GitHub/quartz
1
0
Fork 2
mirror of https://github.com/jackyzha0/quartz.git synced 2025-12-27 14:54:05 -06:00
Code Issues Actions Packages Projects Releases Wiki Activity

vault backup: 2022-10-30 13:55:53

Browse Source
This commit is contained in:
Jet Hughes 2022-10-30 13:55:53 +13:00
parent c5ae995627
commit c8bc1f4697
1 changed files with 47 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

48
content/notes/19-cpu-scheduling.md
View File

@ -24,7 +24,53 @@ criteria
- throughput - number of processes completed per time unit - throughput - number of processes completed per time unit
- turnaround time - (for a single process) length of time from submitted to completed - turnaround time - (for a single process) length of time from submitted to completed
- waiting time - (for a single process) time spent waiting for the CPU - waiting time - (for a single process) time spent waiting for the CPU
- response time - (for a signle process) average timefrom the submission of a request to a process util the first response in produced - does include time wating for IO
- response time - (for a signle process) average time from the submission of a request to a process until the first response is produced
Preemption
- 4 situations for scheduling
1. a process switches from running to waiting
2. a process switches from running to ready (due to interrupt)
3. a process switches from wauting to ready (due to comletion of IO)
4. a process terminates
- non preemptive - scheduling take splace only under 1 nd 4, which has no choice
- premptive - scheduling can take place under 2 and 3 as well. that means a running process has to give way to the processes with higher priority
if you allow interruption it is preemptive
# Scheduling algorithms # Scheduling algorithms
## Gantt charts
The operation of scheduling algortihm is commonly represented in a Gantt chart
![example chart|400](https://i.imgur.com/haqjMOW.png)
![gantt evaluation|400](https://i.imgur.com/iqtT90Q.png)
## First come first served (FCFS)
![diagram|400](https://i.imgur.com/yAd4LRl.png)
simplest method- executes processes in the ready queue on a first come first served basis
- when a proces become ready it is put at the tail of the queue
- when the currently executig process terminates, or waits for IO, the process at the front of the queue is selected next
- processes can come at any time - what if two at the same time? choose process with smallest ID to serve first
non-preemptive
pro
- easy to understand and implement
con
- waiting time bad
- convoy effect: lots of small process can get stuck behind a big on
- bad response time
fair but not efficient
## Shortest job first (SJF)
choose the job in the queue with the shortest burst time.
- equal burst time are executed FCFS (by process id)
![SJF example|400](https://i.imgur.com/nzwBwjc.png)
pro
- prova