Database Management Systems - Case Study Example

DATABASE MANAGEMENT SYSTEMS Student’s Name: Code + Course name Professor’s name University City, State Date Introduction A relational database management system (RDBMS) refers to a database system that consists of data element files arranged in two-dimensional arrays of columns and rows. The database management system (DBMS) has the potential of recombining data elements to form various relations that enhance the flexibility of using data (Nichols 2007). The user perceives an RDBMS to be an array of two-dimensional tables manipulated as a single set at a time rather than one record at a time. Manipulating RDBMS necessitates the use of the SQL. On the other hand, logical data models describe data at both the view and conceptual models thereby providing structuring capabilities that are fairly flexible. The report covers the popular DBMS products under both the relational and logical data models. Moreover, the paper discusses the strengths and weaknesses associated with the use of relational and logical data models of database management systems. The Relational DBMS There are several products under the relational database management system. They encompass MS SQL Server 2008, Oracle 11g, IBM DB2, Ms Access 2010, and My SQL 5.5. MS SQL Server 2008 MS SQL Server 2008 is a relational database management system (RDBMS) developed by Microsoft. It uses Transaction-SQL as its primary language. The language is an implementation of the ISO/ANSI standard SQL used by both Sybase and Microsoft. The server supports consistent, atomic, durable and isolated transactions. It also has the capability of supporting database clustering and mirroring. The cluster of an SQL server entails a collection of servers that have an identical configuration that enables the effective distribution of workload among various servers. The server also handles distributed databases following its support for data partitioning. The data mirroring capability enables the server to create mirrors of the contents of the database besides the transaction logs. The creation of the mirrors that reflect the database contents depends on particular predefined triggers (Rankins et al. 2010). IBM DB2 IBM has developed a number of relational database management systems. DB2 is one of the systems. The system runs on UNIX, Linux or Windows server machines. The administration of the DB2 takes place either through a GUI interface or a command line. Prior to using the command line approach, it is imperative that the user should have adequate knowledge of the system. However, automation and scripting is easy with the use of the approach. In the case of the GUI interface, it entails a multi-platform Java Client that consists of several wizards appropriate for novice users. The server supports both XQuery and the Structured Query Language (SQL). The server has a native XML data storage implementation. As a result, it stores XML data as XML to enable enhanced access using XQuery. The server also supports integration into Visual Studio and Eclipse .NET integrated development environments. The error processing capability is an important feature of the DB2 server that enables it to use the SQL communications area structure to return error information to the server for an SQL statement after every API call (Zikopoulos et al. 2007). Oracle 11g Oracle 11g is a relational DBMS developed by the Oracle Corporation. It encompasses at least a single application instance as well as data storage. The instance entails a set of memory structures and different processes of the operating system that enables the interaction with data storage. The server also has online redo logs that hold history about previous transactions. It also consists of archive logs that refer to the archived online redo logs thereby providing the basis for data recovery and replication. The database stores data physically using data files and logically using table spaces. The data files consist of at least one data block that have varying sizes. The features of the Oracle DBMS include data dictionary, clusters and indexes. The version also has the grid computing capabilities that enable an instance application to utilize CPU resources from another grid node (Loney 2008). Ms Access 2010 The database produced by Microsoft combines software development tools and the graphical user interface (GUI) with the Microsoft jet Database Engine. The compatibility of the database with SQL queries is one of the major advantages of the Ms Access 2010. The Jet Engine does not have stored procedures and database triggers as is the case in other databases. However, it boasts of its potential to provide a special syntax that uses parameters to create queries. It is proper to compare the capability to the creation of stored procedures. The only difference is its limit of one statement pr procedure. The changes made to the table contents do not trigger any codes since the forms do not have the codes. Hobby programmers, departments of large corporations and small businesses use Ms Access to develop customized desktop systems that are responsible for creating and manipulating data. Ms Access is also applicable to rapid application development particularly the creation of stand-alone applications and prototypes to aid transit businesses (Conrad & Viscas 2010). My SQL 5.5 This entails a multiuser, open-source, free and multithread SQL database management systems used in various corporations. The program enables multiple users to access several database systems. It encompasses a subset of ANSI SQL 99, its extensions, stored procedures, cross-platform support, updatable view, cursors, an X/Open XA distributed TPS and triggers. Additionally, it provides support for a two-phase commit engine, embedded database library, replication entailing many slaves per master and one master per slave, query catching, SSL support, and the InnoDB clustered engine enabled ACID compliance (Kofler 2005). DBMS products based on a logical model The IBM IMS The IBM Information Management System (IMS) was the first hierarchical database system developed by IBM. The application modeled systems in which the segments or entities comprised of one-to-many relationships arranged logically in a descending order. “Child” and “twin” pointers established the relationships. The prefix of each of the database records contained the pointers. The system has concurrent control and a reliable recovery and backup mechanism to protect the system from its potential failure, the failure of the database and the failure of an application program. The recovery mechanism also protects the system from a failure of the operating system (OS) or the network control program. In the recovery of application programs, the mechanism record images both before and after the altering of the images. By so doing, it is possible to “roll back” the database in case the transaction is incomplete. In the event of a disc failure, it is possible to “roll the images forward”. The EyeDB Since its first development in 1993, the primary role of the EyeDB OODBMS was to store and enhance the access to human genome mapping data comprising of both genetic and physical maps. The system was more efficient in the storage and retrieval of human genome data as compared to any other existing system. It became a real OODBMS following a complete rewriting exercise carried out by Sysra Informatique. The key features of the system include the standard OODBMS features such as recovery system and client/server model, language orientation, orthogonality and genericity of the object model, supporting data distribution, scalability, efficiency and the support for large databases (Viara et al. 1999). The Objectivity/DB The Objectivity Inc. developed the Objectivity/DB object oriented database management product. The heterogeneous networks of client applications that can execute Objectivity/DB encompass DEC Station under Ultrix 4.2, Sun4/SPARC under Solaris 2.0, Solaris 2.1 or Sun OS 4.1, and VAX under VMS or Ultrix 4.2. The other networks that can execute the system encompass HP 9000 Series 300 under UX 8.0/HP, HP 9000 series 700 or 800 under UX 8.0/HP or UX 9.0/HP and IBM 6000/RISC System under AIX. Silicon Graphics Iris under IRIX 4.0 and NCR system 3300 under SVR 4 version 2.0 are the other client application networks that can execute the OODBMS. In the event that the applications run on any of the above platforms and share a local area network, they can access one database. The system also provides interfaces for ANSI C and AT & T compatible C++. The system is an open product designed to work with any AT & T compatible C++ and ANSI C compiler. The persistence of the system depends on its class. As a result, the system-defined classes are responsible for the attributes that enable the persistence storage of information. The database administration tasks performed by Objectivity/DB include data configuration, locking support, transaction support, recovery and backup of federated databases and the reclamation of unused storage in the databases (McFarland et al. 1999). The ObjectStore 2.0 The Object Design Inc developed the ObjectStore OODBMS. The platforms that can handle the system include Sun under both Solaris 1.x and 2.x, HP under HP/UX, DEC under Ultrix, NCR under SVR 4, Univel under SVR 4 and Olivetti under SVR 4. The other platforms that can host the system include IBM 6000/RISC system under AIX, Silicon Graphics and IBM PC under OS 2 and Windows 3.1. The application interfaces provided by the system include AT & T compatible C++ and ANSI C. The persistence of the system depends on the object. The programmer can create either a transient or a persistent object. Loading of the databases occurs either through the batch mode or incremental mode. The administration and monitoring utilities provided by the ObjectStore databases encompass starting, verifying and stopping the running processes, forcing application flushing and server checkpoints, installing schema definitions and setting parameters for the cache and serve manager process configurations. The user-level database manipulation utilities include controlling file access, copying, deleting and moving databases, exporting and importing database contents and identifying the host that serves the database. The other utilities that the user can perform include listing all the databases, deleting and creating database directories, performing schema evolution, reporting the size of the database and its segments and verifying all the database pointers (McFarland et al. 1999). Strengths and Weaknesses of the RDBMS At the time of the development of the RDBMS in 1970, the concept of the relational database management system emanated from the necessity of guaranteeing the independence of user-written application programs and providing a “natural” description of data. There are several advantages of the RDBMS. To begin with, the rigorous methodology associated with the design of the system relied on sound theoretical bases. The database also bears the potential of reducing all the other database structures into a set of relational tables. By so doing, the relational database is capable of providing a generalized representation of data. Thirdly, it is easy to implement and use the system as compared to other systems. It is also easy to modify the system thereby enabling the easy addition of new tables and data rows. Finally, the system boasts of its flexibility in the retrieval of ad hoc data arising from the powerful facilities of the query language and the relational joint mechanism. However, the use of the system necessitates greater process requirements thereby requiring more resources as compared to the other systems. In fact, an increase in the number of users implies a consequent increase in the resources required by the system. Secondly, there are slower response times in the event of heavily loaded systems since the queries of such systems involve multiple relational joints. However, the effective use of optimization strategies such as indexing can play a major role in alleviating the weakness. The continuous improvements in computer hardware performance can also reduce the response times (Suri & Sharma 2011). Therefore, it is evident that RDBMS require powerful hardware for effective performance. It is also proper to observe that the ease of design of the RDBMS may be an avenue for a poor design of the system. It is not mandatory that the user of the system should understand the complex processes associated with the storage of physical data. The ease associated with the design, implementation and use of the system may result in the development of a poor design. Even though the database is efficient, it is certain that the efficiencies of the database will not manifest in the event of handling small data. However, as the amount of data increases, data corruption and performance degradation will manifest thereby revealing the poor design of the system. The ease of using and implementing relational database systems may also result in the information island phenomenon. The departments or individuals may opt to develop their applications and database thereby impacting negatively on the integration of information in the organization that is a prerequisite for the smooth running of the organization. As a result, data redundancy, data inconsistency and data duplication may arise from the phenomenon (Suri & Sharma 2011). The existence of a lot of information in the system may yield information complexity thereby setting back relational databases. Apparently, the relational database systems collect data with reference to specific characteristics. The increase in the complexity of information may result in complex images, design products, multimedia and numbers thereby impeding easy categorization that forms the basis of the object-relational database management system (ORDBMS). Scalability is one of the main features of the system that enables it to handle complex information. Finally, some relational database systems have limited field lengths that may result in the loss of data if the data input cannot fit into the field (Suri & Sharma 2011). Strengths and Weaknesses of the logical database systems A hierarchical database system is a good example of a logical database. There are several advantages of the logical database. Better performance is one of the major advantages of logical database systems. With reference to a hierarchical database system, it is apparent that it is easy to navigate through the records. The ease of navigating the records arises from the fact that pointers implement the relationship between the parent and child entities from both parent to parent and child to child. Similarly, the sideways relationships from parent to parent and child to child also have pointers. As a result, it is not imperative for the program to search for the index after finding the first record. Moreover, the program does not have to perform a table scan to determine the subsequent record. The program only follows one of the existing child record pointers, the single parent record pointer, or the single child record pointer to gain access to the “next” record. The ease of understanding is the second advantage of the logical database system. The database organization parallels the family tree or the corporate organization chart. As a result, even non-programmers have a familiar feel of the database system. Logical systems also guarantee the easy detection of database relationships such as where entity 1 is part of entity 2. Since object oriented databases also store information in a logical model, their advantages mirror the advantages of logical models of database management systems. As a result, the capability of handling different data types is the other advantage of object oriented databases. The system allows the user to store different types of data including texts, numbers, voice, video and pictures. The high efficiency of the system also emanates from the fact that the system is a combination of both database management and object-oriented programming to yield an integrated application development system. Including the definition of both data and operations as is the case under OODBMS is advantageous to the system. Firstly, it implies that the defined operations are independent of the database application and applied to the intended function. Moreover, it is possible to extend the data types to enable the handling of complex data. The inheritance attribute of the OODBMS that uses the logical model also improves the performance of the system. Inheritance enables making reference to the already defined past objects in the definition of new objects. With the use of dynamic binding and polymorphism, programmers can define operations for an object and share the operation specifications with other objects. The latter objects can extend the operation to include unique attributes applicable to them. Finally, the explicit representation of relationships in OODBMS enhances data access since it supports both associative and navigational information access. However, starting with the hierarchical database systems that also used the logical model, it is proper to note that that they have a rigid structure. For instance, it is imperative for the system to create a new table to accommodate larger records when the user intends to add a new table or record. This is not the case in an SQL database that has an ALTER TABLE command to handle such functions. Moreover, in the event that the user wants to add a new relationship, the individual has to construct a new and redundant relationship. Some of the disadvantages of OODBMS that uses the logical model include the absence of a universal data model and adequate experience on the part of the users of the system. There are also no defined standards associated with the use of the system. Furthermore, the optimization of queries compromises encapsulation. The ability of the system to lock the level of an object may also impact negatively on its performance. The inclusion of object-oriented programming into the database technology also increases the complexity of the system thereby making OODBMS products more expensive and difficult to use as compared to the predecessors. Finally, there is inadequate support for views and security in the use of the system (Maronsi et al. 2000). Conclusion A relational database management system consists of data files arranged in two-dimensional arrays of rows and columns. Database management systems that have a logical model describe data at both the conceptual and view levels. Some of the common RDBMS products include MS SQL Server 2008, IBM DB2, Ms Access 2010, My SQL 5.5 and Oracle 11g. Some of the DBMS products based on a logical model include IBM IMS, EyeDB, Objectivity/DB and ObjectStore 2.0. The ability of the relational databases to provide a generalized representation of information suffices to be the main strength of relational databases. However, the effective performance of the system necessitates powerful hardware architecture. Better performance arising from the ease of use and implementation is the major advantage of the database systems that utilize a logical model. However, certain systems such as OODBMS are complex. Reference List Conrad, J & Viescas, J 2010, ‘Microsoft Access 2010 Inside Out’, Microsoft Press. Kofler, M 2005, ‘The Definitive Guide to MySQL 5’, Apress. Loney, K 2008, ‘Oracle Database 11g: The Complete Reference’, McGraw-Hill Osborne Media. Maronsi, H, Löfqvist, D & Larsson, M 2000, ‘Object-Oriented databases’. McFarland, G, Rudmik, A & Lange, D 1999, ‘Object-oriented database management systems revisited’, Modus Operandi Inc. Nichols, L 2007, ‘A Comparison of Object-Relational and Relational Databases’ (Doctoral dissertation, California Polytechnic State University San Luis Obispo). Rankins, R, Bertucci, PT, Gallelli, C & Silverstein, AT 2010, ‘Microsoft SQL Server 2008 R2 Unleashed’. Suri, P & Sharma, M 2011, ‘A comparative study between the performance of relational and object oriented database in data warehousing’, International Journal of Database Management Systems, vol. 3, no. 2, pp.116. Viara, E, Barillot, E & Vaysseix, G 1999, ‘August. The EYEDB OODBMS. In Database Engineering and Applications, 1999. IDEAS'99. International Symposium Proceedings (pp. 390-402). IEEE. Zikopoulos, P, Baklarz, G, Katsnelson, L & Katsnelson, L 2007, ‘IBM DB2 9 New Features’, McGraw-Hill. Read More