TASK $1$:

Set$-$up and verify the CUDA computational environment in your Google Colab.

TASK $2$:

Write, compile, and execute a simple $$helloGPU$.$cu$$ CUDA program $/$

application.

TASK $3$:

Each thread that executes the kernel is given a unique thread ID that is accessible

within the kernel through the built$-$in threadIdx.x variable. Modify the kernel

function in $$helloGPU$.$cu$$ script with the thread index to let the output be:

$ $./$helloGPU

Hello VIT$\_$ from CPU!

Hello VIT$\_$ from GPU thread $5$