Tags

instructors seminar get_involved atomic Chip multiprocessor cilk cilk_for cilk_scope cilk_spawn cilk_sync core cpu Cilksan span critical section data race deadlock determincacy race determinism distributed memory execution time fake lock false sharing global variable hyperobject instruction knot lock linear speedup multicore multiprocessor nondeterminism mutex opencilk nonlocal variable parallel loop parallelism perfect linear speedup process processor race condition receiver running time reducer response time scale down scale up serial execution sequential consistency serial projection serial semantics shared memory software performance engineering spawn strand speedup sync task-parallel thread throughput view work work stealing worker events competition concurrent agent lock contention commutative operation