Distributed Database: Types, Architectures, Advantages & Disadvantages

A Distributed Database (DDB) encompasses multiple interconnected databases that are logically related and spread across a computer network.

Table of Contents

A Distributed Database Management System (DDBMS) serves as the software responsible for managing the DDB. It offers an access mechanism that renders the distribution transparent to users. A distributed database system permits the physical storage of data across various sites. Each site or node is governed by an independent DBMS capable of functioning autonomously from other sites.

Unlike a centralized setup where storage devices are all linked to a single processing unit, in a distributed database, the storage can be distributed across multiple computers. These computers might be co-located or scattered over a network of interconnected machines.

System administrators can disperse data collections, such as databases, across various physical locations. A distributed database can be hosted on network servers over the Internet, corporate intranets, or other company networks.

Get A Free USA Scholarship, Visa & Accommodation

To ensure the distributed database remains up-to-date, two processes are utilized:

Replication: This involves specialized software that identifies changes in the distributed database. Once changes are recognized, replication processes synchronize all databases to match. This process can be intricate and time-consuming, depending on the number and size of the distributed databases.
Duplication: This process is simpler; it designates one database as the master and then duplicates it. Duplication typically occurs at specified intervals. This maintains uniform data across distributed locations. Users generally modify only the master database to prevent overwriting local data.

A Distributed Database Management System is tailored for heterogeneous database platforms, focusing on managing diverse database management systems. Desirable properties include:

Distributed Data Independence: Users should query data without specifying where referenced relations, copies, or fragments of these relations are located.

Distributed Transaction Atomicity: Users should compose transactions that access and update data across multiple sites, similar to working with local data.

Types of Distributed Databases

There are two primary types of distributed database systems:

Homogeneous Distributed Database: Conditions for a homogeneous database to be satisfied include using the same operating system at each location, along with compatible or identical data structures and database applications.
Heterogeneous Distributed Database: Conditions for a heterogeneous database include the possibility of different schemas, software, hardware, and data structures across nodes.

Architectures of Distributed Database Systems

The three major architectures for distributed DBMS are:

Client-Server: In this model, clients handle data presentation or processing, while servers manage storage, security, and significant data processing. Clients are responsible for user interfaces, while servers execute transactions and manage data. Client-server systems consist of client processes initiating requests to server processes, allowing queries to be sent to any server.
1. Advantages: Clear functional separation, ease of implementation, graphical user interfaces, distribution of roles, and security control.
2. Disadvantages*: Inability to span queries across servers, potential client-server distinction complexities, overlapping issues, and network traffic problems.
3. Collaborating Server: This setup comprises database servers capable of performing transactions on local data. These servers work together to execute transactions spanning multiple servers, addressing the limitations of client-server architecture.

Middleware: Web transactions occur on servers. Web servers handle communication with browsers, while database servers store the necessary data.

Advantages of Distributed Databases

Data is distributed across nodes.
Interconnected processors through a network.
Complete database functionality.
Reliable transactions due to replication.
Hardware, OS, network, fragmentation, and location independence.
Continuous operation even if nodes go offline.
Improved performance with distributed query processing.
Scalability and local autonomy.
Protection against data loss through distributed copies.
Modularity without affecting other systems.

Disadvantages of Distributed Databases

Complex data integrity maintenance.
Complexity in managing distributed data.
Increased infrastructure costs.
Lack of standardized framework.
Need for additional software.
Complexity in database design.
OS support for distributed environments.

Storing Data in DDBS

Data storage in distributed databases involves two concepts:

– Fragmentation: Splitting a relation into smaller fragments and storing them possibly at different sites. Horizontal fragmentation involves rows, while vertical fragmentation involves columns.

Study in US, Canada, UK, Germany For Free

– Replication: Storing multiple copies of a relation or fragment. Entire relations or fragments can be replicated at various sites.

Parallel DBMS vs. Distributed DBMS

Parallel Distributed System: Improves performance through parallel operations like data loading, indexing, and querying.
Distributed Database System: Physically stores data across sites, managed by DBMSs that can run independently. Distribution depends on factors like ownership and availability.