What are the differences of execution configuration syntax?

I am a total newbie in CUDA programming. And I want to know for 1 dimension of <<>>, this is means I only got one concurrent unit, And if I use <<<2, x>>>, this means I got two concurrent units?

This is covered in the first few units of this training series as well as the programming guide.

The <<<...>>> syntax as two mandatory arguments, the first is the number of blocks per grid, the second is the number of threads per block:

<<<blocks_per_grid, threads_per_block>>>

A block is a group of threads. Threads are the individual workers that run “concurrently”. Threads in a block (may) run concurrently. And blocks in the grid (may) run concurrently.

The grid is the set of all workers that will participate in a kernel launch.