CPU Scheduling Algorithm Simulator

FCFS (First Come First Served): sort by arrival time, execute in order, no preemption. Simple but suffers convoy effect (short processes wait for long process ahead). SJF (Shortest Job First): non-preemptive — select shortest burst process from arrived queue. Preemptive SJF (SRTF — Shortest Remaining Time First): on each arrival and CPU cycle, switch to process with smallest remaining burst time. SJF minimizes average waiting time but requires knowledge of burst time (impractical in real OS — estimated using exponential averaging of past bursts).

Round Robin: maintain circular queue, give each process a time quantum (Q). After Q ms, preempt and add to end of queue. Configure Q: too small → excessive context switching overhead, too large → degrades to FCFS. Priority: select process with highest priority from ready queue. Preemptive variant: if higher-priority process arrives, preempt current. Problem: starvation (low-priority processes never execute). Solution: aging — periodically increase priority of waiting processes.

MLFQ (used in macOS, Unix) has multiple queues with different priorities and quantum sizes. Queue 1 (highest priority): quantum=8ms. Queue 2: quantum=16ms. Queue 3 (lowest, FCFS): no preemption. Rules: new processes enter queue 1. If process uses full quantum, demote to lower queue (CPU-bound detected). If process blocks for I/O before quantum expires, stay in same queue (I/O-bound processes kept high priority). Boost: periodically move all processes to queue 1 to prevent starvation.