Quantifying software vulnerability
Proceedings of the 2008 workshop on Radiation effects and fault tolerance in nanometer technologies
Online Estimation of Architectural Vulnerability Factor for Soft Errors
ISCA '08 Proceedings of the 35th Annual International Symposium on Computer Architecture
Phaser: phased methodology for modeling the system-level effects of soft errors
IBM Journal of Research and Development
Architecture Design for Soft Errors
Architecture Design for Soft Errors
Power-efficient, reliable microprocessor architectures: modeling and design methods
Proceedings of the 20th symposium on Great lakes symposium on VLSI
Analysis of single-event effects in embedded processors for non-uniform fault tolerant design
IIT'09 Proceedings of the 6th international conference on Innovations in information technology
MICRO '43 Proceedings of the 2010 43rd Annual IEEE/ACM International Symposium on Microarchitecture
Soft error benchmarking of L2 caches with PARMA
Proceedings of the ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Soft error benchmarking of L2 caches with PARMA
ACM SIGMETRICS Performance Evaluation Review - Performance evaluation review
Nanoscale technologies: prospect or hazard to dependable and secure computing?
LADC'07 Proceedings of the Third Latin-American conference on Dependable Computing
| Hi-index | 0.00 |
This paper concerns the validity of a widely used method for estimating the architecture-level mean time to failure (MTTF) due to soft errors. The method first calculates the failure rate for an architecture-level component as the product of its raw error rate and an architecture vulnerability factor (AVF). Next, the method calculates the system failure rate as the sum of the failure rates (SOFR) of all components, and the system MTTF as the reciprocal of this failure rate. Both steps make significant assumptions. We investigate the validity of the AVF+SOFR method across a large design space, using both mathematical and experimental techniques with real program traces from SPEC 2000 benchmarks and synthesized traces to simulate longer real-world workloads. We show that AVF+SOFR is valid for most of the realistic cases under current raw error rates. However, for some realistic combinations of large systems, long-running workloads with large phases, and/or large raw error rates, the MTTF calculated using AVF+SOFR shows significant-discrepancies from that using first principles. We also show that SoftArch, a previously proposed alternative method that does not make the AVF+SOFR assumptions, does not exhibit the above discrepancies.