Why would OS use paged segmentation or combined paging and segmentation?

• A. To avoid paging entirely
• B. To force fixed-size segments in memory
• C. To provide logical divisions (code, data) with variable sizes while benefiting from fixed-size paging to manage memory and reduce fragmentation; combined approach reduces external fragmentation and simplifies hardware
• D. To simplify hardware by removing MMU

Answer: (C)

The idea being tested is how an OS balances flexible program structure with efficient memory management. Segmentation lets you divide a process into logical pieces like code, data, and stack, each with its own variable size that matches the program’s needs. Paging, on the other hand, forces memory to be managed in fixed-size units, which makes allocation simpler, eliminates external fragmentation, and provides straightforward protection through page-level mechanisms.

Using both together gives the best of both worlds. You keep the natural, variable-sized segments to reflect the program’s organization, while backing those segments with fixed-size pages so memory can be allocated in any free frames. This means a segment doesn’t have to be stored contiguously in physical memory; its pages can be scattered, which greatly reduces external fragmentation. At the same time, you avoid wasting large chunks of memory within a segment by sticking to page-sized units, mitigating internal fragmentation as long as the page size is chosen wisely. The hardware can manage memory more predictably with a paging system (and its MMU/TLB), while the segmentation structure preserves meaningful boundaries for protection and sharing.

In short, combining paging with segmentation supports logical, flexible division of a program with the practical advantages of fixed-size memory management, reducing fragmentation and simplifying hardware design.