Data warehouses, what are they
The development of information systems and the web has created a vast amount of data assets. The use of these information assets of the company is done by using Data Warehouse technologies.
Data Warehouse (DWH) is a database that stores all information regarding the company’s activities and it is used by the management to make strategic decisions.
With a data warehouse, the company can store all company’s data in a single place, in fact, data is normally located on different information sources. Furthermore, this data warehouse is well organized according to its dimensions to be easily accessible by users at different levels of the company’s hierarchy.
From this point of view a DWH facilitates a “knowledge creation process”: starting from the organization’s scattered data (which is both in the operational environment and in external data sources), progressing through a series of steps, and with the adoption of appropriate data processing and transformation tools, the company can obtain useful information for the user, which can be easily consulted and, therefore, represents the basis of company knowledge.
Data warehouse, how is it used in the company?
The fundamental objectives of data warehouse are:
- making company information easily accessible: the content of the data warehouse must be understandable, intuitive and obvious for the business user, not just for the developer. Content must be tagged meaningfully so that users who wish to separate and combine data in almost infinite combinations have minimal waiting times;
- presenting company information consistently: data must be carefully assembled from a wide range of sources around the company, it must be cleaned, meaning that its quality must be guaranteed and it must be distributed only when it is suitable for consumption by the user;
- being adaptable and flexible to changes, to user needs, to business conditions: data and technology change over time. Data warehouse must be designed to handle these unavoidable changes;
- being a safe bastion to protect information: a data warehouse must effectively control the access to company confidential information;
- serving as a basis for making better decisions: a data warehouse has only one true output: the decisions made after the data warehouse has presented its evaluations;
- sharing and having the company accept the data warehouse. Unlike rewriting an operational system, where business users must use the new system, data warehouse usage can be optional. The acceptance of business users is based mainly on simplicity.
Architecture and design of a data warehouse
Data warehouse life cycle begins with the project planning. In this phase the company degree of readiness is evaluated for a data warehouse initiative, a primary objective and a justification are established, resources are obtained and the project is launched.
The second main operation focuses on defining business needs. The alignment of the data warehouse with company needs is absolutely essential. The most advanced technologies will not recover a data warehouse which does not focus on the company. Data warehouse designers need to understand the needs of the company and translate them into design considerations.
The design of the technical architecture establishes the general structure which supports the integration of multiple technologies. Using the features identified in the architectural design as a shopping list, it is possible to proceed with the evaluation and selection of specific products.
The central part of the life cycle focuses on data. It starts by translating the needs into a dimensional model, which is then transformed into a physical structure. Focus is on performance tuning strategies, such as aggregation, indexing, and subdivision, during physical design activities. Last, but not least, the extraction-transformation-loading (ETL) processes of data staging are designed and developed.
The final group of operations generated by the definition of business needs consists of designing and developing analytical applications. The data warehouse project does not end when data is produced. Technology, data and traces of analytical applications are brought together with a healthy dose of training and support for a well-organized distribution. After that, continuous maintenance will be necessary in order to guarantee that the data warehouse meets all the needs that time and a greater knowledge of data can give. Finally, the future growth of the data warehouse is managed, pulling the trigger to consecutive projects, each of which returns to the beginning of the life cycle.
Differences between a data warehouse and a database
The main difference between an operational database and a data warehouse is the type of queries. In databases, queries execute transactions that typically read and write a reduced number of records from several tables linked by simple relationships. This type of processing is commonly called On-Line Transactional Processing (OLTP). On the contrary, the type of processing for which data warehouses are created is called On-Line Analitycal Processing (OLAP) and it is characterized by a dynamic and multidimensional analysis that requires the scanning of a huge amount of records to calculate a set of synthetic numerical data that quantify company performance.
The unique characteristics of OLAP queries shall ensure that data in the data warehouse is normally represented in a multidimensional form. The basic idea is to see data as points in a space whose dimensions correspond to as many possible dimensions of analysis where each point, representative of an event occurred in the company, is described through a set of measures of interest for decision making.
What are the advantages that companies and public administrations can obtain using data warehouses?
Companies need to be “agile” and quick to respond to changes in a changing business environment, so they can increase their competitive advantage. To be successful in the market, companies need information that enables them to understand the facts, the ability to make decisions quickly and the experience to act without problems. By improving business agility through the adoption of a robust data warehouse, companies get three specific benefits:
- a better view;
- a greater efficiency in complex situations;
- the opportunity to be innovative.
The ability of a company to perceive changes and respond to them effectively and efficiently makes their company agile. All investments and all business processes adopted must contribute to improving business agility. The highest degree of business agility is reached with knowledge, visibility and solutions that enable a company to process the data in its possession faster, in order to effectively respond – producing profit – to unexpected changes. Companies, therefore, become able to manage and overcome an ever-increasing complexity, anticipating the knowledge of market trends, generating profits and pursuing new opportunities, all this while their own competitors are still trying to understand what is happening.
How does Consulthink help companies that want to implement data warehouse solutions?
Consulthink is able to support its customers in all phases of a data warehouse project, thanks to its highly specialized team in Data Warehouse and Business Intelligence projects. Experience gained in the field, working for major customers, public and private. Our professionals are at your service, offering you the data warehouse model that best meets the needs of your company and offering you, over time, a maintenance service that can always better analyze data, allowing you to respond quickly and better than your competitors, to market trends, permitting you to benefit from new opportunities.