algorithm - why do we use RAM model of computation for algorithmic analysis when it corresponds to harvard architecture? -
as per wikipedia , rasp model of computation corresponds von neumann architecture while ram model of computation corresponds harvard architecture , von neumann architecture closest computers since have ram memory in our computers stores data instructions in , while ram model has different program memory , different data memory , why use ram model algorithmic analysis ?
https://en.wikipedia.org/wiki/random-access_stored-program_machine
unless algorithm uses self-modifying code, distinction doesn't matter, reasons margaret points out.
also, far real hardware concerned, (almost?) modern cpus have cache have split instruction/data l1 caches, actual cpu core more harvard-like von neumann. instruction-fetch competes data access.
wikipedia's modified harvard architecture article goes more detail, , mentions other ways mostly-harvard architectures can still programmed programs.
Comments
Post a Comment