Bit Map for Dynamic Partitioning

The Main concern for dynamic partitioning is keeping track of all the free and allocated partitions. However, the Operating system uses following data structures for this task.

1. Bit Map
2. Linked List

Bit Map is the least famous data structure to store the details. In this scheme, the main memory is divided into the collection of allocation units. One or more allocation units may be allocated to a process according to the need of that process. However, the size of the allocation unit is fixed that is defined by the Operating System and never changed. Although the partition size may vary but the allocation size is fixed.

The main task of the operating system is to keep track of whether the partition is free or filled. For this purpose, the operating system also manages another data structure that is called bitmap.

The process or the hole in Allocation units is represented by a flag bit of bitmap. In the image shown below, a flag bit is defined for every bit of allocation units. However, it is not the general case, it depends on the OS that, for how many bits of the allocation units, it wants to store the flag bit.

The flag bit is set to 1 if there is a contiguously present process at the adjacent bit in allocation unit otherwise it is set to 0.

A string of 0s in the bitmap shows that there is a hole in the relative Allocation unit while the string of 1s represents the process in the relative allocation unit.

Disadvantages of using Bitmap

1. The OS has to assign some memory for bitmap as well since it stores the details about allocation units. That much amount of memory cannot be used to load any process therefore that decreases the degree of multiprogramming as well as throughput.

In the above image,

The allocation unit is of 4 bits that is 0.5 bits. Here, 1 bit of the bitmap is representing 1 bit of allocation unit.

Therefore, in this bitmap configuration, 1/5 of total main memory is wasted.

2. To identify any hole in the memory, the OS need to search the string of 0s in the bitmap. This searching takes a huge amount of time which makes the system inefficient to some extent