Global Signal Vulnerability (GSV) analysis for selective state element hardening in modern microprocessors

Mihalis Maniatakos, Chandrasekharan Tirumurti, Rajesh Galivanche, Yiorgos Makris

Research output: Contribution to journalArticle

Abstract

Global Signal Vulnerability (GSV) analysis is a novel method for assessing the susceptibility of modern microprocessor state elements to failures in the field of operation. In order to effectively allocate design for reliability resources, GSV analysis takes into account the high degree of architectural masking exhibited in modern microprocessors and ranks state elements accordingly. The novelty of this method lies in the way this ranking is computed. GSV analysis operates either at the Register Transfer (RT-) or at the Gate-Level, offering increased accuracy in contrast to methods which compute the architectural vulnerability of registers through high-level simulations on performance models. Moreover, it does not rely on extensive Statistical Fault Injection (SFI) campaigns and lengthy executions of workloads to completion in RT- or Gate-Level designs, which would make such analysis prohibitive. Instead, it monitors the behavior of key global microprocessor signals in response to a progressive stuck-at fault injection method during partial workload execution. Experimentation with the Scheduler and Reorder Buffer modules of an Alpha-like microprocessor and a modern Intel microprocessor corroborates that GSV analysis generates a near-optimal ranking, yet is several orders of magnitude faster than existing RT- or Gate-Level approaches.

Original languageEnglish (US)
Article number6280561
Pages (from-to)1361-1370
Number of pages10
JournalIEEE Transactions on Computers
Volume61
Issue number10
DOIs
StatePublished - Sep 7 2012

Fingerprint

Signal analysis
Microprocessor
Hardening
Vulnerability
Microprocessor chips
Fault Injection
Workload
Ranking
Masking
Performance Model
Scheduler
Experimentation
Susceptibility
Completion
Buffer
Monitor
Partial
Module
Resources
Simulation

Keywords

  • AVF
  • control logic
  • GSV
  • modern microprocessor
  • reliability
  • vulnerability analysis
  • workload

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Software
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

Global Signal Vulnerability (GSV) analysis for selective state element hardening in modern microprocessors. / Maniatakos, Mihalis; Tirumurti, Chandrasekharan; Galivanche, Rajesh; Makris, Yiorgos.

In: IEEE Transactions on Computers, Vol. 61, No. 10, 6280561, 07.09.2012, p. 1361-1370.

Research output: Contribution to journalArticle

Maniatakos, Mihalis ; Tirumurti, Chandrasekharan ; Galivanche, Rajesh ; Makris, Yiorgos. / Global Signal Vulnerability (GSV) analysis for selective state element hardening in modern microprocessors. In: IEEE Transactions on Computers. 2012 ; Vol. 61, No. 10. pp. 1361-1370.
@article{3da4f9d2630548b3b7d22e9aaa0129a5,
title = "Global Signal Vulnerability (GSV) analysis for selective state element hardening in modern microprocessors",
abstract = "Global Signal Vulnerability (GSV) analysis is a novel method for assessing the susceptibility of modern microprocessor state elements to failures in the field of operation. In order to effectively allocate design for reliability resources, GSV analysis takes into account the high degree of architectural masking exhibited in modern microprocessors and ranks state elements accordingly. The novelty of this method lies in the way this ranking is computed. GSV analysis operates either at the Register Transfer (RT-) or at the Gate-Level, offering increased accuracy in contrast to methods which compute the architectural vulnerability of registers through high-level simulations on performance models. Moreover, it does not rely on extensive Statistical Fault Injection (SFI) campaigns and lengthy executions of workloads to completion in RT- or Gate-Level designs, which would make such analysis prohibitive. Instead, it monitors the behavior of key global microprocessor signals in response to a progressive stuck-at fault injection method during partial workload execution. Experimentation with the Scheduler and Reorder Buffer modules of an Alpha-like microprocessor and a modern Intel microprocessor corroborates that GSV analysis generates a near-optimal ranking, yet is several orders of magnitude faster than existing RT- or Gate-Level approaches.",
keywords = "AVF, control logic, GSV, modern microprocessor, reliability, vulnerability analysis, workload",
author = "Mihalis Maniatakos and Chandrasekharan Tirumurti and Rajesh Galivanche and Yiorgos Makris",
year = "2012",
month = "9",
day = "7",
doi = "10.1109/TC.2011.172",
language = "English (US)",
volume = "61",
pages = "1361--1370",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "10",

}

TY - JOUR

T1 - Global Signal Vulnerability (GSV) analysis for selective state element hardening in modern microprocessors

AU - Maniatakos, Mihalis

AU - Tirumurti, Chandrasekharan

AU - Galivanche, Rajesh

AU - Makris, Yiorgos

PY - 2012/9/7

Y1 - 2012/9/7

N2 - Global Signal Vulnerability (GSV) analysis is a novel method for assessing the susceptibility of modern microprocessor state elements to failures in the field of operation. In order to effectively allocate design for reliability resources, GSV analysis takes into account the high degree of architectural masking exhibited in modern microprocessors and ranks state elements accordingly. The novelty of this method lies in the way this ranking is computed. GSV analysis operates either at the Register Transfer (RT-) or at the Gate-Level, offering increased accuracy in contrast to methods which compute the architectural vulnerability of registers through high-level simulations on performance models. Moreover, it does not rely on extensive Statistical Fault Injection (SFI) campaigns and lengthy executions of workloads to completion in RT- or Gate-Level designs, which would make such analysis prohibitive. Instead, it monitors the behavior of key global microprocessor signals in response to a progressive stuck-at fault injection method during partial workload execution. Experimentation with the Scheduler and Reorder Buffer modules of an Alpha-like microprocessor and a modern Intel microprocessor corroborates that GSV analysis generates a near-optimal ranking, yet is several orders of magnitude faster than existing RT- or Gate-Level approaches.

AB - Global Signal Vulnerability (GSV) analysis is a novel method for assessing the susceptibility of modern microprocessor state elements to failures in the field of operation. In order to effectively allocate design for reliability resources, GSV analysis takes into account the high degree of architectural masking exhibited in modern microprocessors and ranks state elements accordingly. The novelty of this method lies in the way this ranking is computed. GSV analysis operates either at the Register Transfer (RT-) or at the Gate-Level, offering increased accuracy in contrast to methods which compute the architectural vulnerability of registers through high-level simulations on performance models. Moreover, it does not rely on extensive Statistical Fault Injection (SFI) campaigns and lengthy executions of workloads to completion in RT- or Gate-Level designs, which would make such analysis prohibitive. Instead, it monitors the behavior of key global microprocessor signals in response to a progressive stuck-at fault injection method during partial workload execution. Experimentation with the Scheduler and Reorder Buffer modules of an Alpha-like microprocessor and a modern Intel microprocessor corroborates that GSV analysis generates a near-optimal ranking, yet is several orders of magnitude faster than existing RT- or Gate-Level approaches.

KW - AVF

KW - control logic

KW - GSV

KW - modern microprocessor

KW - reliability

KW - vulnerability analysis

KW - workload

UR - http://www.scopus.com/inward/record.url?scp=84865675531&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865675531&partnerID=8YFLogxK

U2 - 10.1109/TC.2011.172

DO - 10.1109/TC.2011.172

M3 - Article

AN - SCOPUS:84865675531

VL - 61

SP - 1361

EP - 1370

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 10

M1 - 6280561

ER -