A Cross-Layer Multicore Architecture to Tradeoff Program Accuracy and Resilience Overheads

被引:15
|
作者
Shi, Qingchuan [1 ]
Hoffmann, Henry [2 ]
Khan, Omer [1 ]
机构
[1] Univ Connecticut, Dept Elect & Comp Engn, Storrs, CT 06269 USA
[2] Univ Chicago, Dept Comp Sci, Chicago, IL 60637 USA
基金
美国国家科学基金会;
关键词
Resilience; soft-errors; program accuracy; multicores;
D O I
10.1109/LCA.2014.2365204
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
To protect multicores from soft-error perturbations, resiliency schemes have been developed with high coverage but high power/performance overheads (similar to 2x). We observe that not all soft-errors affect program correctness, some soft-errors only affect program accuracy, i.e., the program completes with certain acceptable deviations from soft-error free outcome. Thus, it is practical to improve processor efficiency by trading off resilience overheads with program accuracy. We propose the idea of declarative resilience that selectively applies resilience schemes to both crucial and non-crucial code, while ensuring program correctness. At the application level, crucial and non-crucial code is identified based on its impact on the program outcome. The hardware collaborates with software support to enable efficient resilience with 100 percent soft-error coverage. Only program accuracy is compromised in the worst-case scenario of a soft-error strike during non-crucial code execution. For a set of multithreaded benchmarks, declarative resilience improves completion time by an average of 21 percent over state-of-the-art hardware resilience scheme that protects all executed code. Its performance overhead is similar to 1.38x over a multicore that does not support resilience.
引用
收藏
页码:85 / 89
页数:5
相关论文
共 50 条
  • [1] Cross-Layer Resilience
    Mitra, Subhasish
    2016 21TH IEEE EUROPEAN TEST SYMPOSIUM (ETS), 2016,
  • [2] SATSIX Cross-layer Architecture
    Melhus, Inge
    Gayraud, Thierry
    Arnal, Fabrice
    Pietrabissa, Antonio
    Fan, Linghang
    2008 INTERNATIONAL WORKSHOP ON SATELLITE AND SPACE COMMUNICATIONS, CONFERENCE PROCEEDINGS, 2008, : 203 - +
  • [3] Design Techniques for Cross-Layer Resilience
    Carter, Nicholas P.
    Naeimi, Helia
    Gardner, Donald S.
    2010 DESIGN, AUTOMATION & TEST IN EUROPE (DATE 2010), 2010, : 1023 - 1028
  • [4] Cross-Layer Exploration of Heterogeneous Multicore Processor Configurations
    Sarma, Santanu
    Dutt, Nikil
    2015 28TH INTERNATIONAL CONFERENCE ON VLSI DESIGN (VLSID), 2015, : 147 - 152
  • [5] A cross-layer architecture for autonomic communications
    Razzaque, M. A.
    Dobson, Simon
    Nixon, Paddy
    AUTONOMIC NETWORKING, 2006, 4195 : 25 - 35
  • [6] The Resilience Wall: Cross-Layer Solution Strategies
    Mitra, Subhasish
    Bose, Pradip
    Cheng, Eric
    Cher, Chen-Yong
    Cho, Hyungmin
    Joshi, Rajiv
    Kim, Young Moon
    Lefurgy, Charles R.
    Li, Yanjing
    Rodbell, Kenneth P.
    Skadron, Kevin
    Stathis, James
    Szafaryn, Lukasz
    PROCEEDINGS OF TECHNICAL PROGRAM - 2014 INTERNATIONAL SYMPOSIUM ON VLSI TECHNOLOGY, SYSTEMS AND APPLICATION (VLSI-TSA), 2014,
  • [7] Cross-Layer System Resilience at Affordable Power
    Gupta, Meeta S.
    Rivers, Jude A.
    Wang, Liang
    Bose, Pradip
    2014 IEEE INTERNATIONAL RELIABILITY PHYSICS SYMPOSIUM, 2014,
  • [8] Cross-Layer Resilience Challenges: Metrics and Optimization
    Mitra, Subhasish
    Brelsford, Kevin
    Sanda, Pia N.
    2010 DESIGN, AUTOMATION & TEST IN EUROPE (DATE 2010), 2010, : 1029 - 1034
  • [9] The Resilience Wall: Cross-Layer Solution Strategies
    Mitra, Subhasish
    Bose, Pradip
    Cheng, Eric
    Cher, Chen-Yong
    Cho, Hyungmin
    Joshi, Rajiv
    Kim, Young Moon
    Lefurgy, Charles R.
    Li, Yanjing
    Rodbell, Kenneth P.
    Skadron, Kevin
    Stathis, James
    Szafaryn, Lukasz
    2014 INTERNATIONAL SYMPOSIUM ON VLSI DESIGN, AUTOMATION AND TEST (VLSI-DAT), 2014,
  • [10] Cross-Layer Error Resilience for Robust Systems
    Leem, Larkhoon
    Cho, Hyungmin
    Lee, Hsiao-Heng
    Kim, Young Moon
    Li, Yanjing
    Mitra, Subhasish
    2010 IEEE AND ACM INTERNATIONAL CONFERENCE ON COMPUTER-AIDED DESIGN (ICCAD), 2010, : 177 - 180