Process generator PID# State(ready/Running) CPU Burst (Accu.) I/O Burst (Accu.) Arrival Time Estimated CPU Burst
Analytic Metrics Throughput Turnaround Response
PCB All the info generated (list of P’s)
Uniprocessor Algorithms R.R. Queues that schedule their own version of the PCB (Time Quanta: 15, 30, 50) FCFS Based off process info and how they schedule SPN Sort of create virtual Gantt Charts that add up wait times as they are being sorted As they are scheduling add up to analytics & print to a text doc
Multi-core Distribution Load sharing Queue of processes, next available core takes the next process in the queue. Divide work across 8 cores
Multilevel Feedback Queue Three round robin queues followed by an FCFS queue. For the Round robins, the time quanta increases as we get deeper into the multilevel feedback queue. After 2 runs of a queue the processes that still remain will be moved to the lower queues until they reach the bottom FCFS. That guarantees that every queue is run through at least once.
Work Breakdown: Group: Process Generator Tristan: RR, MLFQ Eben: SPN, Analytics Aggregation Andrew: FCFS, Multi-Core Distribution
Deliverable Dates Process Generator: March 23 Uniprocessor Algorithms: March 30 + Weekend Multi-core, MLFQ, Analytics: April 6