allows multiple programs to reside in separate areas of core at the time.
is a method of memory allocation by which the program is subdivided into equal portions, or pages and core is subdivided into equal portions or blocks.
consists of those addresses that may be generated by a processor during execution of a computation.