Thrashing occurs mainly due to which of the following conditions ?

Concept: Thrashing is the operating-system condition in which the total memory demand of the currently running processes exceeds the number of physical frames actually available to them, so each process is given far fewer resident frames than its working set (its actively-referenced set of pages) needs. Under the working-set / locality-of-reference model, once a process's resident-frame count falls below the size of its working set, it must continually replace pages that it will reference again almost immediately.

Application: As the degree of multiprogramming keeps increasing, the operating system tries to accommodate more processes by shrinking each one's frame allocation. Once frames drop below each process's working-set size, every process starts faulting almost continuously: an instruction needs a page that is not resident, the CPU traps into the page-fault handler, the process blocks on a disk I/O request to fetch that page, and while it waits, the CPU has few other runnable, resident-ready processes to switch to, so it sits mostly idle. Because the disk I/O needed to service a fault is far slower than instruction execution, this state shows up directly as a high page-fault rate, since nearly every memory reference misses, and as low CPU utilization, since the CPU spends most cycles waiting on the resulting swap-in operations rather than executing user code. This is exactly why operating-systems theory plots CPU utilization rising and then sharply collapsing as the degree of multiprogramming crosses the thrashing threshold, even as the page-fault rate keeps climbing.

Cross-check:

• High CPU utilization and low page fault rate describes the healthy, non-thrashing regime, the opposite pairing of both indicators, so it cannot be the cause of thrashing.

• High main memory fragmentation is a free-space layout problem in memory allocation; it does not by itself force repeated page faults or drive CPU utilization down, so it is a distinct operating-system concern from thrashing.

• A faulty disk scheduler causing excessive swapping is a hardware or driver malfunction, not the emergent, workload-driven condition of over-committed frames relative to working sets that operating-system theory identifies as the mechanism of thrashing.

• Only low CPU utilization and high page fault rate matches both empirical signatures of thrashing described above, confirming it as the condition being asked about.