Changes

K2

213 bytes added, 20:20, 16 April 2018

→‎Assignment3

The bitonic sorting algorithm is devised for the input length n being a power of 2.

To check the noticeable time gap, we put 2^1516, 2^20, 2^2524.

void bitonicSort(int* array, int N){

Let's see how this algorithm works to sort and explanation of several terms.

[[File:BitonicSort1.png]]

Below is the comparison between bitonic sort using GPU, quick sort using CPU, bitonic sort using CPU.

[[File:BitonicGraph.jpg|500px|thumb|left|sorting algorithms comparison result]]

==Assignment3==

While working on the project, we discovered that cudaMemcpy(HtoD) and (DtoH) takes long time.

~~Use~~ so, we decided to use pinned memory instead of pageable memory ~~cudaHostAlloc();~~to improve its performance.

"It cudaHostAlloc() will allocate pinned memory which is always stored in RAM and can be accessed by GPU's DMA(direct memory access) directly"

[[File:Resultone.png|500px|thumb|left|pinned memory vs pageable memory]]

~~[[File:Resultone.png|500px|thumb|left|~~ using pinned memory vs is 2 times faster than pageable memory]]

44

edits