The growing complexity and shrinking geometries of modern manufacturing technologies are making high–density, low–voltage devices increasingly susceptible to the influences of electrical noise, process variation, transistor aging, and the effects of natural radiation. The system–level impact of these errors can be far–reaching. Growing concern about intermittent errors, unstable storage cells, and the effects of aging are influencing system design and failures in memories account for a significant fraction of costly product returns. Emerging logic and memory device technologies introduce several reliability challenges that need to be addressed to make these technologies viable. Finally, reliability is a key issue for large–scale systems, such as those in data centers. The SELSE workshop provides a forum for discussion of current research and practice in system–level error management. Participants from industry and academia explore both current technologies and future research directions (including nanotechnology). SELSE is soliciting papers that address the system–level effects of errors from a variety of perspectives: architectural, logical, circuit–level, and semiconductor processes. Case studies are also solicited.<br>Key areas of interest are (but not limited to):<br>Technology trends and the impact on error rates.<br>New error mitigation techniques.<br>Characterizing the overhead and design complexity of error mitigation techniques.<br>Case studies describing the tradeoffs analysis for reliable systems.<br>Experimental silicon failure data.<br>System–level models: derating factors and validation of error models.<br>Error handling protocols (higher–level protocols for robust system design).<br>Characterization of reliability of systems deployed in the field and mitigation of issues.<br>
Abbrevation
SELSE
City
Boston
Country
United States
Deadline Paper
Start Date
End Date
Abstract