Best way to get the min value from an array

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi Everyone,

The topic title says it all. Sorry if this seems like a stupid question- I’m still getting used to parallel programming. Also I’m somewhat confused as to whether the parallel prefix-sum example using CUDPP has something to do with this- I remember reading that the scan was useful as a building block for parallel sorting algorithms, but I can’t see how for the life of me. Again I apologize if this is something obvious or if it’s been explicitly dealt with before. If it has could you please tell me where I can read up on this? Thanks a lot!

Cheers,

Paul

2

i think that the best way is using reduction …

3

Sorry for being a bit dense mate, but do you mean anything specific when you use “reduction” or are you referring to the general meaning of reduction? Thanks

4

He’s talking about data parallel reduction. There’s an example here.