Might that not imply that the implementation is relatively little tied to a particular OS? Taking your example, as long as there's a way to get the block size for a disk, there's not much more OS-specific code to deal with -- in particular, adapting your code to the file system doesn't need to be done since the DB will essentially create its own anyway. Similarly, only a few threading primitives need to be used -- the DB isn't going to tie itself heavily to the vagaries of the async IO implementation, which does tend to differ heavily and pervasively between platforms.