Transition of Traditional Relational Database

¶ … Transition of Traditional Relational Database Technology to Object Based and Knowledge Database Technology

This capstone project will address the issue of the transition of traditional relational database technology to object based and knowledge database technology.

Relational database

This type of database stores all of its data within tables only. All the actions on the data contained in a relational database take place within the tables and as a result an additional table may be generated. Nothing else is observable except the tables that are created for the purpose of data storage. A table is defined as a set of rows and columns and its elements do not possess any form of predefined order. Every row of the table comprises of a set of columns having only a single value. It is however worth pointing out that all rows that are derived from the same table do possess the same columns sets. In certain cases however, the columns get assigned NULL values. This is to mean that the value in the particular row is never initialized. It is worth pointing out that an uninitialized column does not have the same meaning as a NULL- valued column.

It must however be noted that relational databases do support relations (Codd, 1970; Date, 1985).The relations which it supports is made up of a set of tuples which comprises of certain fixed number of data which is primitive in nature. As a simple analogy. The rows in a certain relational database can be viewed as being analogous to a record while the columns are analogous to a field.

Below is an example of a table and the corresponding SQL statement which generates the table.

CREATE TABLE ADDR_BOOK (

NAME char (25),

COMPANY char (15),

E_MAIL char (30) )

NAME | COMPANY | E_MAIL

| Smith Kevin| Software Systems |

| John | HP | -- |

It is important to note that there are basically two forms of operations that can be performed on a relational database. The first operation involves the retrieving of data from the database. The actual process involves the retrieving of various subsets from the columns of the database and then the second operation involves the retrieving of subsets from the various rows of the relational database. Below is an illustration of the two actions on a relational database.

SELECT NAME, E_MAIL FROM ADDR_BOOK

| NAME | E_MAIL |

| Smith Kevin| -- |

| John | -- |

SELECT * FROM ADDR_BOOK WHERE COMPANY =

'Software System'

| NAME | COMPANY | E_MAIL |

| Smith Kevin|Software System | -- |

The two operations can also be combined as follows:

SELECT NAME, E_MAIL FROM ADDR_BOOK WHERE

COMPANY = 'Software system'

| NAME | E_MAIL |

| Smith Kevin| -- |

The two operations can also be executed between two tables while treating the two tables as two separate sets. A Cartesian product of the tables can then be evaluated and then the intersection between the two tables can be evaluated. One table can be summed up with another as an example.

Object oriented databases.

In our analysis we focus on object oriented systems through a careful introduction and explanation of all the basic concept of object oriented systems. In the process some opinion is presented of the suitability of incorporating the concepts themselves into future database designs. It is however worthwhile to note that the introduction into the market of very versatile and applicable systems like the ObjectStore (Lamb et al. 1991) have made it very possible to conclude that it is high time that a thorough analysis and investigation be launched in order to find out the possibility of migration from traditional relational database technology to both object based and knowledge database technologies.

Reasons for the transition to object oriented database systems

The increasing need for the transition from the traditional relational database systems to the object oriented database systems is fuelled by a variety of reasons and applications that are ripe to make use of the dynamics that can only be actualized through the use of the latest technologies in database systems. A perfect example is the applications of the object oriented database technology in the field of Computer Aided Manufacturing .Other examples include the use of object oriented technology in the process of designing office automation systems that are intended to handle various forms of data. The other application of the objects oriented database technology is in the hospital sector where it is used in order to design various forms of patient care tracking systems. A common characteristic of all these applications however is the fact that they are involved in the management of highly complex operations which involved the sorting out of various types of interrelated information. It is worth pointing out therefore the fact that relational database systems have failed in addressing key issues such as the handling of highly complex information systems.

The other main problem with the relational information system is the fact that in forces the developer of the system to forcefully apply an information system/model over tables in which the interrelationship between the entities are only definable by values. As pointed out by Loomis, the architect versant, an object oriented database management system; the relational and object oriented database systems can be compared as follows.

The process of designing a relational database involves actions devoted towards figuring out of ways of representing real-world objects by means of tables in manner that allows for the preservation of both the data integrity as well as the performance of the database. It is worthwhile to point out that the process of designing the object database is critical part of developing every application.

The classes of objects that are utilized by the various programming languages are the exact classes that are exploited by the object oriented database management system in use. This is made possible by the consistency in the models used. There is therefore no inherent need for any further transformation of the model of the program being used in order to allow for a perfect fit with the unique model that is presented by the database management system in use. It is important to acknowledge that fact that the initial area of application of object oriented database systems was mainly in the field of Computer Aided design (CAD), Computer Aided Software Engineering as well as Computer Aided Manufacturing (CAM) applications (Alam et al. 2006)

Once again it is important to point out that the primary characteristic of all the applications that make use of object oriented database management systems is their ability and need to offer efficient management of complex information. The other fields in which object oriented database systems find applications are office and factory automation. A perfect example would be the manufacturing of a motor vehicle which demands the tracking of several independent parts that must be assembled using different configurations. The only database management system that offers a viable solution to this kind of arrangement is the object-oriented database management system.

It is therefore important to acknowledge the fact that object-orientation is part of the effort by various database architects efforts to come up with a method of solving the traditional relational database related problems in a way that is more natural and even easier to fathom. This is better expressed by Brodie (1985) in his book regarding Conceptual Modeling as "The fundamental characteristic of the new level of system description is that it is closer to the human conceptualization of a problem domain." In other words, the adoption of object oriented programming systems can aid in improving the communications between various system designers. This would have the advantage of improving the time required to conceptualize, plan and execute the design of particular object-oriented database applications.

The database market is undergoing rapid transition

The database market the world over is undergoing a paradigm shift in terms of its need and offerings. This is because the period in which we are in is marked with intensive changes and innovations as regards the various database technology and its products (McClure, 1997)

Several years back saw the introduction of additional data types.these additional data type are commonly referred to as the complex data. These data types include multimedia data types such as text, image, audio and video. The other types of complex data are time series as well as geospatial data types.

Literature review

The current trend in database technology is one that will potentially expand the area to meet modern needs for information sharing and data management. The newest ideologies are functionally different from relational database technologies in that they demand much larger memory stores and most importantly are offered in a platform that to some degree thinks for the user or organization. The traditional database model, which requires searching and some initial knowledge of what exists on the system and how to access it is being replaced by systems that do a great deal of the thinking for the user. They in short link and map ideas that already exist, creating real time answers to questions as they are applied to new and archived knowledge or action plans.

According to IDC reports, most data warehouses will be stored in a columnar fashion and not in rows, reporting and data collection problems will be solved with databases that have no formal schema at all, horizontal scalability through clustering will be achieved by large-scale database servers; and most OLTP databases will either reside entirely in memory or be augmented by an in-memory database. These new systems will encourage companies to forget disk-based partitioning schemes, buffer management, indexing strategies and embrace a world of large-memory models, many processors with many cores, clustered servers and highly compressed column wise storage. (Prakash, 2010)

The formats mentioned her by Prakash (2010) are fundamentally responsive to a new form of data storage and formatting, where objects are saved as objects, rather than as object types or keyword searched data bits. In Kopteff's (2008) technical work the author demonstrates a direct comparison between an object relational database (Hibernates) to an object oriented database program (Versant), on several levels. In this work Kopteff (2010), concludes that the object database program was in fact easier to use and provided more rapid access to the user, due almost entirely to the fact that the object-based format did not have imbedded issues of mismatching, as well as secondary issues of development that allowed the developer to avoid the concerns of object relational mapping conversions.

From the usage section we can conclude, that using Versants interfaces is at least equally easy than using Hibernates interfaces and in most cases easier, since it lacks the object-relational mismatch configuration. Using an object database lets the developer focus on actual design problems and not on the problems of the ORM tool or the conversion of objects to relational form. When considering the results of the performance comparison, Versant was faster in most cases, hence it is more efficient to use. When looking at the results as a whole we can conclude that Versant is more useful for developing applications, since users can use Versant to develop applications more efficiently and with less workload, than with Hibernate. This also results savings in costs, which could be one of the key factors for the usage of object database to spread. (Kopteff, 2008, p. 11)

The overall intentions of the work were to present the viability and applicability of replacing object relational mapping programs, which take objects and codify them rather than simply organizing and storing them as objects in their true form, with object database systems. Clearly in the application of these two systems in a controlled setting the researcher makes a case for stepping up next-generation development, rather than simply attempting to place older templates on new knowledge-based demands.

Discussing the issue of Knowledge Management can be secondary to the highly technical discussion of how such information is stored, and in what form as relational database technology can answer many of the existing needs for knowledge management. Yet, clearly the object database format, with its many pros can be seen as a plausible answer to knowledge management. As Segev points out in a 2010 article on application of Knowledge Management to a usable database system most knowledge management databases are geared toward larger organizations, and the ease of development is largely ignored, as the need to develop outweighs potential cost of doing so. The point Segev makes is that small and medium sized organizations are now seeking knowledge management systems that offer ease of use and rapid access to applicable data. This transition implies that the databases that provide access to archival, new and transformative data must streamline and become more accessible for both internal and web based application. (2010) the easy leap is that they should do so using object oriented database technologies, as they remove a great deal of obstacles, and will prove to be time-saving devices that can be programmed to meet specific needs of the users. Given that small business make up the majority of businesses in the U.S. The need for applicable database technology for knowledge OBD knowledge management tools is obvious. According to the U.S. Census Bureau in 2004 there were just short of 5 million (4,980,165) businesses in the U.S. that employ between 1 and 99 employees. (U.S. Census Bureau, 2010) These businesses form the backbone of the American economy and if successful continually seek to improve technology, where applicable to better organize, manage and communicate. OBD and knowledge management technology may seem an illogical choice for some, but even the smallest business needs to stay competitive as few businesses successfully endure based on sheer luck and the skill of a single person on plan. These businesses, to stay competitive will be seeking database technology and will likely seek skilled it consultants if not employees to help them better organize and grow their business.

Rationale and Systems Analysis

The focus of database technology over the last 30 years has been placed almost exclusively in the development of traditional database technology. Ultimately, though exceedingly useful this technology does not consider or handle items as objects, offering only very limited utilization of multimedia all in one next-generation technology nor does it solve the many problems associated with mismatching errors, or most importantly offer search free time savings. Knowledge database technology, which is in a sense a form of artificial intelligence offers the ability of users, be they machines or humans the opportunity to solve knowledge-based questions or concerns using historical data and logical rapid reasoning. These two monumental but distinct forms of database technology are the next generation of database systems and offer users countless productivity improvements. This being said the time frame for such developments has been set aside, in large part due to the increases in hardware and software technology that can manage much larger database forms either in house or in secondary server locations. The need to create and offer systems that manage much larger compilations of data and at a much more rapid rate is likely the source of setting such technologies aside. Though there are forms of the technology that can be added to existing databases these are but temporary fixes and serve only to delay a monumental need that will arise in the future.

Though it is clear that information technology and database technology have led many to believe that knowledge and object-based database technology has significant merit, and will likely be the wave of the future current technology alone cannot create such as system. The technology to create such systems lies not only in the conception of the possibilities, but in the actual development of standalone database systems that are knowledge and object based. We now know that a chemist in Iowa can rapidly access all the information he or she needs to determine the current state of research (usually within a company or academic database) or he or she can contact other chemists anywhere in the world yet, the systems of contact and knowledge gathering are still laborious. The result is then individuals still feeling as if they are working in a vacuum or that they cannot gain such knowledge without taking time away from the main focus of research their research. This analogy is one of many that can be posed as proof that this technology must expand to next-generation knowledge and object database standards. Knowledge management databases in congruence with object database systems would, to some degree allow real time interaction between individuals, organizations, and the data they need to make decisions and streamline information. Rationale for this study is to support users, both large and small to expand database technology in a system that will allow for the mass storage and access capabilities for knowledge and object databases rather than continued reliance on traditional relational databases. "The widespread popularity of object-relational mapping (ORM) tools still raises the question of using an object database instead of a redundant data mapping tool and persisting data in its natural form -- as objects." (Kopteff, 2008, p. 1)

A systems analysis is clearly due in this area, as current database technology has developed in a very random pattern, to meet the immediate needs of programmers, users and organizations as well as to fit within the parameters of current technologies. These current technologies according to most now support the ability of databases to be object-based rather than the stop gap solution of object relational or simply relational in format. (Kopteff, 2008) Yet, even those same users have clearly come to the end of their ability to use database system as they are. This is in part due to the expansion of knowledge itself, and particularly archived knowledge and due to the fact that those same users can see the possibilities of object based and knowledge-based databases but have yet to see them take hold. (McDermott, 2000, pp. 21-36) the systems analysis that was conducted to both better understand and rationally develop and ideology to move forward with a technical project in the areas of next-generation database technology demonstrated that the technology is cutting edge, and demands usable systems that are both easy to find and easy to implement. The work also demonstrated that knowledge management databases, translate well into object-based databases, as such databases require less coding and can function on a multi-media dimensional level. Multi-media applications (increasing in use almost exponentially) require formats and function of objects because they differ and the user is in need of databases that have less up front programming concern. New databases also must confront the need of rapidly allowing decisions to be made, as objects are grouped and linked according to need, and are mapped less rigidly and stored in ways that make them easier to find. It must also be noted that object-based databases (OBDS) are not currently standardized and therefore the work of organizations that are creating such standardizations is essential to producing products that are cost-effective, especially given the increased cost of technology needed to run them. (OBDMS.org)

Goals and Objectives

Goals and objectives for this work should commence with a review of available OBD programs. Given the stated need in the literature for affordable software solutions for knowledge management, for smaller organizations and even individuals the first goal of the work will be to determine the current state of OBD technology, in both universalization and available technology through current available reviews of technology. The need for a basic review of available software is evident from this brief review of research, provided in the proposal. It is therefore important that one of the goals or objectives of this work be to create a functional review of object-based databases, by issues of cost, technology needs, database capabilities and usability. There is currently a great deal of available literature on the issue as well as several comparison works, which will be utilized to develop a more user friendly way to find and choose available software. This review will serve as a tool for recommendation of OBD software to users in the future and serve as the initial crux of the project deliverables.

The second, although primary goal is to apply the use of an OBD to an existing small-scale knowledge management need. For the purpose of this project, the existing need will be to apply the OBD (Versant) to a student problem. The student problem will be to increase organization and hopefully productivity through OBD technology. The student application program will map all the cumulative knowledge available for the completion of a degree. This will include current (cumulative) and archived projects as well as all the electronic data collected for each project, a timeline for course completion that aligns the plan with course availability over the length of the degree and allows the student to map his or her course load according to an existing work schedule. Additionally, all syllabi for current courses and any available for past or future courses will be available in the knowledge management database, concurrent with electronic course notes, where available. At least one course text will be scanned into the system to determine the nature of the object-based system for multi-media use and usability of electronic picture files for reading. The system would also be used to time-map and schedule study and research time around class schedules, based on personal estimates of time needed to complete projects and course work. The application of this program will be developed using my own course load and history and then be expanded to allow the communications aspects of the program to be tested. I will attempt to recruit at least five other students, in the it department, to aide me in this project, allowing communication and knowledge sharing in a secure format so materials can be read but not copied (for the purpose of reducing possible plagiarism). The initial database development, using my own materials and course work will allow other students to do additional input for variations in degree seeking and course timelines as well as projects and materials. Students will then be asked to fill out additional reviews and use the technology to communicate the reviews to me as well as any other pertinent information, such as concerns about how to use the database and best apply the materials to it. Student participants will be encouraged to at the least minimally, input information into each aspect of the database, i.e. A single current course, its syllabus, textbook (if possible) and a single current project as well as to utilize the time scheduling aspect as much as possible and to input data for degree and course requirements, or long-term-goals aspect of the work. The researcher will be available to input course data and degree requirements for each student if they choose to ask for such assistance and to cut down on time demands of the beta test.

The objective of this goal is to demonstrate the ease of use of a single currently available OBD product, for the purpose of both developing user specific database development and creating a viable real life application. The objective will be met not only by application but by a review of the ease with which information was mapped and usable after it has been developed and programmed into the OBD. This review will also serve as a crucial aspect of the project deliverables. Although, a multitude of programs are available to better manage individual time, and information all of them are fragmented and require the use of many different programs or tools. The utilization of a single OBD and its functional and updatable format will allow all the information to be collected and used in a single database, which is the primary purpose of all databases.

The ultimate goal will be to determine the efficacy of OBD software as it is applied to a small-scale knowledge management need, for the purpose of determining if a small business or other organization or detail oriented individual could functionally benefit (through time saving and better organization) in production and development with existing available software. (Segev, 2010) the purpose of this aspect of the goals is to determine the need for the creation of new software, and to gage, in a real life example the state of existing software. Developing the database and reviewing it will serve a multi-tiered purpose, not the least of which is to better understand how OBD works in this application and how to better manage it in a small-scale yet real world application.

The choice of this application demonstrates an analogous comparison to both long and short-term goals of organizing in a small business. Short-term goals being the development of an existing project and long-term goals to map the entire course of the next few years, noting variations in time according to available course offerings and decision making regarding when and how to complete them. Ultimately, the most efficient way of achieving an end would be stressed, as individuals could choose to map out the next few terms, ticking off requirements as they are offered, rather than seeking to blindly drudge forward with the possibility of being a few courses or credits shy of completion in the final few terms of their coursework. In a small business there are usually multiple projects occurring, in multiple periods, all which must be managed according to short- and long-term goals. The project based applications as well as an overall goal of efficiency are key to organizing in many smaller organizations and the ability of a few people to rapidly access all the archived and current knowledge is crucial to progress and profitability. Additionally, in a small organization the owner/manager is likely willing to delegate only tasks he or she is ensured others can effectively complete. For this reason, knowledge mapping could serve as a way for the owner/manager to track progress without holding meetings or constantly asking for updates from employees. The ability to delegate then would be enhanced as he or she became more adept at checking progress within a database.

On a personal level the development of a complex knowledge mapping problem using OBD technology, which is fully supported for student use (Versant Academic Program) is essential to development as an it professional, as OBD technology will surely be the wave of the future and requires alternative skills to manage and develop for organization use. This work will demonstrate a clear route to applying this development tool to a real life and markedly interesting problem, i.e. The need to learn and better organize time and knowledge on a small-scale individual level. The work will also help build my technical writing skills. Students are clearly at risk of falling behind in their studies as a result of poor time and information management and as degrees become more demanding, i.e. At higher levels the requirements and time constraints are far more extreme. To built the efficacy for utilizing such a tool, as supported by a group of individuals also using it would provide informed feedback, in a timely manner that could ultimately make a big difference in costly time errors made by students regarding project and even graduation requirements. The application could be used across colleges, as individuals seeking nearly any type of degree could apply the tool to reformulate the manner in which they manage time. The use of the tool for it degree seekers is the result of the need for the program evaluation from a development standpoint as such informed users could offer assistance to others with regard to the best and most practical ways of using the OBD software for development and application.

Success, associated with this project will be measured by comparison of time saving tools in organizing goals, as opposed to a previous period where such information was not mapped but was instead kept in non-electronic form in various notebooks, calendars, books, online databases and stored in personal memory. The reviews of the program application will serve as the most logical outcome measures, as the reviews will ask a series of questions that address usability, additional or variant development options and communicate all these in real time. The review will then ask the user to offer feedback regarding the measure in which the program has helped them better manage time and projects. Even if users choose only to apply some of the functions of the developed database, the outcomes can still communicate scale measures of improved time/project management. Users will also be asked to utilize the database as much as time allows, hence scanning textbooks, lecture notes, syllabi, but most importantly the time scale measure for course completion. The review will also ask the user to identify all the aspects of the database he or she utilized and briefly explain why he or she did not use others (if this is the case).

Why Versant?

The Versant object-oriented database is chosen due to the various powerful advantages that it offers applications that make use of various complicated object models with a relatively high level of concurrency needs as well as diversified data sets (VERSANT.com).

It is a fact that the process of mapping objects into various relational databases is hard, expensive and time consuming. The choice of versant is suitable as it offers solutions that can offer the performance, better navigational access as well as scalability that is demanded by the various applications that are developed in Java and C++.

The following however are the key reasons why Versant's object oriented database solution is better than other the use of the traditional relational database systems in executing the project at hand.

The complex interrelationships that exists in the application to be developed

Since there are several relationships involved in the Versant Academic Program, it is important to select the best platform to use to handle the complex data schema that is to be present and executed within the database. This is because it is necessary to select the best and yet affordable object-oriented solution that is capable of perfectly executing the various relationships that are to be handled in the system to be designed.

The complexity of the data to be handled

It is fact that the system to be developed will handle a variety of complex data. The only way of taming this complexity is the use of an object-oriented database system such as Versant in the process of developing the application. The use of Versant system allows one to avoid disasters associated with the use of other architectures that combine technical requirements for example persistence in SQL, with domain models. The choice of Versant offers a solution that allows for the development of applications using objects that only require the domain behavior as such frees the developer from the dreaded concerns related to persistence.

The lack of a Mapping layer

It is indeed very difficult, expensive and time wasting to map various objects into the traditional relational database systems. The whole process is also detrimental to the performance of the application under development. The ability of the Versant object oriented database system to store objects as objects while allowing for other formats of storage such as the ability to store data sets in the form of many-to-many as well as network and tree relationships is a merit. This allows for relationships that are bi-directional and therefore eliminates the need for having to use JOIN tables. This therefore means that a lot of valuable time will be saved during the programming phase as well as the phase of information retrieval from the database.

The fastness and ease of development using the Versant object-oriented database system as well as the ability of the solution to cope with the ever present evolution of data

The Versant object-oriented database solution is often the best choice for developing application with a lot of ease

Project Deliverables

Project deliverables will include a formal report, which will reiterate the rationale for the report by offering technical and definitive explanation of OBD and knowledge management database technology. The given demand for such technology for small scale users, i.e. small organizations make this aspect of the formal report crucial as a rationale for applying an OBD knowledge management database to such a small scale project.

The second aspect of the formal report will provide a software overview, which will summarize the applicability of different OBD knowledge management database software to different types of user needs. It will determine the technological requirements, i.e. added hardware, server support and number of systems each price range allows use on. The work will also seek out information on usability reviews, programming language and platform, user support and training opportunities and costs, update timing and availability and as many other key bits of information that can make the decision of choosing OBD or knowledge database software easier for novice users. The emphasis will be on programs with wide availability as well as those, which emphasize use by smaller organizations.

The formal report will then detail the technical project, explaining the rational, objectives and goals and most importantly the results. The application of innovative OBD knowledge management databases to businesses is clearly a new and emerging issue in it and one that deserves serious consideration, as it is to a large degree a serious paradigm shift in database technology. The ability of new it professionals to develop software formats using this technology is likely key to individual as well as global success of business. Due to the fact that a great number of businesses in the U.S. are small businesses the applications can no longer focus solely on large scale projects or organizations. The formal report will then focus on the application of such databases in small scale.