Kernel Launch Overhead

Problem

Compute the total kernel launch overhead in milliseconds: num_kernels * overhead_us / 1000.

The signature is kernel_launch_overhead(num_kernels, overhead_us).

Hint

Launch overhead matters most for small kernels and eager execution.

case 1

in: [100,5]

expected: 0.5

case 2

in: [1,3]

expected: 0.003

case 3

in: [1000,2.5]

expected: 2.5

case 4

in: [0,5]

expected: 0

Starter codePython

def kernel_launch_overhead(num_kernels, overhead_us):
    # Your code here
    pass

4 test cases · run your code against them in the editor