Front Page

The Liberty Research Group

Automatic Instruction-Level Software-Only Recovery Methods [abstract] (IEEE Xplore, Original Full Paper, PDF)
George A. Reis, Jonathan Chang, and David I. August
IEEE Micro, Volume 27, Number 1, January 2007.
IEEE Micro's "Top Picks" special issue for papers "most relevant to industry and significant in contribution to the field of computer architecture" in 2006.

As chip densities and clock rates increase, processors are becoming more susceptible to transient faults that can affect program correctness. Computer architects have typically addressed reliability issues by adding redundant hardware, but these techniques are often too expensive to be used widely. Software-only reliability techniques have shown promise in their ability to protect against soft-errors without any hardware overhead. However, existing low-level software-only fault tolerance techniques have only addressed the problem of detecting faults, leaving recovery largely unaddressed. In this paper, we present the concept, implementation, and evaluation of automatic, instruction-level, software-only recovery techniques, as well as various specific techniques representing different trade-offs between reliability and performance. Our evaluation shows that these techniques fulfill the promises of instruction-level, software-only fault tolerance by offering a wide range of flexible recovery options