Introduction to multiprocessor systems, parallel programming models including Pthreads, MPI, hardware and software transactional memory, synchronization primitives, memory consistency mdels, cache coherence, on-chip shared cache architectures, on-chip interconnects, multi-chip interconnects, multi-chip bus-based and general-purpose interconnect-based shared-memory systems, clusters.
The course will include weekly readings, discussions, and student reviews and reports on publications (besides the text book) of seminal and recent contributions to the field of computer architecture. Student reviews, class discussions, and an independent research project will account for a significant fraction of the grade. Feedback on performance will be given only upon request by a student. There will be no recitation classes.
The course will also include an independent and original research project, in which students study, improve, and evaluate multiprocessor innovations using a software simulation infrastructure. There will be a list of project ideas given out, but students can suggest and work on their own ideas with potentials for advancing the state of the art.