Segmentation in Operating System (OS) | Core Computer Science

What is Segmentation?

Segmentation is another memory management technique that gives the user's view of a process. The user's view is mapped into the Physical Memory.

In Segmentation, a process is divided into multiple segments. The size of each segment is not necessarily the same which is different from paging. In Paging, a process was divided into equal partitions called pages. The module contained in a segment decides the size of the segment.

Why Segmentation is needed?

We used Paging as one of the memory management techniques. In Paging, the process was divided into equal-sized pages irrespective of the fact that what is inside the pages. It also divides some relative parts of a process into different pages which should be loaded on the same page. It decreases the efficiency of the system and doesn't give the user's view of a process.

In Segmentation, similar modules are loaded in the same segments. It gives the user's view of a process and also increases the efficiency of the system as compared to Paging.

Types of Segmentation

There are two types of Segmentation. These are:

1. Simple Segmentation - In Simple Segmentation, each process is divided into multiple segments and all of them are loaded into the memory at run time. They can be at non-contagious locations.
2. Virtual Memory Segmentation - In Virtual Memory Segmentation, each process is divided into multiple segments and all of them do not reside at one point at a time.

Segment Table

• Segment Table stores the information about all the segments of a process.
• It helps in the mapping of the two-dimensional logical addresses to the physical addresses.
• It is stored in the main memory.
• There are two entries in the Segment Table.

1. Base Address - It is the starting physical address of the particular segment inside the main memory.
2. Limit - It denotes the size of a particular segment.

Translation of Logical Address into Physical Address

The CPU generates the logical address which consists of two parts:

1. Segment Number(s) - It is the number of bits required to represent a Segment. It is used as an index in the Segment Table.
2. Segment Offset(d) - It is the number of bits required to represent the size of a Segment.

Segment Number from the Segment Table gives the base address of a Segment in the main memory and Segment Offset gives the size of that Segment. These two together give the Physical Address of a Segment in the memory. If Offset(d)<Limit then the segment is fetched from the memory else there is an error. 

Advantages of Segmentation

• There is no Internal Fragmentation.
• It gives the user's view of a process.
• It helps in dividing a program into similar modules which are called segments.
• Segment Table consumes less space than Page Table.

Disadvantages of Segmentation 

• When Process is removed from the memory, it breaks memory into small pieces which leads to External Fragmentation.