Saturday, December 12, 2009

Open Source Business Intelligence: The Quiet Evolution

In the current economic climate, organizations have to review and rationalize expenses associated with all enterprise software. As a direct consequence, open source business intelligence (BI) is emerging as an important choice for new as well as existing BI implementations. Even though most analyst research indicates that its evolution may have been understated thus far, open source BI is growing rapidly. Open source BI solutions have already been proved to complement and integrate well with traditional BI environments. In their own right, open source BI vendors offer competitive technologies and present the irrefutable advantage of cost savings.

The term "open source software" is often assumed to mean "free access to source code." However, the scope of open source software has widened considerably. The open source software license (often referred to as copyleft) is subject to regulations defined by the Open Source Initiative (OSI)—which dictates that open source software cannot discriminate against groups or technology and must be free to distribute.

Such software can be modified, but redistribution of modified licenses may be conditional (for instance, the license may require that changes be patch files rather than be integrated with the original code) to protect the integrity of the original author's work. The best-known open source package is the LAMP solution stack, which is comprised of the Linux operating system, the Apache hypertext transfer protocol (HTTP) Server, the MySQL database management system, and programming languages including PHP, Perl, and Python.

Open Source and Business Intelligence: The Common Thread

"Open source applications" is the term that describes systems built using open source software in the form of frameworks or libraries. Although copyleft licenses do not permit organizations to resell software developed using open software, mechanisms such as dual-license models have arisen, whereby commercial vendors can deliver their software under a community license that follows the open source license regulations and offers a commercial license with an attached fee. Vendors may charge users for services such as support, training, consulting, and advanced features.

In the past two years, commercial open source vendors have been working actively towards establishing a long-term position in the enterprise applications space. In February 2007, the Open Solutions Alliance (OSA) was formed to bring together commercial open source software businesses; its main purpose is to broaden the horizon of open source applications and most importantly, foster interoperability between them. JasperSoft, one of the pioneers of open source BI is among the founding members of this alliance. Pentaho, another open source BI vendor, has set itself apart by leading and sponsoring all of its core projects, implementing open industry standards and establishing partnerships with vendors of data warehouse technology, such as InfoBright and ParAccel.

BI has some of the most challenging technology problems among all enterprise software applications. These challenges include the design of very large databases; complex data integration between disparate and multiple data sources; the ability to search across a surfeit of information; and some of the most stringent performance and latency requirements. Even with proprietary solutions, organizations need a team of experienced professionals—including database administrators, business analysts, and programmers—to implement and support a data warehouse and BI environment.

Open source BI goes one step further: it encourages organizations to use and modify the software as needed and share advances with the rest of the community. It seems only natural that open source and BI technologies have converged. A crucial factor to consider when adopting an open source BI solution is that underlying technologies are often, if not always, open source themselves; although not mandatory, it is prudent to have technical teams acquire the necessary skills. For instance, most open source BI software is built on the LAMP stack. In order to adopt and maintain the applications, technical teams need to have development and administration skills using the LAMP stack.

No comments:

Post a Comment