Buy custom Data Warehouse essay
A data warehouse (DW) is a subject-oriented, integrated, unchanged, and supporting the data set chronology as well as decision making warehouse. According to Ponniah (2011), there are two different functions incorporated in this definition: a) collecting, organizing, and preparing data for the analysis in the continuously scalable database form; b) the analysis is a part of the decision-making.
The basic data warehouse’s principle is very important, because once listed in the data, it can be repeatedly extracted from it and used for analysis. Thus, one of the major advantages of using DW in the company is control over the critical information from various sources, such as a productive resource (Ponniah, 2011, p. 147).
Data Warehouse in Business Setting
From the business perspective, the most vulnerable point of DW in the enterprise is the correctness of data retrieved from the different sources. The data prior loaded in DW must be “free from the noise” or must be processed by the fuzzy logic methods, which allows the existence of the contradictory evidence. For example, information about the company’s partner may be obtained from the various experts, whose estimates are diametrically opposed.
Also, the integration in the data warehouse’s definition does not mean the information’s integration from all sources of the functional activity in the enterprise; it is a consistent view of data from the different sources according to their type, dimension, and meaningful description. It is the integration of data from the business processes and not of the business processes. The business processes are integrated within the corporate information system (CIS) of the company.
According to Kimball (2011), the information technology used on the DW basis involves objective approach in its organization. Data warehouse is created to address specific, well-defined tasks of data analysis. A range of tasks can be expanded over time, but a defining moment in the DW building is the problem of data analyzing, which must be solved to achieve the goals of the business.
The main objectives of DW are to make the main business management data accessible in a standardized form and suitable for the analysis and to obtain the necessary reports. Data warehouse creation involves: extracting data from the existing internal and external machine-readable sources to achieve it; identifying the purpose of the DW by listing the tasks that should be addressed in the management process and take into account the data which is necessary to address them; creating a working group of the project and include in its membership all the specialists from the different areas of DW; creating an initial sample of data from all sources, analyzing it, and determining what problems and issues of decision support systems can be solved in the data space; standardizing the presentation of data to analyze and check them for completeness and inconsistency; determining the extraction procedure from their sources; carry out modeling; circuiting design; and building a DW (Kimball, 2011, p. 211).
The following aspects should be taken into consideration when creating the data warehouse:
1) If your organization is engaged in the production and distribution of finished goods, then the main tasks include the investigation of the relevant market segments, sales analysis, optimization of supply chain components of the composite products, etc.
2) A created working group must be headed by the project manager or his/her deputy. In addition to specialists in computer technology, the group should include the database administrator and the senior specialists of the main DW objectives.
3) The main issue for the working group to resolve is the question of who should be invited for consultations. It may be independent consultants and experts, but usually, it is the representatives of the company whose software products are used in the organization. At this stage (to reduce the overruns) members of the working group should be explained that the financing of the project is incremental. While the purpose of the stage is not reached, no additional funds will be released.
4) At this stage of the DW implementation, all the major organizational and technological issues must be addressed, which include : selection of the basic information technology (for example, a network solution in the Intranet), choosing of the base software, development of curriculum and staff training, preparation of time schedule of the project, etc. Furthermore, individuals responsible for the development of linguistic software system should be identified. Their task will be to create a normative reference database as well as data dictionaries and thesauri management.
5) An initial sample of all the machine-readable data sources should be created. In practice, too much of the necessary data is circulating in the information systems company, which hinders solving of the problem of analysis as quickly as it is wanted, and critical information is not available at the right time. Therefore, this phase is one of the most important points in the development of DW plan (Kimball, 2011, p. 236).
At this stage, it is crucial to:
1. Identify all the data that will be stored in a DW and by the subject area and fix all the sources of these data in the information system. Such objects may be selling, purchasing, personnel, etc. It is important to determine a set of objects subjected to analysis, based on the control problems.
2. Standardize the data for the DW. It means to bring all the data used to a single representation in DW. This will be the first iteration in the metadata's data warehouse creation.
3. Estimate the rate of updating data in the DW. Since the data will come from internal or external sources and will be transformed and monitored for consistency with developed special procedures, it is important to know the time of these procedures for all the data sets. There will be a relative lag time of the DW integrated snapshot data. For example, the bills are paid during the working day and recorded in the corresponding system. The data of the bills' payment received for the processing determines how realistic the picture of financial flows in the enterprise is. It may happen that the current situation involves the money to be prepaid on a large enterprise account, but in fact, there is a separation from other accounts that have not yet been recorded in the DW. To fix the financial flows, it is better to choose more stable object – a daily balance.
4. Document the type, size, description, and the name of all DW object attributes in the metadata.
5. Implement software tools and procedures for conversion of data for loading into the data warehouse. At this stage, the data warehouse's prototype will be received.
At the stage of standardization and data cleansing, it is necessary:
- to test and evaluate all the data sources for all the data elements to establish the actual content of these elements and determine their standard form for the enterprise;
- to standardize reporting across the enterprise;
- to standardize the data for all the major data warehouse's facilities;
- to standardize the event, the cryptographic codes, and the state of objects in the data warehouse;
- to implement the adopted standards in the data warehouse’s software.
According to Ullrey (2007), a DW database can be non relational. In principle, for the typical data warehouse’s “star” scheme, the implementation of any database type can be used. It all depends on the prevailing stereotype of the organization’s information and financial capacity of the project. The following tasks should be kept in mind:
- Developing and loading the data dictionary of common DW elements (metadata) and reference database. Here, the actions to standardize the information in the normative database should be elaborated. Anyway, something similar will have to be developed, because any standardization requires standards and reference information.
- The development or adaptation of standard software tools for data preprocessing (data cleaning procedures).
- Downloading data to SQL-server or any other server.
- Evaluation of progress and technology’s implementation (Ullrey, 2007, p. 78).
However, how the invested means in the creation of DW will justify themselves depends on the individuals who implement DW.
Data Warehouse in Decision Support
According to Taniar (2009), a decision support data warehouse system enables skillful workers to create and make their tactical and strategic business decisions and business support based on factual data that is stored in the DW business intelligence. Decision support systems should enable users to manipulate data flexibly, quickly, and intuitively using customary terms to provide analytical understanding.
Decision support DW are implementations of an informational database used to keep common data that is sourced from database-of-record in the operational system. It is a subject database that enables users to detect the enterprise’s store of operational data to control different business trends, contribute forecasting and other data mining, react to data models, facilitate planning and strategic efforts, and measure productivity against the tactical aims; all with the aim of supporting and aiding in decision making. The ultimate aim of every decision support DW system is to improve corporate productivity by providing different decision makers with quality and realistic information (Taniar, 2009, p. 195).
According to Martin (2008), MicroStrategy business analytics software reaches this vision by supplying a combined business intelligence platform in order to provide easy using and self-service. This platform supports all the users in the company – from the best educated analyst to the newcomer with only a single web interface. Users can create their own reports, analyses, dashboards, and scorecards with minimum technology information involvement. Also, MicroStrategy offers a highly measurable architecture, which includes full sixty four-bit support, multithreading, and scale-out.
Access to company’s data includes many operational databases: relational, cube, flat files, and other different databases. Such engines produce highly optimized and specific database MDX and SQL in order to resolve any issue with the highest performance (Martin, 2008, p. 263).
Failures in Data Warehouse System
Some problems may occur during the data warehouse creation. For example, it was already decided to make a data warehouse; the tasks and goals are set. What can stop it (if not to take into account the lack of money)? As it is known, a good venture can be spoiled with “good” people. In the creation of any long-term intensive systems, the human factors could be the major aspects to consider. When creating successful DW projects, only a half of funds will be spent on hardware and software, while the other half will be used for counseling and training. Thus, the information system with DW will cost twice more than similar information systems without it.
According to Vercellis (2011), disruptive failures can be caused not only by decision-makers, but persons performing the project may also slow the process of solving the problem. As experience shows, it is possible to avoid failures and time delays in the development of information systems, especially when creating a DW.
Important Moments in Data Warehouse Creation
Uniformity of Technological Concepts
When one creates an EIS (Executive Information System) technology, he/she uses an Internet / Intranet – the so-called Groupwise or its reasonable combination. Regardless of the DBMS (Database Management System) choice, the technology is formed under these different corporate data processing technologies, differently distributed load on the server and client side, and so on. Uniformity of approach provides a rhythm of work.
Organizational and Technological Factors
On the one hand, the unit can be not ready to work on a new information technology and may refuse to accept innovations in a number of objective and subjective reasons (this phenomenon is also common). On the other hand, the underestimation of the organizational and technological support development can lead to permanent deferral of the project and, sometimes, even to its disruption. The staff's preparedness degree at all levels of the hierarchy should be monitored at all development stages and system implementation.
The Importance of Linguistic Support
This is especially critical for the DW construction. Usually, during the creation of a system, it is assumed that everything is contextually clear for the customer and that all the business procedures do not cause any troubles. For example, the development of accounting software built using special computer languages suggests that the accountant should have a good enough computer science and programming base as a part of a university course. In the successful DW usage, a great role plays metadata, on which base the user gets access to the data. The semantics and meaning of all the DW data must be clearly and precisely defined. In practice, if it is not the dominant purpose of the system, attention will not be paid to the development of linguistic software, attributing it to the next. DW construction should begin with the creation of linguistic support.
DW is created mostly for problems solving and decision support. Thus, if you are not sure what information is needed to solve the problem, it is better to wait until your competitors and partners will get their own DW and then borrow their experience.
Market Information Services’ Factor
Judging by the trends, in the near future, the information services market will appear ready for using DW, accumulating data external to company’s sources. Some of the consulting firms have already rendered services such as recommendations and analyzes, so why not to give them a specialized form of DW, which can be incorporated into a KIS or used independently?
It is unreasonable not to combine the compatible. For example, transactional and analytical parts within a single development technology should not be combined in programs like “Trading house”. It will be a long and difficult process, because the approaches to the modeling and design of these parts are very different. Implementing them separately is much easier and faster. However, there are other examples where the concept is used to develop a higher level of abstraction. For example, solutions in Baan or ROSS Systems are successful. The development of a methodology has been worked out by merger transaction and analytical pieces in a single EIS, but they are expensive (Vercellis, 2011, p. 47).
Factors to Consider
According to Wright (2010), by creating a DW in the company, it is important to think about the possible benefits and to act similarly. For example, it is essential to consider the areas where the introduction of DW has already yielded positive results. It will not turn your work in scientific research with an uncertain period of completion. The task list is as follows:
- the segmentation of the market;
- sales planning, forecasting, and management;
- customer care;
- the development of loyalty schemes;
- design and development of new products;
- integration of supply chain;
- incorporation of the intelligent technology in business organization;
- DW distribution of strategic planning for current operations (Wright, 2010, p. 95).
Firms which in recent years support the concept of DW in their products will help to solve these problems. If to talk of the benefits of DW, one should consider the fact that a competitive advantage in the market is becoming harder and harder to achieve and corporate alliances are becoming more and more popular. Moreover, the problems of the enterprises’ automation are often discussed in the press, but the need for publications on this subject does not lose its relevance. This is determined primarily by the trends and needs of the domestic business world that is involved in a mad race (Wright, 2010, p. 156).
There are multiple benefits of implementation data warehouse in decision support system and business. Majority of the companies create their corporate data warehouses to improve their manageability. Quick and appropriate response on changes in the environmental conditions ultimately determines the position taken by the market. The quality of information, the ability to process and analyze it, the speed of decision-making based on the results of this analysis – these are the factors that determine the success or failure of the competition.
In order to adequately assess the situation and work out the optimal solutions, there is must be some way to classify and organize data, filter out irrelevant facts and data available to project a specific business problem. It is important to note that there are significant differences between the concepts of “data” and “information”. Transaction data is significant primarily for operational-level employees. Managers of a higher level are not enough knowledgeable about a particular operation or even a set of business operations. To assess the state of business processes, one needs to operate on qualitative indicators; it means that multi-angle information is needed.