Business intelligence is a vital component for successful business management. It introduces capabilities for effective decision-making, resulting in higher income and increased growth for the organization. BI programs must keep strategic goals and organizational missions in mind, while reducing the cost and time of implementing solutions.

Open source BI may be evaluated against the parameters of total cost of ownership, performance, scalability and user requirements.

Why is Business Intelligence important to an Organization?

Organizations can make intelligent decisions when timely information is consistently made visible to decision-makers at all levels, as this endows them with the ability to monitor important drivers of organizational performance. A well-designed BI system collects the organization’s operational data from different sources, presenting it to decision makers and stakeholders simply and meaningfully through use of a user-friendly tool. A good BI solution helps organizations gain better insight into their businesses, improve decision-making and optimize enterprise performance.

An Open Source BI Overview

Open source BI has come a long way compared to other commercial BI products, and is becoming widely recognized as an important component for enterprise-level applications.  Open source BI projects such as Pentaho and Jasper have evolved from community-driven tools to viable technology with professional support for enterprise-wide adoption and witness growing demand. Organizations can use open source BI software to replace custom-coded applications. The open source BI tools can also be considered for BI components that complement the existing proprietary solution to reduce license cost. Because organizations are not locked into proprietary vendor’s platforms, open source enlarges organizational flexibility.                                            

TCO:  Critical Factor in Implementing BI Solutions

While few will deny the importance of BI, the most important factor to be considered for BI is the total cost of owning the application. The TCO concept measures costs related to the acquisition of a BI solution, its deployment and ongoing use. Though TCO estimation methods vary for different BI implementations based on requirements and business needs, certain proportions may be assumed to calculate the TCO in most projects. For example, typically staffing costs account for 50 percent while the hardware costs account for 8 percent of the project value. Based on market trend reports published by leading industry analyst organizations, the cost breakdown  in Figure 1 may be assumed as the TCO breakdown for most BI implementations.

Open source helps reduce TCO on all the parameters in Figure 1. Open source BI helps in reducing costs and risks for prospective BI users.  Though this does not suggest that open source BI is the right choice for every organization in every BI deployment, it can be used as an alternative for reducing BI costs if it satisfies user requirements.

Major TCO Components

Hardware: This covers the cost incurred in procuring the hardware throughout the organization, including all client machines, servers, storage solutions and networking devices attached to servers. As most software licenses are based on the number of CPUs, it directly impacts the cost of hardware. Using a scale-out approach, low-cost servers can be used to deliver open source BI solutions.

Software: The cost related to the software is one of the significant factors in the overall TCO. Open source BI is available at a fraction of cost as compared to commercial products. Open source BI customers have the flexibility to choose the components and their support level according to the requirements of the end users.

Staffing: Staffing constitutes 40 percent to 50 percent of the BI application’s cost, including the cost of resources during the analysis, development and maintenance phases.  The ability of the vendor to provide the documentation and technology of expert resources makes a big impact on the TCO. Open source is based on public standards and public domain technologies.

Selection Criteria for Open Source BI Solution

Though TCO is the commonly accepted financial measure for evaluating the BI solution, factors such as user requirements, complexity of development and scalability of the solution have to be analyzed to perform the TCO calculation. The five factors that affect a BI solution are:

  • BI product selection and user requirements,
  • Complexity of development,
  • BI project timelines,
  • Product support and third party support and
  • Performance and scalability.

These points can be used to compare the open source BI solution with proprietary vendor’s solutions.
BI product selection and user requirements. The objective of collecting BI user requirements is to establish the outcome of the BI solution and other aspects of the projects relating to time, cost and resources. In the case of open source BI solutions, organizations can verify the requirements without contacting the product company, because organizations can initiate a proof of concept and refine the requirements without buying the BI tools.

Complexity of development. Developing a BI solution is not only dependent on the user requirements but also on the product features and technology. As compared to proprietary vendors, open source BI products are based on the technologies available in the public domain. The resources for developing and maintaining applications are easily available. Most open source BI solutions allow a design approach in which a prototype can be done rapidly with regular testing and feedback from the BI users.

BI project timelines. Any BI solution requires orchestrated efforts by the team to complete the solution on time. Selection of proprietary and open source technology affects the human cost. While considering the open source tool, organizations must consider developers and supporting people such as database administrators and testers to understand and learn the technology. Open source BI products have simplified the use of tools and added features that can reduce the development timelines.

Product support and third party support. All open source companies provide support for the products at very low subscription prices compared to proprietary vendors. A systems integration partner is usually brought in to support the solution.

Performance and Scalability

The performance of the BI solution is dependent on factors such as data source performance, server hardware, content complexity and user requests. Most open source BI solutions support scale-up and scale-out architectures and can scale linearly.

Integration with existing infrastructure. Open source BI solutions provide a comprehensive integration interface wherein customization can integrate with the existing infrastructure. Also, this solution can be embedded on compliant servers. Information like cubes and report can be integrated through XML, HTML or JSR-168 portlets. Open source solutions are compatible with multiple operating systems.

End users and supporting personnel training. End users are business people who understand business terms. Open source solutions have the capability to put up a semantic layer that hides the complexity of the data and allows end users to exploit information using business metadata. It removes the necessity for end users to learn the coding language or syntax related to products. System integrators or open source BI companies can provide training to the support team when the BI solution moves into production.

A Complete BI Solution with Open Source

A typical BI solution has the following architecture: Source systems are the starting point for data. A second layer consists of data transfer to the data warehouse in a structured format. This is commonly referred to as extract, transform and load. The third layer is a data warehouse, where all the data is stored for analysis in normalized or denormalized format. The fourth layer consists of different applications (like data mining, OLAP and reporting engines) talking to the data warehouse. The final layer is a presentation layer. Information is utilized by end users through client applications, portals and browsers. 

Open Source ETL. ETL is one of the important pillars of the BI solution. Open source is a good option for customers who want to avoid custom codes and reduce the high cost of proprietary ETL products. Open source projects have recently added an option for on-demand ETL solutions. Another advantage with open source ETL is that organizations can develop their own customized plugins and extend the capability of ETL tools.

Reporting and dashboards. Open source BI reporting tools can be used as standalone products or can be embedded into the existing BI solution. Reporting engines directly access relational, OLAP and XML data sources, and reports can be delivered in PDF, HTML, XML, Excel and text format. Ad hoc reporting capabilities through Web browsers is also available in some tools. Dashboards provided by open source BI vendors integrate reporting components and exercise improved control on the content through the use of JSP.

OLAP Analysis 

Multidimensional analysis turns data into highly explorable structures of cubes, and OLAP can provide quick answers to business questions. Mondrian is a leading open source tool for OLAP that allows users to analysis and navigates the cubes. Mondrian connects to a JDBC data source and does multidimensional analysis. OLAP tools provide optimized response time even for large data sets and support MDX language. Users can slice and dice data to explore trends, patterns and correlations.

Open source BI solutions effectively promote an enterprise system with low TCO, high reliability and high-performance. Open source software can reduce initial and operating costs  and can customize systems to meet user requirements.  Open source tools can be compared with traditional BI products by using the basic parameters of software, hardware and staffing costs. The functions, performance, reliability and security of open source tools are continuously evolving and contrast with the high acquisition, integration and lock-in costs of traditional BI solutions. 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access