On Using an Hybrid MPI-Thread Programming for the Implementation of a Parallel Sparse Direct Solver on a Network of SMP Nodes.
Pascal HénonPierre RametJean RomanPublished in: PPAM (2005)
Keyphrases
- shared memory
- parallel implementation
- parallel computers
- message passing interface
- parallel algorithm
- address space
- shared memory multiprocessor
- parallel computing
- network structure
- massively parallel
- interconnection networks
- message passing
- parallel programming
- local area network
- distributed memory
- graphics processing units
- parallel architectures
- data transfer
- sparsely connected
- load balance
- parallel computation
- parallel architecture
- parallel processing
- processing elements
- network nodes
- programming language
- parallelization strategy
- peer to peer
- fully connected
- network traffic
- network topologies
- high performance computing
- path length
- high level programming language
- hidden nodes
- bit parallel
- mobile networks
- unstructured peer to peer
- shortest path
- load balancing
- complex networks
- parallel genetic algorithm
- neighboring nodes
- propagation model
- parallel machines
- mobile nodes
- programming course
- centrality measures
- distributed processing