In this project five different sorting algorithms are analysed with different approaches using CUDA.

​

1. Selection Sort

2. Bitonic Sort

3. Quick Sort

4. Merge Sort

5. Radix Sort

​

Different approaches are discussed with optimizing getting better results with the algorithm. In improved sorting methods merge sort is implemented which is designed for many-core GPU. Performance is analysed on GPU and CPU and comparison is performed. Use of Thrust concept provides the improved performance which is implemented with radix sort and analysis is done.

​