🖥️ CPU Scheduling — FCFS, SJF, Round Robin

For each process, turnaround time equals completion time minus arrival time, and waiting time equals turnaround time minus burst time. The averages shown update live as the cursor reveals completed processes, so you can watch the figures settle as the Gantt chart fills in.

Round Robin preempts the running process whenever its quantum expires and sends it to the back of a circular queue. With a small quantum the CPU changes process frequently, which improves response time and fairness for interactive workloads but raises the context-switch count, each one carrying real overhead in a true system.

The convoy effect occurs under FCFS when a long CPU-bound process arrives first and several short processes queue behind it, inflating their waiting times. Randomise until a long job arrives early, run FCFS, then switch to SJF on the same set: the average waiting time usually drops sharply.

Yes. SRTF and Round Robin are genuinely preemptive in the model, and when no process has arrived yet the timeline records an idle unit shown as a faint block. Those idle gaps are still counted in totalTime, so they correctly delay the completion and turnaround of later processes.

Each process is assigned a priority number, and the scheduler runs the available process with the lowest number, treating that as the highest priority. It is non-preemptive here, so a chosen process finishes before the next decision. Without aging this scheme can let a low-priority process starve if higher-priority work keeps arriving.

The algorithms match the textbook definitions taught in operating-systems courses, with integer time units and known burst lengths. Real schedulers add complications such as unknown future bursts, I/O blocking, multiple cores, priority aging and per-switch overhead, so this is an accurate teaching model rather than a faithful reproduction of a production kernel.