Comparison between centralized and distributed DBMS
1- Comparison between centralized and distributed DBMS:
There are many aspect that let us make a comparison between centralized and distributed DBMS:
- Database management system is any software that manages and controls the storage, the organization, security, retrieval and integral of data in a specific database, whereas DDBMS consist of a single database that is divided into many fragments. Each fragment is integrated on one or more computer and controlled by independent database (DBMS) (Connolly & Begg, 2004).
- In centralized DBMS the data is distributed across the network computers, and the data is stored on many sites and under the management responsibility of DDBMS. But in the DBMS data is stored and controlled in a central site.
- Both of DDBMS and centralized DBMS provide the access to database using the same interface, but for this function centralized DBMS faces less complication than DDBMS.
- For distributing data over network we can use replication or fragmentation. The objective of replication and fragmentation is to make a transparency of this allocation to make the details of implementation hidden on users. In centralized DBMS is not need to make transparencies.
- In DDBMS design we can find three issues which are not in centralized DBMS design. These issues are: How to split the database to fragments, and fragments to replicate, and in which locate we can find these fragments.
- Consequently, centralized DBMS is less sophisticated than DDBMS because it not supports the organizational structure of today's widely distributed enterprises, and DDBMS more reactive and reliable (Blurtit, 2010).
2- Advantages of DDBMSs
The DDBMS have many advantages over DBMS. We can gather them in the following points:
Reflects organizational structure: We can distribute the data base over any organization offices which has distributed locations
Improved shareability and local autonomy: Users can use the data of other sites, which mean the data can be near of the users who use it. Also by this way the data can be controlled by the user.
Improved availability: Unlike centralized DBMS, the failure at one site or link of communication makes only some parts of system inaccessible, which means the entire system is still working.
Improved reliability: The replication system make the data exist in many site. So this insures the possibility of accessing to this data if there is any failure happened.
Improved performance: We can improve the accessing speed to data base if we use remote centralized database. Also, there may not be same conflict for CPU and using the services, like DBMS.
Economics: For making the organization systems more cost-effective to obtain separate computers, DBMS allows us to create systems of smaller computer, its power equal the power of one large computer.
Modular growth: This one refers to the flexibility of DDBMS, where we can add a new site without any affects on the operation of other sites.
Complexity: The features of DDBMS like replication, availability flexibility make the implementation of DDBMS more complex than centralized DBMS.
Cost: The cost will increase due to the complexity in DDBMS than centralized DBMS. It requires the maintenance and additional hardware to create a network between the posts.
Security: Unlike centralized system, access to data not only has to be secured, but the entire network also has to be made secure.
Integrity control more difficult: In a DDBMS, the costs of processing and communication that we have to make to implement integrity constraints may be prohibitive.
Lack of standards: Despite there is effective communication when we use DDBMS, There are no methodologies or tools to assist users convert a centralized DBMS into a distributed DBMS.
Lack of experience: Unlike what we have with centralized DBMSs, still our experiences are lack in industry of DDBMS.
Database design more complex: The design of DDMBS is more complex than centralized database, because the data fragmentation has to be implemented in this design, and also data replication, and fragments allocation to specific sites
4- Possibility of using the replication server:
The replication is a mechanism enables the companies to provide their users accessing to current data at the time they need it (Connolly & Begg, 2004).
The DreamHome company can use the replication server to enhance the dealing with requirements, because if it want to make any updates to a replication object at one site are will be applied to their copies at all other sites. Furthermore, replication server enables the company users to replicate many objects, such as the tables and Functions and Procedures, Indexes, Views and Object Views, Packages and Package Bodies and so on (Urbano, 2003).
Implementation of replication server can be useful. Because by it users can transfer data to any local site needs it. And the company can insure availability of data and applications at all the time. Companies can insure that important applications are available at all times. Also, replication server allows any user access to its data from distance (DCI, 1996). So that it is possible to use server of replication in this company.
Finally the recommendations that we can give this company are
- It's better to implement DDBM to manage the data of this company.
- If this company use same operating system like Windows or Linux, it is better to integrate Homogeneous DDBMS, because, Its design is easier than heterogeneous DDBMS, as well as data managing, also It helps the company to add new sites to DDBMS. As well as It allows augmented performance by using the parallel processing capability of multiple sites (Edward, 2007).
Blurtit, d. (2010). Ask it answer it. Retrieved April 11, 2010, from blurtit: http://www.blurtit.com/q828920.html
Connolly, T. M., & Begg, C. E. (2004). Data base solution. A step by step guide to buildind database . England: Pearsoned.
DCI, D. (1996). Replication, The Next Generation of Distributed Database Technology. Retrieved March 13, 2010, from DCI's Event Central: http://www.dciexpo.com/geos/replica2.htm
Edward, C. ( 2007, Novamber 05). Distributed DBMS Concepts and Design. Retrieved March 13, 2010, from SJSU Department of Computer Science: http://www.cs.sjsu.edu/~lee/cs157b/fall2003/Edward_Chen_Chapter%2022.ppt.
Urbano, R. (2003). Oracle Database Advanced Replication. Oracl.
If you are the original writer of this essay and would like it removed from UKEssays please email the URL of this essay to us via the link below.