Distributed DBMS Distributed Databases
The concept of DDBMS is introduced in this chapter. There are a variety of databases in a distributed database that can be distributed geographically all over the world. The distributed database is handled in a way by a distributed DBMS such that it appears to users as one single database. We continue to research the factors that relate to distributed databases, their benefits and drawbacks in the later part of the chapter.
A distributed database is a set of multiple interconnected databases that interact through a computer network, physically spread across different locations.
- Databases are interrelated logically with each other in the collection. They also represent a single logical database.
- Across several sites, data is physically processed. Data on each site can be handled independently of the other sites via a DBMS.
- Via a network, the processors on the sites are connected. They have no configuration of any multiprocessor.
- A distributed database is not a file system which is loosely linked.
- A distributed database incorporates the transaction processing, but is not synonymous with a processing system for transactions.
Distributed Database Management System
A distributed database management system (DDBMS) is a centralised system of software that manages a distributed database as if it were all stored in a single place.
- It is used for the creation, retrieval, updating and deletion of distributed databases.
- It synchronises the database on a regular basis and provides users with access mechanisms by which the distribution becomes transparent.
- It ensures that the information modified on any site is updated universally.
- It is used in application areas where large data volumes are processed and accessed simultaneously by multiple users.
- It is designed for database platforms that are heterogeneous.
- The confidentiality and data integrity of the databases are maintained.
Factors Encouraging DDBMS
The dependent factors encourage the move to DDBMS-
- Distributed Nature of Organizational Units-Most companies are subdivided into various units physically distributed across the globe in the current times. Each unit includes a collection of local data of its own. The organization's overall database is thus distributed.
- Need for data sharing-Different organizational units also need to communicate and exchange their data and resources with each other. This includes common databases that should be used in a synchronized manner or replicated databases.
- Support for Both OLTP and OLAP − − Online Transaction Processing (OLTP) and Online Analytical Processing (OLAP) operate upon diversified systems which may have common data. Through offering synchronized data, distributed database systems facilitate all these processing activities.
- Database Recovery-Data replication across various sites is one of the standard techniques used in DDBMS. Data replication automatically helps to recover data if there is damage to the database on any site. While the damaged site is being reconstructed, users can access data from other sites. Thus, database failure may become almost inconspicuous to users.
- Multiple Application Software Support-Most organizations use a range of application software, each with specific support for their databases. For the use of the same data between different platforms, DDBMS provides a uniform functionality.
Advantages of Distributed Databases
The benefits of distributed databases over computer systems are described below.
Modular Development Of centralized database systems, whether the system is to be extended to new locations or new units, the action involves significant effort and disruption to the existing functioning. In distributed databases, however, the job simply involves adding to the new site new computers and local data and eventually connecting them to the distributed system, without disrupting current functions.
More Reliable −The complete system of centralized databases is stopped in the case of database failures. In distributed systems, however, the functioning of the system will continue to be at a reduced performance when a part fails. That makes DDBMS more reliable.
Better Response- If data is efficiently distributed, then user requests can be fulfilled from local data itself, providing a faster response. In centralized systems, on the other hand, all queries have to pass through the central processing computer, which increases the response time.
Lower Communication Cost − In distributed database systems, if data is located locally where it is mostly used, then the communication costs for data manipulation can be minimised. In centralized systems, this is not feasible.
Adversities of Distributed Databases
Some of the problems associated with distributed databases are below.
- Need for complex and expensive software − DDBMS requires complex and often expensive software to provide data transparency and co-ordination across the many sites.
- Overhead processing-Even simple operations may require a large number of communications and extra calculations to provide information uniformity across the sites.
- Data integrity- The need to update data on multiple sites raises data integrity problems.
- Overheads for improper distribution of data- Query responsiveness is largely based on proper distribution of data. Improper distribution of data often leads to a very slow response to requests from users.