Design of data marts is a fundamental task when preparing data destined for implementation of a Decision Support System (DSS). To answer questions on the underlying users information needs, a data-mart designer is challenged to distill the relevant information from various data sources, enable in-depth data analysis and provide ease of access of information to targeted DSS users. This paper presents the design of data marts for analysis of livestock datasets using dimensional modelling techniques. The designed data marts are then implemented for supporting information and knowledge extraction, leveraging large quantities of livestock data from livestock data source that was identified in the case study environment. Appropriate data mart dimensions and facts were modeled in order to ease data analysis and queries in a role-based information decision support model that was adopted in studied context. Data models based on dimensional modelling (star schema model) are provided and discussed. A comprehensive example showing how a piece of data is loaded from livestock data repository to ‘fact’ and ‘dimension’ tables using an open source CloverETL Designer tool is also given. The paper concludes with an overview of the overall detailed schema for the livestock data mart that will serve as a backend engine for On-Line Analytical Processing (OLAP) analysis, reporting, data visualization and information querying via mobile and web access. It is anticipated that the overall DSS once implemented, can be used for improving information delivery, sharing and decision making process to smallholder livestock keepers and livestock experts in Tanzania.