Terabytes, petabytes and exabytes, terms that weren’t in common use 10 years ago, now seem commonplace as new business processes and technologies create massive amounts of data. Tools like Apache Hadoop—an open source software project that enables multiple computers networked together to run large computations—can process and generate insight from large data sets efficiently and quickly. However, business leaders are challenged with how to manage these large data sets, identify trends, and make data-driven business decisions. To overcome these challenges, this global media and entertainment company engaged Point B to set up and operationalize data capture of its large online ad tracking system. The result? A fully operationalized system and greater insight into online consumer behavior.
Laying the foundation for success
By working with stakeholders across the organization, Point B helped our client clarify and refine their strategy for the online ad tracking system and bring it into greater alignment with the overall goals of the organization. Once defined, we led the client to:
- Define and implement requirements for a production-ready system, including SLAs, uptime requirements, 24x7 support model, and testing thresholds.
- Determine key milestones and integration points between the Operations team, Hadoop engineering team, and the Digital Ads team.
- Implement the Hadoop cluster on the production environment.
- Assist with test plan development, including 100% parallel testing, to ensure a successful system launch.
- Drive communications with internal stakeholders to inform them of the change, state project objectives, anticipated benefits, and
- encourage buy-in and feedback on the project.
Getting big data and big results
With our help, the client transformed a rigid ad tracking and reporting system to a scalable parallel storage and processing architecture that provides historical, current, and projected future data that are always available. The data give them greater insight into consumer behavior and demand. In addition:
- Because the system captures all event data, the client is now able to tie together new data sets for greater reporting/business intelligence capabilities.
- We helped the client make this transition with zero downtime and zero loss to existing data.
- The new system reduces storage costs, scales inexpensively, and enhances capabilities to support data outputs to downstream systems and reports.