Data Warehouse

Data Warehousing

Corporate operational data is not readily usable for strategic decision making. These large quantities of operational data are very useful for running business operations, but hardly suitable for making decisions about business strategies and objectives. A data warehouse is an environment where executives can find strategic information to make better decisions. InfoTelica has the expertise and competence to build robust decision support environments from massive amounts of corporate operational data.

ETLDatamodellingAnalyticsReports &dashboardsKPIs &scorecardsCollaborativeBIMobileBIBI & DWBI & DWETLExtract, clean and loaddata from every sourceData modellingConceptual, logical andphysical modelsAnalyticsOLAP analysis acrossevery business dimensionReports & dashboardsAd-hoc reports andvisual dashboardsKPIs & scorecardsTrack performanceagainst your targetsCollaborative BIShare insightsand decide togetherMobile BIDashboards and alertson any device
Services

Data Warehouse Services

  • Project Management.
  • Business Requirement Analysis.
  • Source System Analysis.
  • Data Warehouse Architecture.
  • Enterprise Data Warehouse and Data Mart Modeling.
  • Source-to-Target Mapping.
  • ETL Development.
  • Inconsistent Source Data Processing.
  • Query and Report Generation.
  • Data Warehouse Maintenance.
OperationalsystemsERPCRMFlat filesETLExtractTransformLoadMetadataSummary dataRaw dataData warehouseOLAPanalysisReportingDataminingOperationalsystemsERPCRMFlat filesETLExtractTransformLoadMetadataSummary dataRaw dataData warehouseOLAPanalysisReportingDatamining
Approach

How a project runs

1
Project Management

Managing data warehouse projects balanced among scope, time, and budget constraints while assuring high quality in accordance with PMI (Project Management Institute) norms. DWH projects are managed through integration, scope, time, cost, quality, human resources, communication and risk management processes across the initiating, executing, monitoring, controlling and closing phases.

2
Business Requirement Analysis

InfoTelica uses 5 core requirement components (entity, attribute, process, external agent and business rule) for in-depth business requirement analysis by

  • Identifying core data requirements beginning with project initiation.
  • Identifying excellent data requirements at the appropriate level of detail.
  • Detailing the data requirements.
  • Identifying and detailing attributive, associative, subtype and super type entities.
  • Detailing complex data related business rules.
  • Discriminating between business data (logical data) and database design (physical data).
  • Utilizing normalization techniques.
  • Validating data requirements with activity requirements.
3
Source System Analysis

Source system analysis provides the key understandings of the source data by

  • Column & Domain analysis.
  • Table/Primary Key analysis.
  • Foreign Key analysis.
  • Cross-Domain analysis.
4
Data Warehouse Architecture

Building data warehouse architecture in accordance with customer’s requirements and source system analysis.

Modelling

Enterprise Data Warehouse and Data Mart Modeling

Data modeling methodology has three phases: Conceptual, Logical, and Physical Data Models. A Conceptual Data Model is a prerequisite for a Logical Data Model, which is a prerequisite for a Physical Data Model. The physical database and table structures are based directly on the data structures outlined in the Physical Data Model.

Conceptual Data Model

How will the enterprise be organized within the data warehouse? The Conceptual Data Model defines the subject areas and major entities.

Logical Data Model

Which entities exist within each subject area? How will those entities relate to each other? The relationships extend beyond just primary key/foreign key relationships. They include Super Type, Subtype, Attribute, and Associative Entities.

Physical Data Model

Databases hold the data of a data warehouse. How will that data be assigned to columns, which are inside tables that are organized into databases that collectively hold the data of a data warehouse? Though not quite the Data Definition Language (DDL) of the tables and databases, the Physical Data Model is very close to the DDL of the tables and databases.

Build and run

From mapping to maintenance

5
Source-to-Target Mapping

After designing target database logical and physical architecture in accordance with business requirements, we determine where and how the source data for each target table column will be obtained.

6
ETL Development

ETL stands for Extract-Transform-Load. In our approach, ETL has two phases: the Relational Data Store (RDS) and the Dimensional Data Store (DDS). The RDS phase transfers source data to a staging area. The DDS phase builds data marts according to reporting requirements. Data marts hold de-normalized, aggregated, historical business data that is used for reporting purposes.

7
Inconsistent Source Data Processing

Building mechanisms to handle inconsistent source data during the RDS phase of ETL. Some examples of inconsistent data are duplicate source data, source data without primary key values, data with wrong data type and so on.

8
Query and Report Generation

Building reports for companies’ decision support systems using the data marts’ denormalized, historical data prepared for reporting purposes in accordance with business requirements.

9
Data Warehouse Maintenance

Speeding up prepared reports and ETL processing by tuning the data warehouse OLAP database, source OLTP databases, ETL packages and report queries.

Services

Other services

DBA ServicesData ServicesModern Data PlatformsPrivate & Generative AIInfoTelica AI TowerBusiness IntelligenceBig DataArtificial IntelligenceSoftware DevelopmentOutsourcingIT ServicesAll services

Not sure where to start?

Tell us about your systems and goals. We will suggest the right mix of services and a first step.

Talk to an architect