If all DCD is doing is ripping bytes from disk without any processing then I wouldn't expect much dependence on parallelism at all. It should be entirely determined by the I/O bandwidth to the CPUs.
That's exactly what the data show.
Maybe it is not clear that we would like to avoid any I/O cost and we would like to get our performance (reduced time to solution) by working on the compute problem in parallel. Make clearer in intro.
That's exactly what the data show.
Maybe it is not clear that we would like to avoid any I/O cost and we would like to get our performance (reduced time to solution) by working on the compute problem in parallel. Make clearer in intro.