Lecture 33 Optimizing Reduction Kernels Contd

Summary

Sorting bitinic sequence, All Prefix Sum , Inclusive and exclusive scan. Steel inclusive scan, Prefix Sum Implementation, Blelloch Scan Algorithm and Implementation. Comparator, Sorting subproblem, Bitonic Sort Parallel Implementation. Sorting, Sorting Networks, Bitonic Sort Serial Implementation, Recursion. Now the other one this is the most well-known Sebastien Bubeck, Microsoft Research simons.berkeley.edu/talks/sebastien-bubeck-09-21-2016 Transpose: Resolving Shared Memory Bank Conflicts, Memory Padding. Transpose Using Shared Memory, shared memory load transactions; store transactions. Transpose Operation: Naive Row and Naive Col Implementations.