ICSA Colloquium Talk - 05/02/2019

Talk title: ParaMedic: Parallel Error Detection and Correction

Abstract: Microprocessor error detection is increasingly important, as the number of transistors in modern systems heightens their vulnerability. However, current detection techniques require duplication of all hardware structures, causing a considerable increase in power consumption and chip area. ParaMedic is a novel hardware-only solution for error detection and correction, that exploits parallelism in checking code which may not exist in the original execution. We pair a high-performance out-of-order core with a set of small low-power cores, each of which checks a portion of the out-of-order core’s execution. The system enables the detection of both hard and soft errors and full correction of the latter with low area, power and performance overheads.

Bio: Timothy Jones is a Reader in Computer Architecture and Compilation at the University of Cambridge Computer Laboratory and a Fellow of Gonville and Caius College. He received an MEng degree in Computer Science from the University of Bristol and a PhD in Informatics from the University of Edinburgh. He has previously held a Postdoctoral Research Fellowship from the Royal Academy of Engineering and EPSRC, and an EPSRC Early Career Fellowship, and has been a visiting researcher at Harvard University and Arm. His research interests include computer architecture and compiler optimisation, with a focus on extracting all forms of parallelism, security, and reliability.

http://www.cl.cam.ac.uk/~tmj32/