GPU code using CUDA, C programing

Need the best result GPU EXPERIMENT: You will create an array A with 5000 random chars in host Transfer A to device In the device, create 10 threads Each thread will (insertion) sort its own chuck, such as Thread 0 sorts from A[0] to A[499], Thread 1 sorts from A[500] to A[999] Transfer back A to device Measure how long does it take the whole process, including transferring A to device and from device back to host Print each chuck