“A database is a system that converts a large collection of data into an abstract tool, allowing users to search for and extract pertinent items of information in a manner that is convenient to the user.”(Brookshear, 2012, p383)
According to Özsu and Valduriez(2011, p3) “a distributed database as a collection of multiple, logically interrelated databases distributed over a computer network.” and is managed by a distributed database management system, DDBMS for short, figure 1.0. This data will be distributed across numerous physical locations as a result distributed databases can improve performance and efficiency by allowing for transactions to be processed on multiple machines, rather than one. To maintain all distributed databases and have them up to date, two processes are employed: duplication and replication.
Whilst, on the other hand, a localized or centralized database is one in which a database is stored, maintained, and located in a single location, figure 1.1. Historically, this was the first version of databases, that gave organizations the ability to organize store data in one central location whereby increasing the data’s integrity.
Figure 1.0 Illustrates a conceptual overview of a distributed database. (Oracle, n.d)
Figure 1.1 Illustrates a conceptual overview of a localized or centralized database.(Neffenger, n.d)
Advantages and Disadvantage of Distributed Databases:
- Increased reliability and availability
- Faster response time
- Data Protection – In case of disasters
- Lower communication costs (More Economical)
- Cost – more expensive to implement
- The operating system should support a distributed environment
- Complex Software Implementation
- Increased Processing overhead
- Data integrity
- Deadlock is difficult to handle.
- Handling failures is a difficult task.
Advantages and Disadvantage of Local/Centralized Databases:
- Data integrity is maximized
- Easier Implementation
- Decreased Risk – data is manipulated and stored in one place
- Single point of entry
- Integration is easy
- Upgrades, mirroring, and backups are easier.
- no two persons can access the same data
- Single Point of failure
- Slower Response Time — pending on the size and frequency of requests
It is preferential to use a distributed rather than a local database when multiple sites need data synchronized and data will be manipulated at each respective site. An example of which is a banking system. I believe choosing which to implement depends on the volume of data and how it will be used.
If all computers were localized in one building, A local/centralized database would be preferential.
If all computers were spread across your country — Jamaica, It is feasible for a distributed database configuration, as you can have one local site and have all the other computers or sites communicate with it. The National Commercial Bank, NCB in Jamaica is a prime example. They have many branches spread across the country with each attending to its customer, therefore maintaining its own local copy. All this information is then available at the Headquarters.
If all computers were spread around the world, Again a Distributed Database is feasible in this sense. We can look at the banking example again with Branches around the world. Each country has it’s localized copy and which communicates with an HQ. The data is duplicated and replicated as stated in the definition above.
This can be used in place of a distributed database, in a small implementation of a centralized database. Mirror offers fault tolerance in data availability. Thus, if a hard disk fails, the entire system isn’t brought to a halt. According to Bradford(n.d), “Mirroring is often described as a backup method. This is not the case. Although the terms are regularly used interchangeably, they are two different tactics designed to meet two different objectives.”
In conclusion, with the advent of database systems, a lot has become possible from the reduction in data duplication, data integrity, data retrieval and data mining for insight into the information collected. Implementing a centralized or distributed database is dependent on the organization’s requirements, data size, and the applications using the same.
Bradford, C. (n.d) Can Mirroring Replace Backups In Your Disaster Recovery Strategy?, Available at: http://www.storagecraft.com/blog/can-mirroring-replace-backups-disaster-recovery-strategy/ (Accessed: May 8, 2016).
Brookshear, J. G., Smith, D. and Brylow, D. Computer Science: An Overview, 11th Edition. Reading, MA: Pearson (Addison-Wesley), 2012
Kandasamy, S (2014) Advantages and Disadvantages of Distributed Databases, Available at: https://exploredatabase.blogspot.com/2014/08/advantages-and-disadvantages-of-distributed-databases.html (Accessed: May 8, 2016).
Neffenger, J. (n.d) Member Database Support, Available at: http://www.volano.com/docs/volano2/database.html (Accessed: May 8, 2016).
Oracle (n.d) Distributed Database Concepts, Available at: https://docs.oracle.com/cd/B10501_01/server.920/a96521/ds_concepts.htm (Accessed: May 8, 2016).
Özsu, M.T. and Valduriez, P., 2011. Principles of distributed database systems. Springer Science & Business Media.