Data Warehousing - Report Example

Abstract This paper deals with various segments of Data Warehouse. It discusses deeply all the components and necessary processes and techniques which are used in a Data Warehouse setting. A Data Warehouse is a repository of integrated information, collected from any number and variety of data sources, including various databases and legacy data sources. The size of a data warehouse is generally enormous and the data warehouse normally stores a wide range of information that has been generated over long periods of time. Data related to business subjects such as products, markets, and customers are all collected, integrated, and housed under the data warehouse umbrella. When this vast wealth of information is interfaced to decision support tools that offer powerful data access and analysis capabilities, the data warehouse can be fully exploited by its users. There are available a number of data warehousing tools and products presently on the market. All of these products provide comparatively easy access to the data warehouse, both at the enterprise and the data mart level. All of these tools also provide positive facilities for administering and managing a data warehouse. However, creation, maintenance, and daily administration of data warehouses are still alarming tasks that are far from being fully automated. The triumphant administration and management of a data warehouse demands skills and proficiency that go beyond those required of a traditional database administrator (DBA). There is a need for the creation of a position that is designated a Data Warehouse Administrator (DWA). The tasks of a DWA include those of a traditional DBA, but the DWA's job is significantly more multifaceted because of the nature of the data warehouse and its position within enterprise data architecture. [Professor’s surname] [Course title] [Date] Data Warehouse Introduction Data Warehouse: A data warehouse stores all the meaningful information that a business accumulates. The information in a data warehouse must not only make data available to organizations, but accessing the information must also be efficient. Data warehousing is comprised of stored information in a database from an organization's activities. This database was developed to store information that would aid a business in making strategic decisions. Organizations rely on a data warehouse because there is a greater chance of transactions being completed in a specified time limit. Analysis Data-Warehousing Cycle: The data-warehousing cycle consists of a series of processes that begin with a collection of required data, followed by cleaning the data, storing the data, and finally dissemination for achievement of business goals. We now elaborate on each phase 1. Data Gathering: Necessary data for decision making and analysis from various sources need to be gathered. Data sources can be both internal and external. Internal sources are mostly a company transaction processing systems or operational systems, and external sources include government, industry, and competitor information. Accumulating data from both internal and external sources helps to limit the effects of the bounded rationality phenomena. Care should be taken to gather only data items needed for decision making. 2. Data Cleaning: After the data is gathered it needs to be cleaned before it can be used. Data cleaning is the most tiring task as it takes a lot of time. It involves a multistep procedure, this includes three steps i.e. validation, aggregation, as well as transformation. 3. Data Summarizing: A vital advantage of a data warehouse is the summarization level related to data. In transaction processing systems you have to summarize individual data elements to obtain the complete picture. Data warehouses take little effort out by summarizing queried information on regular basis .The summarization level will differ, depending completely on organizational needs as well as the system design. Once summarized, data are ready to be loaded into the warehouse. 4. Data Loading: There are a huge number of tools as well as approaches available for data loads. Today, with advances in networking and distributed processing, it is customary to have parallel load approaches 5. Data Retrieval: Most data warehouses are implemented in distributed client-server architecture, in which the server holds the data warehouse and users access these warehouses via client machines. This architecture has several benefits, as it allows for a central repository of data and only the most important server needs to have the necessary space as well as the computing power to hold the warehouse. Benefits of Data Ware House: Data warehouses have been around for the last decade and their benefits are well documented. They provide a single outlet for all validated and summarized data needed for decision making within an organization. Hence, they have associated benefits, such as simplicity in use, better-quality data, and faster access times resulting in sustaining competitive advantage through strategic management of data. 1. Simplicity: Data warehouses provide a single image of the business’s outlook and help users analyze a coherent set of data. They give benefits of a single client-service application containing very important data. Analysts do not need to have computer programmers in order to access the data. All that is needed are basic computing skills, as the whole thing is handled through a number of simple keystrokes or mouse clicks. 2. Better-Quality Data: Data warehouses validate each element of data prior to loading. This allows for a well-defined, concise, and error-free basis for decision making. Transaction systems, on the other hand, suffer from high volatility and their distributed nature does not allow for feasible analysis and planning. Without quality data, analysis performed will be faulty and will lead to failures. Enterprises using a warehouse enable predictions to be made using high-quality data on customer behavior in addition to analysis of market patterns, which is vital for attraction of prospective customers. Unrealistic Expectations: Management frequently scopes out a data warehouse to solve all business problems. A data warehouse should never be used to solve more than a few clearly defined business goals. Companies often try to combine warehouses with other operational systems, which leads to complexity and failure. So the manage should always set goals from the initial stages and should also be realistic about expectations. Data Acquisition and Quality: Data quality is a critical issue, as the users of the warehouse trust the information implicitly and it determines the usefulness as well as the relevance of decisions. Without quality data, we go back to an old computer. It is hard to convince users to use the data warehouse is they don’t have trust on the used data. Uncoordinated Development Efforts: Data warehouses have need of the dexterity of efforts in the development phase from approximately each and every functional area in a venture. The data-warehouse team cannot in any way consist of just programmers along with a single manager. Representatives from each and every department have to be willing to share equal accountability for the assignment. If we were to construct a data warehouse for investigation of customer behavior, we would require a marketing representative to assist scope out what information is required to be placed in the warehouse, research as well as development would be required to contribute to their views on how to integrate new product details, finance would require to look for feasible alternatives to finance the project in addition to calculate return on investment, in addition to systems analysts would need to help in the design of the user interface . Budget and Time Constraints: Data-warehouse projects are expensive and time consuming. The project should not be rushed to be attempted. A significant amount of time should be devoted to analyzing and understanding user needs and expectations. A clear-cut strategy and business goals should be laid out. Recruiting the right talent pool or consulting firm is also critical. Approximately data ware house costs $2 and $3 million and it takes two to three years to be completed. . Hence, sizable investments need to be made. Data warehouse projects also need to be delivered on time, as failure to do so can lead to significant losses of time, money, and effort. Executive Information Systems: Executive information systems provide a snapshot view of business processes and functions. These systems serve as dashboards for high level managers who want to see various signals of their business’s health. Data from a warehouse can be repopulated automatically into reports for managers. Executive information systems have limited use, as they are mostly static. They give an overall picture but cannot be used for diagnosis. Generation of accounting statements such as profit-and-loss statements and balance sheets are to a large extent a kin to executive information systems. They provide investors with a snapshot view of a business’s state but have low analytical or decision-making value. Useful and useless Internet sites: When the Internet was first introduced to the public, it marked an extraordinary change in every aspect of our lives. Today, with the explosive increase of sites on the Internet, technology has taken another dramatic rise. There are not doubts that Internet has become a very interesting tool in doing business. But in order to survive in such a competitive environment that faces constant changes and revisions everyday, it is necessary that entrepreneurs develop more skills than those mentioned above in the traditional way of managing small business. Despite the high importance of those skills, which are still essential in today's management of a business, managers have to develop a sharper sense to identify opportunities, have a stronger vision of future, develop a marked sagacity about timing, start thinking globally, think in a more entertaining way to attract customers and offer customers added value through customer service For finding useful information is challenging, tedious, and time consuming. A large portion of the information on the Internet is commercial and useless for research. Also, determining whether the information on a certain website is based on actual facts or a point of view is virtually impossible - especially when you have two or more websites that contradict each other. Conclusion: From this report we can derive one important message: companies and organizations, now more than ever, mainly compete on the basis of their intelligence, on their capacity to combine the knowledge of their members as well as of their networks, in other words through their intangible resources. For executives, the integration of these various dimensions, allows us to foresee the emergence of new challenges, especially those related to the trade-off between a short-term logic of optimization and a long-term logic of potential creation via the accumulation and the valorization of knowledge The paradigms of innovation explained above suggest that the new challenges for companies are more inside their frontiers and around their borders than in market structure analysis. An essential paradox arises from this assertion: The definition and implementation of a management of intangibles can be done only by the recognition of a primacy to people; whereas this element is generally treated as a resource whose costs are to be 'optimized'. Far be it from me to minimize the importance of this dimension for the management of organizations and the credibility of their leaders. But the mode of treatment of the human factor in organizations remains problematic, if only because organizational frontiers are constantly changing. (books.google.com/books?isbn=0415224934). Useful and Useless Internet Sites during Research books.google.com/books?isbn=0415224934 retrieved on March 8, 2007 the reason why I found this site good for my research was that it had a book which helped me with my conclusion au.answers.yahoo.com/question/index?qid=20070108043918AAaYZAy the reason why I fount this website useful was that it helped in understanding the concept www.datawaregames.com/html/cbook7.htm this website was completely useless for me as it was a game website and had nothing to do with my research Work cited books.google.com/books?isbn=0415224934 retrieved on March 8, 2007 Bibliography Ganti, Narsim & William Grayman. The Transition of Legacy Systems to a Distributed Archictecture, John Wiley & Sons, Inc., New York, 1995. Inmon, W.H. Building the Data Warehouse, John Wiley & Sons, Inc., New York, 1992. Inmon, W.H. and Hackathorn, Richard. Using the Data Warehouse, John Wiley & Sons, Inc., New York, 1994. Inmon, W.H. and Kelley, Chuck. Rdb/VMS: Developing the Data Warehouse, QED Publishing Group, Boston, 1993. Kimball, Ralph. The Data Warehouse Toolkit--A Guide to Building Dimensional Warehouses, 1995 Read More