4k views

Consider an operating system capable of loading and executing a single sequential user process at a time. The disk head scheduling algorithm used is First Come First Served (FCFS). If FCFS is replaced by Shortest Seek Time First (SSTF), claimed by the vendor to give $50 \%$ better benchmark results, what is the expected improvement in the I/O performance of user programs?

1. $50 \%$
2. $40\%$
3. $25\%$
4. $0\%$
edited | 4k views
0
+2
@arjun sir if they mention in question multi user process at a time then what is answer how it improves the Io performances of user pgm

Question says "single sequential user process". So, all the requests to disk scheduler will be in sequence and each one will be blocking the execution and hence there is no use of any disk scheduling algorithm. Any disk scheduling algorithm gives the same input sequence and hence the improvement will be $0\%$ for SSTF over FCFS.

Correct Answer: $D$
edited
+3
+4
@Arjun sir is "single sequential user processes" same as "single processor system"?
+55

@Akhil Then we can believe the vendor and chose answer as A. (Of course then this would not have been asked in exam)

@Srestha No, they are not the same. Because "single processor system" can also be multi threaded. That is even if only a single process is executing different part of the same process might be run in parallel. So, some of these threads can generate disk requests paralelly. But if the process is also sequential, then only one request will be live at any time. (the request should be blocking) and hence there is no need of a scheduling algorithm.

+4
Since sequentially one-by-one user processes will come to the disk scheduler so they will be scheduled in FCFS order even if the scheduler uses SSTF algo. Thus 0% improvement.

Is my understanding correct?
+16
If the disk scheduler had known all the disk requests priori then it could have scheduled the order of servicing the requests. That is, for FCFS it would service in the same order as the requests came. For SSTF the disk header services the request which is closest to the current one(in other words where the header movement will take the shortest time to reach).

In this current scenario, the disk scheduler doesn't know what is the next request(as they come one by one after each request gets served) so it can't manipulate the movement of the header and therefore has to move as per the order of incoming requests.

This was my understanding. Correct me if I am wrong.
+1
@tuhin seems right :)

so sequential is the catch here ?

what if it was just single user process?
+3

@arjunsir please check my procedure, correct or not ..??

one is for FCFS and another is for SSTF ..!

0

according to you all the algorithm will work same. Am i correct?

0

@flash12 Yes..don't you think so..?

0

@MiNiPanda sir

here I/O performance =io access time?

so vendor claims SSTF better for particular instance whereas user program may or may not agree with such request of pattern so performance can not be measured with the argument.

am i right?

0

@MiNiPanda

Even if the disk scheduler knows the requests in advance, if they are sequential like 2,3,5,6,7,10,14....

both FCFS and SSTF will choose same requests each time.

Isn't?

+1

@sushmita Yes if the header is positioned before cylinder 2.

@Abhisek Tiwari 4 Yes i think you are right :)

1
2
3
4
5
6