The project aims to improve the quality of service and lifetime of real-time embedded systems, particularly those implemented on multi-core platforms. In contrast with many other quality metrics, such as performance and power consumption, reliability is difficult to accurately estimate because it is influenced by design decisions, environmental conditions, and process variation during integrated circuit fabrication.

This project consists of three main technical tasks. (1) Develop a reliability modeling and analysis framework that can efficiently and accurately determine the impact of design and runtime management decisions on reliability. (2) Develop a reliability-driven resource management framework, which includes runtime algorithms for assignment and scheduling of real-time tasks to maximize system lifetime while keeping soft error rates low, and a lightweight technique to adaptively adjust the activation frequency of the algorithms (i.e., overhead). (3) Develop wear state monitoring techniques and data collection infrastructure to enable the runtime refinement and validation of system-level reliability models that require long-term in-field system deployment.

The techniques developed in this project will support the production of more reliable and/or less expensive electronic devices, enabling integrated circuits, which are susceptible to wear due to lifetime fault processes to be used in special-purpose computers with strict reliability and performance requirements. In particular, this project aims to ease the use of multicore processors in high-reliability computing applications with deadlines, such as automotive, multimedia, and health care applications. These applications have historically seen slow adoption of multicore processors, despite their price, performance, and power consumption benefits. We believe this is partially due to gaps in knowledge of how to control and optimize reliability on such systems, some of which this project will fill. The involvement of both industry researchers and university students at the undergraduate and graduate level will result in a broad dissemination of the research results.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Network Systems (CNS)
Type
Standard Grant (Standard)
Application #
1319718
Program Officer
M. Mimi McClure
Project Start
Project End
Budget Start
2013-10-01
Budget End
2016-09-30
Support Year
Fiscal Year
2013
Total Cost
$166,222
Indirect Cost
Name
Utah State University
Department
Type
DUNS #
City
Logan
State
UT
Country
United States
Zip Code
84322