Multi-core processors are, in effect, multiprocessing systems on a chip. Consequently, embedded developers must graduate from a serial execution model, where software tasks take turns running on a single processor, to a parallel execution model, where multiple software tasks can run simultaneously. The more parallelism developers achieve, the better their multicore systems perform. This paper looks at how one development team minimized its migration efforts while maximizing system throughput.
Paul Leroux is an award-winning writer, technology analyst, and public relations manager. He has been with QNX Software Systems for more than 20 years, where he is particularly valued for his knowledge of the embedded industry and for his ability to explain complex technical issues to both specialists and non-specialists. Leroux maintains the blog On Q, which he describes as "a personal mashup of QNX, cars, embedded systems, M2M, and photography." When he is not writing whitepapers or blogging, Leroux is an avid snowshoer and wildlife photographer. He holds a B.A. and an M.A., both from Concordia University in Montreal.
02/01/07
Paul Leroux