Data warehouse – in short
Data warehouses
Modern companies succeed using knowledge-based management. Data warehouses, which integrate and consolidate the data available throughout the organization are the foundament for business analytics / intelligence solutions. Data warehouses are the starting point for the so-called IT analytics (data science) and creating deep analysis based on them.
summ-it provides comprehensive service – from business analysis through implementation to maintenance of implemented solutions.
Data warehouse – for whom?
Medium company
Big company
Corporation
INDUSTRIES AND SECTORS: retail and e-commerce, finance and fintech, insurance, industrial production, public sector, health care.
The data warehouse is an ideal solution for an organization with extensive data resources distracted between different systems, departments or even geographical location.
Data warehouse – benefits
- Consistent and accelerated reporting – also online
- Universal data source for all departments
- The right business decisions
- Increased competitive advantage
- Reduced costs
If you know what benefits are crucial for you – you are ready to start the implementation process!
Data warehouse – solution architecture and functionality
Distributed sources supply data warehouses using the ETL solution: Extract, Transform and Load. His task is to make them coherent. Due to the amount of information, based on the central warehouse, thematic warehouses are built (data marts) containing data that describe narrow areas of the company’s operation.
Data warehouses allow:
- ON-LINE ANALYTICAL PROCESSING (OLAP):
analytical processing of data through queries, preparing statistical summaries, charts and reports, summarizing significant amounts of data; as part of OLAP tasks, the needs of management and administrative staff as well as analysts are met - ON-LINE TRANSACTIONAL PROCESSING (OLTP):
information processing on production databases, optimized for high performance and availability - DECISION SUPPORT (DS) / BUSINESS INTELLIGENCE (BI):
complex analysis, buliding tactical and conditional scenarios supporting business decisions, predicting future business analyzes - DATA MINING (DI):
building knowledge resources based on data; automatic data summary, extraction of key information from the possessed data, discovering patterns and models in raw source data
- CORPORATE INFORMATION FACTORY (CIF):
centralization of data from various systems and sources available within the organization, guaranteeing data durability, stability and constancy - BACKUP:
centralized data archive with fast access
- KNOWLEDGE DISCOVERY IN DATABASES (KDD):
Iterative process focused on discovering and building knowledge resources. It consists of:
- DATA CLEANING – data cleaning defined as the removal of “information noise” and irrelevant data
- DATA INTEGRATION – integration of data from many sources for the purposes of analysis
- DATA SELECTION – selection of data relevant for the analysis
- DATA TRANSFORMATION – the process of transforming data into the appropriate form required by the search procedure
- DATA MINING – extracting patterns and models based on the analyzed data
- PATTERN EVALUATION – identification of significance of patterns and models built based on data
- VISUALISATION TOOLS – tools for synthetic and graphic presentation of the results of the whole proces
Data warehouse – description and structure of the solution
Heterogeneity and distributed IT systems are common challenges facing organizations. In such cases, access to data is inefficient, and the picture of the situation built of fragmentary materials – incomplete or incorrect.
The data warehouse not only aggregates, but also integrates data from different systems according to one structure and model. Based on this, systems are created to provide management with knowledge in the form of intuitive tools for preparing analyzes and reports. However, they provide the basis for making accurate decisions.
Data Warehouse:
- It’s a large database containing up to hundreds of terabytes of information
- It integrates both historical and current data
- It is designed to handle queries and perform analysis
- Provides users with data only in read mode
- Contains aggregated data on many levels
Data warehouse – solutions and services:
- Business analysis – scope and business goals for the data warehouse
- Data warehouse design – incl. taking into account the planned use of the warehouse, the size and nature of the data held, the type of transformation required and the frequency of data refresh
- Implementations of OLTP and OLAP systems (multidimensional and tabular)
- Implementation of on-premises and public cloud data warehouses (cloud computing)
- Designing tools to create reports with visualizations for processed information
summ-it will create optimal assumptions for your data warehouse thanks to experience and numerous implementations.
Data warehouse – 3 steps to implement
Determining the purpose of building a data warehouse
Analysis of the organization structure and available data sources
Creating a business case – implementation treated as an investment and KPI definition