When Threading Building Blocks Attack!

 This week (25-29 JUL 2011) I'm lecturing at the UPCRC UIUC Summer School on Multicore Programming. I like teaching parallel programming and I live near the UIUC campus, so it's convenient for me. Overall I have a good time and often learn some things from the students. As an example, I found the following oddity about using lambdas within Intel® Threading Building Blocks (TBB).

Generic Parallel Algorithms for Intel ® TBB - "They're Already in There" Part 1

A high-level overview of some generic parallel algorithms included in Intel® TBB to let you know what's possible: parallel_reduce, parallel_do, parallel_for_each

Observations from Parallel Sorting Part I: the subtleties of tbb::parallel_reduce

In this series of postings, I discuss two common sorting algorithms, mergesort and quicksort, and highlight some of the interesting issues that arise when creating parallel implementations using TBB.  In all cases we’ll assume that, like STL’s std::sort, the input to the sort is a pair of RandomAccessIterators begin and end that denote the beginning and ending of the collection to be sorted, and a comparator object compare that provides an operator() for ordering two given elements.

Подписаться на parallel_reduce