L2 cache optimization in matrix multiplication

[alfin3]

My understanding is that the consecutive program ids are mapped to 2D program ids such that:

• the segments of the B matrix along the K dimension that are non-contiguous in the global GPU memory are more likely to be kept and reused in the L2 cache, and
• the segments of the A matrix along the K dimension that are contiguous in the global GPU memory are more likely to be repeatedly fetched into the L2 cache.

A more efficient use of the L2 cache line would result in fewer misses and higher TFLOPS.

It is also stated that the programs are “launch[ed] in an order that promotes data reuse”. The execution order of thread blocks is undefined, unless co-scheduling in thread block clusters is used on an NVIDIA GPU.

Is the statement regarding the more than 10% improvement in TFLOPS based on experiment(s) where the improvement was specifically attributable to L2 cache and the launch and execution order?

How are programs “launch[ed and executed] in an order that promotes data reuse” in terms of thread blocks?

Thank you.