The naive parallel prefix algorithm is the original version of this algorithm; this algorithm is "naive" because it makes an assumption that given n input elements, 
, with the further assumption that n is dyadic (that is, 
for some positive integer, k), and we can run the algorithm in parallel over n processors (or n threads). Obviously, this will impose strong limits on the cardinality n of sets that we may process. However, given these conditions are satisfied, we have a nice result in that its computational ...
The naive parallel prefix algorithm
Get Hands-On GPU Programming with Python and CUDA now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
