Implement an open source BI tool, head to toe

Tip

Implement an open source BI tool, head to toe

There has been a change in paradigm from wanting to use an open source BI tool for non-mission critical purposes. Open source solutions are now becoming serious alternatives to proprietary software with ever increasing open source projects providing a wide variety of tools for data warehousing and full BI suites.

Open source BI tool: Necessary precautions

Before any organization can begin its journey on an open source BI tool it needs to   assess its specific needs that would make open source BI tool a perfect choice. The considerations should include:

  • What is the licensing cost and TCO (total cost of ownership) of a proprietary BI tool vis-à-vis an open source alternative?
  • Does your organization have large data sets that the traditional, commercial BI tools cannot handle?
  • Does the organization have any special needs like processing data on cloud platform and leveraging cloud for BI?

Pre-implementation assessments

  • Today organizations use cloud for large data processing needs.  Skill set availability needs to be kept in mind to leverage cloud and extend the functionalities of the open source BI tool for the various organizational needs.
  • Evaluate your open source database needs and the open source BI tool requirements along with the systems that would be in use as they will have a bearing on the development of the open source BI tool.

The ground reality

Any open source BI tool development and deployment will need to be evaluated in the context of the following three factors:

  • The hardware which is already existent or needs to be added
  • Software (codes) that the open source solution is offering, and the most imperative,
  • The development team for open source BI tool.

The other certain areas that one would need to focus on would be:

  1. Business need for the deployment of an open source BI tool
  2. Time frame under which open source BI tool will be implemented
  3. Budget considerations for development and deployment of open source BI tool. (Remember, open source does not mean free!)
  4. Support costs associated  with open source BI tool
  5. Scalability of the tool
  6. Future expansion in the analytic space that would necessitate integration of the open source BI tool with other applications.

Deployment considerations

The resource needs for any open source BI tool would have to include an IT team with experience in Java, cloud infrastructure, knowledge of distributed systems, and distributed processing.

The implementation of this will go smoothly only if there is a strong base of trained developers that can help build the foundation of the open source BI tool and take it to the next level. Without these there is a high probability to increase the production cost and overall development cost.  

For example, at Yahoo, we’ve developed an in-house BI tool on Hadoop platform. We factored in all these resource needs well in advance to ensure success in the project. Moreover, since the development of open source business intelligence tools does not come with an option of plug and play as many proprietary products, the implementation challenges are also case specific.

BI training for users

Like any BI solution, open source BI tools also call for a similar amount of training. Business users need to be trained as to the difference in the earlier methodology of report generation and the new one.

Payback time

Lastly, based on our experience about return on investment on an open source BI tool project, it may be said that RoI on open source BI tool development (and deployment) can be realized over a period of three years.

The first year is the investment phase and the next two years are spent in monitoring the usage and increasing the adoption. Once you achieve a high level of adoption, you can start realizing the cost saved on licensing fees, special hardware needs, and support cost savings.

 

About the Author: Rohit Chatter is the Data & BI Architect for Yahoo India with extensive experience in BI research & development.

(As told to Sharon D'Souza)

This was first published in March 2011

Disclaimer: Our Tips Exchange is a forum for you to share technical advice and expertise with your peers and to learn from other enterprise IT professionals. TechTarget provides the infrastructure to facilitate this sharing of information. However, we cannot guarantee the accuracy or validity of the material submitted. You agree that your use of the Ask The Expert services and your reliance on any questions, answers, information or other materials received through this Web site is at your own risk.