Skip to main content

Scheduling prediction practice

  1. Trace A(0,8), B(0,2) with FIFO and RR quantum 2. Compute response, turnaround and waiting for each, not just averages.
  2. Trace A(0,4), B(2,1) under arrivals-before-requeue, then under the alternative rule. Name which part of the result depends on policy versus event semantics.
  3. Use A(0,1), B(5,1) with switch cost 3. Predict idle and dispatch intervals; do not charge the idle gap as switching.
  4. Trace A(0,2), B(0,1), C(3,1) with quantum 2 and cost 2. State which job is already selected when C arrives.
  5. Compare quantum 1,2,4 and cost 0,1 on A(0,8), B(0,2), C(1,1). Keep six reports and identify a tradeoff using individual job metrics.
  6. Explain why SJF's mean-turnaround optimum assumes a stated workload model (including simultaneous availability and known burst sizes). Construct a late-arrival case where nonpreemptive SJF cannot immediately run the arriving short job.

Keep predictions before execution, failed expectations and repairs. After 48 hours, change arrival times and repeat the weakest case without a solution. Do not treat agreement between two copied implementations as independent confirmation.