis , 1998, Proc. IEEE.

[2]  Chin-Long Chen,et al.  Error-Correcting Codes for Semiconductor Memory Applications: A State-of-the-Art Review , 1984, IBM J. Res. Dev..

[3]  Alaa R. Alameldeen,et al.  Trading off Cache Capacity for Reliability to Enable Low Voltage Operation , 2008, 2008 International Symposium on Computer Architecture.

[4]  James D. Meindl,et al.  Impact of die-to-die and within-die parameter fluctuations on the maximum clock frequency distribution for gigascale integration , 2002, IEEE J. Solid State Circuits.

[5]  Jaume Abella,et al.  Power efficient data cache designs , 2003, Proceedings 21st International Conference on Computer Design.

[6]  M. Motoyoshi,et al.  A novel 6T-SRAM cell technology designed with rectangular patterns scalable beyond 0.18 /spl mu/m generation and desirable for ultra high speed operation , 1998, International Electron Devices Meeting 1998. Technical Digest (Cat. No.98CH36217).

[7]  Rochit Rajsuman.  Design and Test of Large Embedded Memories: An Overview , 2001, IEEE Des. Test Comput..

[8]  Diana Marculescu,et al.  Power efficiency of voltage scaling in multiple clock, multiple voltage cores , 2002, ICCAD 2002.

[9]  Kaushik Roy,et al.  A 160 mV, fully differential, robust schmitt trigger based sub-threshold SRAM , 2007, Proceedings of the 2007 international symposium on Low power electronics and design (ISLPED '07).

[10]  Y. Nakagome,et al.  Trends in low-power RAM circuit technologies , 1995 .

[11]  J. Jopling,et al.  Erratic fluctuations of sram cache vmin at the 90nm process technology node , 2005, IEEE InternationalElectron Devices Meeting, 2005. IEDM Technical Digest..

[12]  R. Krishnamurthy,et al.  A 4 GHz 130 nm address generation unit with 32-bit sparse-tree adder core , 2002, 2002 Symposium on VLSI Circuits. Digest of Technical Papers (Cat. No.02CH37302).

[13]  H. J. Arnold.  Introduction to the Practice of Statistics , 1990 .

[14]  Pankaj Agarwal,et al.  A low leakage and SNM free SRAM cell design in deep sub micron CMOS technology , 2006, 19th International Conference on VLSI Design held jointly with 5th International Conference on Embedded Systems Design (VLSID'06).

[15]  Rohit Bhatia,et al.  Montecito: a dual-core, dual-thread Itanium processor , 2005, IEEE Micro.

[16]  Tadahiro Kuroda,et al.  Variable supply-voltage scheme for low-power high-speed CMOS digital design , 1998, IEEE J. Solid State Circuits.

[17]  Michael L. Scott,et al.  Dynamic frequency and voltage control for a multiple clock domain microarchitecture , 2002, MICRO.

[18]  Zhiyu Liu,et al.  High Read Stability and Low Leakage Cache Memory Cell , 2007, 2007 IEEE International Symposium on Circuits and Systems.

[19]  B. M. Gordon,et al.  Supply and threshold voltage scaling for low power CMOS , 1997, IEEE J. Solid State Circuits.

[20]  D. Blaauw,et al.  Single-V/sub DD/ and single-V/sub T/ super-drowsy techniques for low-leakage high-performance instruction caches , 2004, Proceedings of the 2004 International Symposium on Low Power Electronics and Design (IEEE Cat. No.04TH8758).

[21]  David Blaauw,et al.  Drowsy instruction caches: leakage power reduction using dynamic voltage scaling and cache sub-bank prediction , 2002, MICRO.

[22]  Trevor N. Mudge,et al.  On-Chip Cache Device Scaling Limits and Effective Fault Repair Techniques in Future Nanoscale Technology , 2007, 10th Euromicro Conference on Digital System Design Architectures, Methods and Tools (DSD 2007).