Data Warehouses - Essay Example

Question One of the problems in setting up a data warehouse is the sheer volumes of data that must be accomo d and the challenge of finding a way to effectively sort and store the data. The underlying effort is to set up data in such a manner that will be easy to access through the use of a single server, so that it can facilitate an organization’s decision making process(www.esri.com). Data warehouses are useful for creating cash flows and customer reports using meaningful data. Groenfeldt (1998) points out that where consolidation of financial information on derivatives into gigantic data warehouses is concerned, the results have been disastrous, because the operating realities make the process of efficient and optium data storage a messy and complex one. In financial organizations such as banks for example, some of the problems arise because they start to build their data warehouses before figuring out what they want in it. To alleiate this problem, effective preparation is necessary; a specific purpose must be formulated for the data warehouse.(Gronfeldt, 1998).Experts recommend that common denominators be set up for the data, which are accessible to all departments and extensions be created for oother departments to link to. Creating extensions to data tables can make specific relevant informaiotn available to specific departments. An ESRI Report identifies how data warehousing is used in hospitals to enhance Online Analytical processing; including a spatial data model can also facilitate patient profiling and physician profiling.(www.esri.com). Structuring the data warehouse so that it facilitates the aggregation of data and data linking would be helpful in developing suhc profiles. 2. Human beings are able to convert data into information through a process of association using external stimuli as well as internal ones such as memory cues. A similar process occurs in Online Analytical processing of data, where data from different sources are associated or linked together in order to assess, discover and evaluate existing trends wihtin it. Data is associated with additional streams of data available from other sources and a process of elaboration of the exsiting data can be initiated through the process of Online Analytical processing, to generate hypotheses about the data and assess the consequences of those hypotheses.(www.edc.ncl.ac.uk). Alternatively, patterns existing within streams of data can also be evaluated in this manner in order to generate usable information. For example, data on large numbers of individuals may be assessed to discover what common treds exist among them, such as how many are college educated and how many are not. This would be usable information. Knowledge is the outcome of a learning process that links information to its potential use. The ability to utilize learning gathered in a particular context to another situaiton involves the aplicaiton of knowledge. In the case of computers, Online Analytical Processing helps to make necessary associations between data and integrate it accordingly in order to generate secialized information, which is knowledge. In effect, this involves data mining, or digging out associations between data that helps to generate fresh information that is derived from existing data. This constitutes the formation of knowledge. 3. In the case of Online Transaction Processing, information does not need to be stored for a long time and it is superficial kind of information, such as biographic details, like name, age, gender, address, goods purchased, price, order number, etc. As a result, the data warehouse does not need to be very comprehensive, for example an airline company issuing tickets to passengers is unlikely to maintain records of all ticketing transactions that were covered thirty years earlier. Data also tends to be strictly factual, because the purpose it is used for is to issue tickets or process a sales order or generate a financial transaction. Some examples of Online Transaction Processing applications include electronic reservations, electronic transfer of funds, inventory control, telephone switching and centralized command and control services. The kinds of transactions to be carried out in Online Transaction applications will very rarely involve any kind of analysis or deeper study of the data, the focus is on accessing data quickly, hence correlations or patterns in data will not be examined. But in the case of Online Analytical Processing, data available may be exhaustive and compiled from a variety of sources. Since this process is akin to business intelligence, it will focus upon collecting data from a variety of sources. Such data will be correlated and linked, with associations, trends and patterns discernable in the data being examined. Hence data in this instance will not be superficial, neither will it be fairly recent, it will be as comprehensive as possible and will be targeted at deriving as much knowledge as possible out of it, so that it can be used as the basis for intelligent deciison making by managers and other executives of an organization (www.databeacon.com) 4. The kind of data warehouse that a knowledge worker is most likely to use is the kind that is conducive to Online Analytical Processing. This is because a knowlege worker’s job is complex and requires thinking, rather than merely performing simple administrative jobs by accessing and processing data. An individual knowledge worker may not be involved in a job like generating an airline ticket for example, but may need to examine information from a variety of sources in order to derive inferences and conclusions from the data. Hence, a knowledge worker may need a data warehouse that is much more exhaustive and contains data that is well organized through the use of a common pool of information and the availability of relevant extensions that will provide specific informaiotn on the subject the knowledge worker may be researching. For example, a knowledge worker involved in the pharmaceutical industry will need access to common information that may be available over the Internet and other sources like books and journals, etc on drugs and general medical information. But in the development of a specific drug, the worker will need access to information on relevant experiments and other informaiton specifically relevant to the development of the drug. Similarly, a financial analyst working in a bank may need access to the common pool of information available in the bank as well as general financial information available from other sources. But in addition, he will be benefitted if the data warehouse of his organizaiton provides extensions where he can specifically access information on derivatives and securities, etc so that he can use it as the basis to make complex financial assessments that may be used in advising clients accordingly. Bibliography * Databeacon, 2004. “Why OLAP – An overview of online analytical Processing”, Retrieved October 20, 2008 from: http://www.databeacon.com/pdf/OLAP_White_Paper_2.0.pdf * ESRI White Paper, 1998. “Spatial data warehousing for hospital organizations”, http://www.esri.com/library/whitepapers/pdfs/sdwho.pdf, * Groenfeldt, Tom, 1998. “The data warehouse problem”, http://www.derivativesstrategy.com/magazine/archive/1998/0598fea5.asp, * Transforming data into knowledge. http://www.edc.ncl.ac.uk/highlight/rhmarch2007g01.php/, Read More