Article Categories
- All Categories
-
Data Structure
-
Networking
-
RDBMS
-
Operating System
-
Java
-
MS Excel
-
iOS
-
HTML
-
CSS
-
Android
-
Python
-
C Programming
-
C++
-
C#
-
MongoDB
-
MySQL
-
Javascript
-
PHP
-
Economics & Finance
Mass Storage Management
Mass Storage Management deals with the organization and optimization of secondary storage devices, primarily disks, in modern operating systems. Disks provide the bulk of secondary storage and require efficient management algorithms to handle data access requests.
Disk Structure
Modern disks contain concentric tracks divided into multiple sectors. The disk structure can be visualized as follows −
Tracks − Concentric circles on the disk surface
Sectors − Smallest storage units (also called blocks)
Cylinder − Same tracks across all disk surfaces
Read/Write Head − One head per disk surface for data access
Disk Scheduling Algorithms
Disk scheduling algorithms optimize the movement of the read/write head to minimize seek time and improve overall disk performance. Consider the following disk requests arriving in order −
Request Queue: 10, 95, 23, 78, 80 Initial Head Position: 50
First Come First Serve (FCFS)
FCFS services requests in the order they arrive. While fair, it does not optimize head movement.
Total Head Movement = |50-10| + |10-95| + |95-23| + |23-78| + |78-80| = 40 + 85 + 72 + 55 + 2 = 254 tracks
Shortest Seek Time First (SSTF)
SSTF services the request closest to the current head position. This reduces seek time but may cause starvation of requests far from the head.
Service Order: 50 ? 23 ? 10 ? 78 ? 80 ? 95
Total Head Movement = |50-23| + |23-10| + |10-78| + |78-80| + |80-95| = 27 + 13 + 68 + 2 + 15 = 125 tracks
SCAN Scheduling (Elevator Algorithm)
The head moves in one direction, servicing all requests until reaching the disk end, then reverses direction. This ensures no request waits indefinitely.
Service Order: 50 ? 23 ? 10 ? 0 (disk end) ? 78 ? 80 ? 95
Total Head Movement = |50-23| + |23-10| + |10-0| + |0-78| + |78-80| + |80-95| = 27 + 13 + 10 + 78 + 2 + 15 = 145 tracks
LOOK Scheduling
Similar to SCAN but reverses direction at the last request in each direction rather than going to the disk end.
Service Order: 50 ? 23 ? 10 ? 78 ? 80 ? 95
Total Head Movement = |50-23| + |23-10| + |10-78| + |78-80| + |80-95| = 27 + 13 + 68 + 2 + 15 = 125 tracks
Comparison
| Algorithm | Total Head Movement | Advantages | Disadvantages |
|---|---|---|---|
| FCFS | 254 tracks | Simple, fair to all requests | High seek time, poor performance |
| SSTF | 125 tracks | Lower seek time than FCFS | May cause starvation |
| SCAN | 145 tracks | No starvation, predictable | Longer wait for requests just missed |
| LOOK | 125 tracks | Better than SCAN, no unnecessary movement | More complex implementation |
Conclusion
Mass storage management involves organizing disk structures and implementing efficient scheduling algorithms. While FCFS is simple and fair, algorithms like SSTF and LOOK significantly reduce head movement and improve disk performance. The choice depends on balancing performance optimization with fairness to prevent request starvation.
