Par Lab Bootcamp 2012
by Henrik Nordberg
1. Software Architecture
1.1. 13 Motifs of computation
1.2. 10 Structural Patterns
1.3. Image Recognition with SVM
1.4. Speech recognition
2. OpenMP
3. MPI
4. Finding Parallelism
5. GPU
5.1. thread block
5.2. __global__ et al.
6. Programming Languages
6.1. UPC
6.1.1. synchronization
6.1.2. collective functions
6.1.3. work distribution
6.1.4. shared memory
6.1.5. distributed arrays
6.1.6. Distr. Arrays Directory style
6.2. Phalanx
7. autotuning
7.1. minimize communication
7.2. computational intensity
7.3. Sparse Matrix Multiply
7.4. pOSKI
7.5. SEJITS
8. Performance
8.1. tools
8.1.1. PAPI
8.1.2. valgrind
8.1.3. PMPI
8.1.4. Vendor Tools
8.1.4.1. Tau
8.1.4.2. CrayPat
8.1.4.3. ipm