1. Introduction to SRTF (Preemptive SJF):

Definition:

Shortest Remaining Time First (SRTF) is a preemptive CPU scheduling algorithm where the process with the shortest remaining burst time is selected for execution. If a new process arrives with a shorter burst time than the currently executing process, it preempts the CPU.

2. Burst Time and Arrival:

Burst Time:

Each process has an associated burst time, representing the total time it requires on the CPU.

Arrival of Processes:

Processes are added to the ready queue upon arrival.

3. Execution and Preemption:

Execution:

The process with the shortest remaining burst time is selected for execution.

If a new process arrives with a shorter burst time than the currently executing process, preemption occurs.

4. Waiting Time and Turnaround Time:

Waiting Time:

Waiting time is the cumulative time a process spends in the ready queue.

Turnaround Time:

Turnaround time is the overall time taken by a process from submission to completion.

5. Pros and Cons:

Pros:

Optimizes turnaround time by prioritizing processes with the shortest remaining burst time.

Efficient for processes with varying burst times.

Cons:

May result in increased context-switching overhead due to frequent preemption.

6. Starvation:

SRTF is designed to minimize starvation, as shorter processes are given priority, but longer processes may still experience delays.

7. Example Scenario:

Consider processes A, B, and C with burst times 6, 4, and 8, respectively.

A → B → A → C → B → A → C → ... (and so on)

Processes are selected based on their remaining burst time, allowing for dynamic adjustments as new processes arrive.

8. Conclusion:

Shortest Remaining Time First (SRTF) is a preemptive scheduling algorithm that dynamically adapts to the burst times of arriving processes. While it optimizes turnaround time, careful consideration is needed to mitigate the potential increase in context-switching overhead.

Notes From The Slides:

Shortest-Remaining-Time-First (SRTF)

In SJF: take the process with shortest CPU burst time (needed CPU time for the whole process)
However, if we allow preemption to make it more dynamic, then new coming processes may always have shorter CPU burst time and therefore will always get executed BEFORE the long ones. The long ones can get stuck! That’s starvation! to deal with it we can use a technique called Aging, where we keep track of how long (age) each process has been in the system. Then, if a process is in the system for too long (aged enough) will get executed even when it has longer CPU burst time than the others do.
Shortest-Remaining-Time-First (SRTF) = Preemptive SJF

take the process with shortest remaining (still needed) CPU burst time (to finish the process)

Example of SRTF (Preemptive SJF)

Now we add the concepts of varying arrival times and preemption to the analysis

Process	Arival Time	Burst Time
P1	0	8
P2	1	4
P3	2	9
P4	3	5

Example of SRTF (cont.)

Process	Arival Time	Burst Time
P1	0	8
P2	1	4
P3	2	9
P4	3	5

Preemptive SJF Gantt Chart

Average waiting time = [(10-1)+(1-1)+(17-2)+5-3)]/4 = 26/4 = 6.5
=p1[(0-0)+(10-1)] +p2[1-1] +p3[17-2]+p4[5-3] = 26