Demand paging vs. preparing

When a process transitions to the ready state, what pages should be placed in the main memory?

Two possible strategies: demand paging and prepaging.

Demand paging – place none and wait for the page faults.

• inefficient.

Prepaging – place those most likely to be referenced.

• the first time, the two pages mentioned before (code and stack).

• next times, those that were in the main memory when the process was suspended.

• more efficient.