lysa karl
by on September 8, 2023

A distributed database is basically a database that is not limited to one system, and is distributed on different sites, that is, on multiple computers or a computer network. A distributed database system resides at various sites that do not share physical components. This may be necessary when a specific database needs to be accessed by different users around the world. It needs to be managed so that to the user it looks like a database.


1. Homogeneous database:
In a homogeneous database, all different sites store the database in the same way. The operating system, database management system and data structures used are the same for all sites. Therefore, they are easy to manage.

2. Heterogeneous distributed database
In a heterogeneous distributed database, different sites can use different architectures and software, which can cause problems in query processing and transactions. Additionally, a particular site may be completely unaware of other sites. Different computers may use different operating systems and different database applications. They may even use a different data model for the database. Therefore, communication between different sites requires translation.

Distributed data storage:
There are two ways to store data on different sites. all these are:

1. Copy –
In this approach, the entire relationship is stored redundantly on 2 or more sites. If the entire database is available at all sites, it is a fully redundant database. So, in replication, the system maintains a copy of the data.

This is advantageous as it increases data availability across different sites. Additionally, query requests can now be processed in parallel.
However, it also has certain disadvantages. Data needs to be constantly updated. Any changes made at one site need to be recorded at each site where the relationship is stored, otherwise inconsistencies may result. This is a lot of overhead. Additionally, concurrency control becomes more complex as concurrent access now needs to be checked on multiple sites.

2. Fragmentation——
In this approach, relationships are fragmented (i.e., they are divided into smaller parts), and each fragment is stored in a different site where they are needed. You must ensure that the fragment can be used to reconstruct the original relationship (i.e., no data is lost).
Fragmentation is advantageous because it does not create copies of the data and consistency is not an issue.

Posted in: Technology
Topics: science
Be the first person to like this.