DOE Patents title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Masked fault detection for reliable low voltage cache operation

Abstract

Systems, apparatuses, and methods for implementing masked fault detection for reliable low voltage cache operation are disclosed. A processor includes a cache that can operate at a relatively low voltage level to conserve power. However, at low voltage levels, the cache is more likely to suffer from bit errors. To mitigate the bit errors occurring in cache lines at low voltage levels, the cache employs a strategy to uncover masked faults during runtime accesses to data by actual software applications. For example, on the first read of a given cache line, the data of the given cache line is inverted and written back to the same data array entry. Also, the error correction bits are regenerated for the inverted data. On a second read of the given cache line, if the fault population of the given cache line changes, then the given cache line's error protection level is updated.

Inventors:
;
Issue Date:
Research Org.:
Advanced Micro Devices, Inc., Santa Clara, CA (United States); Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1986850
Patent Number(s):
11509333
Application Number:
17/125,145
Assignee:
Advanced Micro Devices, Inc. (Santa Clara, CA)
DOE Contract Number:  
AC52-07NA27344; B620717
Resource Type:
Patent
Resource Relation:
Patent File Date: 12/17/2020
Country of Publication:
United States
Language:
English

Citation Formats

Ganapathy, Shrikanth, and Kalamatianos, John. Masked fault detection for reliable low voltage cache operation. United States: N. p., 2022. Web.
Ganapathy, Shrikanth, & Kalamatianos, John. Masked fault detection for reliable low voltage cache operation. United States.
Ganapathy, Shrikanth, and Kalamatianos, John. Tue . "Masked fault detection for reliable low voltage cache operation". United States. https://www.osti.gov/servlets/purl/1986850.
@article{osti_1986850,
title = {Masked fault detection for reliable low voltage cache operation},
author = {Ganapathy, Shrikanth and Kalamatianos, John},
abstractNote = {Systems, apparatuses, and methods for implementing masked fault detection for reliable low voltage cache operation are disclosed. A processor includes a cache that can operate at a relatively low voltage level to conserve power. However, at low voltage levels, the cache is more likely to suffer from bit errors. To mitigate the bit errors occurring in cache lines at low voltage levels, the cache employs a strategy to uncover masked faults during runtime accesses to data by actual software applications. For example, on the first read of a given cache line, the data of the given cache line is inverted and written back to the same data array entry. Also, the error correction bits are regenerated for the inverted data. On a second read of the given cache line, if the fault population of the given cache line changes, then the given cache line's error protection level is updated.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2022},
month = {11}
}

Works referenced in this record:

Yield-driven near-threshold SRAM design
conference, November 2007


Method for Testing a Memory and Memory System
patent-application, December 2014


Memory System
patent-application, July 2018


Near-Threshold Computing: Reclaiming Moore's Law Through Energy Efficient Integrated Circuits
journal, February 2010


A 400mV active VMIN, 200mV retention VMIN, 2.8 GHz 64Kb SRAM with a 0.09 um26T bitcell in a 16nm FinFET CMOS process
conference, June 2016


Bit Error Protection in Cache Memories
patent-application, March 2020


Trading off Cache Capacity for Reliability to Enable Low Voltage Operation
conference, June 2008


On Characterizing Near-Threshold SRAM Failures in FinFET Technology
conference, June 2017


Energy-efficient cache design using variable-strength error-correcting codes
conference, January 2011


Method and Apparatus for Using Compression to Improve Performance of Low Voltage Caches
patent-application, June 2020


Low-voltage on-chip cache architecture using heterogeneous cell sizes for high-performance processors
conference, February 2011


Archipelago: A polymorphic cache design for enabling robust near-threshold operation
conference, February 2011


Memory Controller and Method of Controlling Memory Controller
patent-application, May 2017


Low Vccmin fault-tolerant cache with highly predictable performance
conference, January 2009


Improving cache lifetime reliability at ultra-low voltages
conference, December 2009


Memory Controller and Method of Data Bus Inversion Using an Error Detection Correction Code
patent-application, June 2016


Memory mapped ECC
conference, June 2009


Soft error and energy consumption interactions
conference, August 2004


Killi: Runtime Fault Classification to Deploy Low Voltage Caches without MBIST
conference, February 2019


Base-delta-immediate compression: practical data compression for on-chip caches
conference, January 2012

  • Pekhimenko, Gennady; Seshadri, Vivek; Mutlu, Onur
  • Proceedings of the 21st international conference on Parallel architectures and compilation techniques - PACT '12
  • https://doi.org/10.1145/2370816.2370870

Operating SECDED-based caches at ultra-low voltage with FLAIR
conference, June 2013


Bit error protection in cache memories
patent, August 2019


Multi-bit Error Tolerant Caches Using Two-Dimensional Error Coding
conference, January 2007


A 45nm dual-port SRAM with write and read capability enhancement at low voltage
conference, January 2007


Reliable Ultra-Low-Voltage Cache Design for Many-Core Systems
journal, December 2012


Application of Special ECC Matrix for Solving Stuck Bit Faults in an ECC Protected Mechanism
patent-application, October 2004


Energy efficient near-threshold chip multi-processing
conference, August 2007


A 160 mV Robust Schmitt Trigger Based Subthreshold SRAM
journal, October 2007


Choosing an Error Protection Scheme for a Microprocessor's L1 Data Cache
conference, October 2006