|
|
|
|
Architecture Review/Recommendation, Proof of Concept
Data warehouse architecture and proof of concept maximize value of customer data
Moviefone, a service of AOL/Time Warner, is the nation's largest movie listing guide and ticketing service, providing interactive telephone information, advertising and teleticketing services to the motion picture industry and moviegoers. Through its 777-FILM phone line and web outlets, Moviefone provides moviegoers with listings for more than 13,000 movie screens in over 30 markets nationwide. In a single week the company has handled as many as 3.7 million calls, and handles more than 300 million calls a year.
Business Challenge
Moviefone was looking to maximize the value of its customer data. With the enormous volume of data being collected, it needed an easy way to maintain and access it for both internal and external marketing and sales initiatives. Irwin Miller, MovieFone's vice president of computer systems development explained, "Our problem was being able to maintain the volume of data we collected, and being able to query it quickly and generate a variety of reports. We needed to be able to take the information from our PC X-base format, normalize it and load it into something that could be queried by SQL."
CSI Solution
Based on the volume of data -and the types of queries it needed to run - Moviefone was interested in a customer information data mart. MovieFone selected CSI to help, plan, test and implement this project.
Because of the complex nature of their inquiries and the sheer amount of data involved, CSI determined that that Sybase Adaptive Server IQ was the ideal solution. To verify this, CSI loaded a couple million rows of data into a traditional RDBMS and into Adaptive Server IQ and ran a number of ad hoc queries against both. The results were dramatic. IQ outperformed the RDBMS by several orders of magnitude.
There was another consideration in choosing IQ. A year's worth of MovieFone data added up to about 40 GB. By the time this amount of data was loaded into a traditional relational database and indexed, the total disk space required would nearly double. IQ, on the other hand, with its bitwise technology and compression capability, was able to load the entire 40 GB of data, plus indexes for every field, in just 40 GB. Rather than requiring in-house IS expertise to maintain a traditional RDBMS, indexing the fields only required weekly loads. That meant that the data mart was essentially maintenance-free. CSI trained Moviefone users on IQ and turned over an easy-to-use interface to the data mart, allowing users to create their own reports.
CSI also helped MovieFone select the hardware on which IQ would run. They chose a Dell NT box with two processors and the ability to expand to four, and a 60 gigabyte hard drive to start. The goal was to build a system that would allow growth over the next five years by adding processors and disk space. In addition, Business Objects was chosen as the business intelligence tool to enable MovieFone staff to do their own ad hoc queries and generate their own reports from their desktops.
Return on Investment
From start to finish, the development and deployment of the turnkey data mart was completed in less than 120 days, considerable quicker than the industry average of six to twelve months. MovieFone users can now create their own reports and answer their own questions from their desktops."
|
|
| Strategic Consulting:
Data Warehousing:
Business Intelligence:
Data Management:
|
|
|
