Lecture
Schedule
Course overview
Computational science algorithms:
parallelism and locality
Graph algorithms: parallelism
and locality
Measurement: execution time and
hardware counters
DAG scheduling
Cache models
MMM and ATLAS
Vectorization (courtesy
Professor David Padua, UIUC)
Cache-coherent shared-memory
multiprocessors
Shared-memory programming: pThreads
and OpenMP
Memory consistency(I)
Memory consistency (II)
GPU programming
CUDA
Message-passing and MPI