Dynamic Prediction of Architectural Vulnerability from Microarchitectural State

Report
Authors:Walcott, Kristen, Department of Computer ScienceUniversity of Virginia Humphreys, Greg, Department of Computer ScienceUniversity of Virginia Gurumurthi, Sudhanva, Department of Computer ScienceUniversity of Virginia
Abstract:

Transient faults due to particle strikes are a key challenge in microprocessor design. Driven by exponentially increas- ing transistor counts, per-chip faults are a growing burden. To protect against soft errors, redundancy techniques such as redundant multithreading (RMT) are often used. How- ever, these techniques assume that the probability that a structural fault will result in a soft error (i.e., the Archi- tectural Vulnerability Factor (AVF)) is 100 percent, unnec- essarily draining processor resources. Due to the high cost of redundancy, there have been efforts to throttle RMT at runtime. To date, these methods have not incorporated an AVF model and therefore tend to be ad hoc. Unfortunately, computing the AVF of complex microprocessor structures (e.g., the ISQ) can be quite involved.
To provide probabilistic guarantees about fault tolerance, we have created a rigorous characterization of AVF behav- ior that can be easily implemented in hardware. We ex- perimentally demonstrate AVF variability within and across the SPEC2000 benchmarks and identify strong correlations between structural AVF values and a small set of proces- sor metrics. Using these simple indicators as predictors, we create a proof-of-concept RMT implementation that demon- strates that AVF prediction can be used to maintain a low fault tolerance level without significant performance impact.

Rights:
All rights reserved (no additional license for public reuse)
Language:
English
Source Citation:

Walcott, Kristen, Greg Humphreys, and Sudhanva Gurumurthi. "Dynamic Prediction of Architectural Vulnerability from Microarchitectural State." University of Virginia Dept. of Computer Science Tech Report (2007).

Publisher:
University of Virginia, Department of Computer Science
Published Date:
2007