This course covers topics related to network-based parallel computing systems. Issues related to clusters and computational "grid" such as interprocessor communications, message-passing and mixed mode paradigms and programming techniques, high performance interconnects, efficient host-network interface for fast messaging, lightweight user-level messaging layers and protocols, (network interface-assisted based) collective communications, communication latency tolerance techniques, power-aware high-performance computing, high performance file systems and I/O, benchmarking and performance evaluation, scheduling and load balancing, system-level middleware and computational grid applications and services will be discussed, Research papers from literature, a term paper and hands-on programming and experiments on a network of workstations will supplement the course. Three term-hours; lectures. A. Afsahi
This course covers topics related to network-based parallel computing systems. Issues related to clusters and computational "grid" such as interprocessor communications, message-passing and mixed mode paradigms and programming techniques, high performance interconnects, efficient host-network interface for fast messaging, lightweight user-level messaging layers and protocols, (network interface-assisted based) collective communications, communication latency tolerance techniques, power-aware high-performance computing, high performance file systems and I/O, benchmarking and performance evaluation, scheduling and load balancing, system-level middleware and computational grid applications and services will be discussed, Research papers from literature, a term paper and hands-on programming and experiments on a network of workstations will supplement the course. Three term-hours; lectures. A. Afsahi
