Read buffer optimizations to support compiler-assisted multiple instruction retryMultiple instruction retry is a recovery mechanism for transient processor faults. We previously developed a compiler-assisted approach to multiple instruction ferry in which a read buffer of size 2N (where N represents the maximum instruction rollback distance) was used to resolve some data hazards while the compiler resolved the remaining hazards. The compiler-assisted scheme was shown to reduce the performance overhead and/or hardware complexity normally associated with hardware-only retry schemes. This paper examines the size and design of the read buffer. We establish a practical lower bound and average size requirement for the read buffer by modifying the scheme to save only the data required for rollback. The study measures the effect on the performance of a DECstation 3100 running ten application programs using six read buffer configurations with varying read buffer sizes. Two alternative configurations are shown to be the most efficient and differed depending on whether split-cycle-saves are assumed. Up to a 55 percent read buffer size reduction is achievable with an average reduction of 39 percent given the most efficient read buffer configuration and a variety of applications.
Document ID
19930067967
Acquisition Source
Legacy CDMS
Document Type
Conference Paper
Authors
Alewine, N. J. (NASA Langley Research Center Hampton, VA, United States)
Fuchs, W. K. (NASA Langley Research Center Hampton, VA, United States)
Hwu, W. M. (Illinois Univ. Urbana, United States)
Date Acquired
August 16, 2013
Publication Date
June 1, 1993
Subject Category
Computer Systems
Meeting Information
Meeting: IEEE, FTCS 23 - International Symposium on Fault-Tolerant Computing