One CPU. Different decisions.

Arrival time is when a process becomes ready. Burst time is the CPU time it needs. Enter arrival 0 for every process if they all arrive together.

CT is the clock time when a process finishes. For example, a process arriving at 2 and finishing at 9 has CT = 9.

TAT = CT − arrival time. It includes all time from arrival until completion. In the example above, TAT = 9 − 2 = 7.

WT = TAT − burst time. If TAT is 7 and burst is 4, WT = 3. This is the total time spent ready but not running.

RT = first start time − arrival time. It measures the wait before the first CPU execution. A preempted process can have a low RT and a much larger total WT.

Total is the sum of a column. Average = total ÷ process count. For waiting times 0, 3 and 6, total WT = 9 and average WT = 3. Total CT is a sum of finish timestamps; it is not elapsed CPU time.

Arrival and burst affect all available algorithms. Priority numbers affect the two Priority variants. Quantum affects Round Robin. The comparison uses the same input rows for every algorithm.

The earliest arrived ready process runs to completion. Equal arrival times use the order of the input rows.

The shortest burst among processes that have already arrived runs to completion. Later arrivals cannot interrupt it.

SRTF is preemptive SJF. Every time unit, it chooses the smallest remaining burst. A shorter arriving process can interrupt the current one. Equal remaining times use arrival time, then input order.

A smaller number means higher priority. Priority 1 runs before priority 3 when both are ready. Equal-priority dispatch ties use arrival time, then input order.

Non-preemptive Priority lets the selected job finish. Preemptive Priority interrupts it when a strictly higher-priority job arrives. An equal-priority arrival does not interrupt the current job.

Each ready job gets at most q CPU time units. If unfinished, it goes to the back of the ready queue. Jobs arriving during a slice or exactly at its end are added before the current job is requeued.

At each dispatch, calculate (current waiting time + burst) ÷ burst. The highest ratio runs to completion. For wait 6 and burst 2, the ratio is 4. Recalculate ratios after every completion. HRRN is non-preemptive.

High (Q0): Round Robin with q. Low (Q1): FCFS. High always takes precedence and can interrupt Low. A process stays in its assigned queue.

Every new job enters Q0 (RR, q). If unfinished after its CPU allowance, it moves to Q1 (RR, 2q), then Q2 (FCFS). Higher queues interrupt lower ones. An interrupted job keeps its queue position and consumed allowance. This version has no aging or periodic boost.

Use +1 unit and −1 unit to move the clock, Play to animate, or the slider to inspect any time. Next event skips to a scheduling event. The state table shows CPU time used, remaining burst and waiting so far. At a boundary, arrivals and completions have already been applied, and Running identifies the process selected for the next unit. The main result table always shows final values.

It includes the selected algorithm, inputs, process results, totals, averages, CPU statistics, the comparison of all available algorithms and the complete execution timeline. Open it in Excel or another spreadsheet app. IDs beginning with spreadsheet formula characters are exported as text. Changed inputs are validated and recalculated before export.

Click Print / Save PDF. In the browser print dialog, choose your printer or Save as PDF. The report includes the selected algorithm, inputs, final results and comparison. A long timeline is limited to its first 100 segments in print; CSV includes every segment. Playback controls are excluded from the report.

It shows CPU execution order. Each block names a process and its time boundaries; Idle means no process was ready. Adjacent intervals for the same process are merged. Widths have a readability minimum, so use the labels for exact durations.

Long schedules need room for their time labels. Swipe or scroll inside the chart to see later blocks. A long schedule shows up to 400 segments at a time, with a notice. Playback moves the chart window to the current time; all calculations use the full schedule.

Utilization = total burst ÷ elapsed time × 100. Throughput = process count ÷ elapsed time. Elapsed time runs from clock 0 to the final completion, including initial idle time.

A direct change from one running process to a different process counts as one switch. The initial dispatch, a transition through Idle, and consecutive slices of the same process do not count. Switching takes zero time in this simulation.

Click its name in the comparison table, or change the algorithm dropdown. The process inputs stay in place, and its timeline, results, totals and averages appear. If you changed an input, it is recalculated before switching.

Green marks the lowest average in that column for your current workload, including ties. A different workload or quantum can change the result. Waiting time and response time can favor different algorithms.

This models one CPU and one CPU burst per process, with no I/O blocking. Use 1–30 processes. Arrival: 0–10,000; burst, quantum and priority: 1–10,000. Times are whole-number abstract units. It simulates scheduling; it does not change your computer’s scheduler.