I can think of reasons why a pure cache would impose costs that are too high -- namely, the device would then be writing during times of very read load.
But: I agree with your bigger point, that it's an interesting programming problem.
We're all familiar with CPU caches, so it's interesting mental exercise to think of this as the same system, but optimized for a different part of design space.
But: I agree with your bigger point, that it's an interesting programming problem.
We're all familiar with CPU caches, so it's interesting mental exercise to think of this as the same system, but optimized for a different part of design space.