|
|
YOUR FEEDBACK
Did you read today's front page stories & breaking news?
SYS-CON.TV SYS-CON.TV WEBCASTS |
POWERBUILDER LINKS YOU MUST CLICK ON Success Story
Well-Informed Business Planning
EMI Music Germany
Mar. 5, 2008 10:00 AM
Digg This!
Page 2 of 2
« previous page
EMI Music: A 70 Year Legacy With its broad organization and far-reaching marketing efforts, EMI Music Germany's decision-makers must be agile and decisive. Being an executive in the music industry means preparing your decisions systematically, monitoring the effects of your decisions directly, and adjusting your strategy rapidly whenever necessary. Contrary to other markets, the music business does not follow a linear pattern but is focused on the time of release of each new recording. Fast, efficient, and flexible business intelligence analysis is of vital importance for all management levels of EMI Music Germany.
Managing and Assessing Performance in Real-time All these activities must not only be planned in detail based on realistic assumptions, but also monitored in real-time while in progress to take full advantage of the narrow time window during which the sales prospects for a new recording are optimal. "Performance assessment at the touch of a button," is how Klaus Giesenow, director of information technology at EMI Music Germany, summarizes the requirements for the business intelligence software and data assets for which he is responsible.
Delivering Multiple Executive Views of Information All departments run segment analysis on current data, as well as long-term trend analysis. Day by day, 50,000 new electronic records of items sold have to be analyzed. Twenty-five million such records are generated each year. Enabling users to analyze this data by defining custom queries that often cover several years allows them to detect trends and patterns while providing crucial information for planning.
Analytics Requirements: High Flexibility + High Performance The rapidly changing market constantly forces EMI to adjust its organizational structure, adapt its product portfolio and sales channels, or rearrange its regions, product groupings and sales infrastructure. "During the last decade we restructured almost once a year," Giesenow reports. The company's BI systems must be flexible enough to support new structures when analyzing data and handle additional users without compromising performance. All this needs to be accomplished using one single, uniform, comprehensive database to ensure meaningful trend analysis. But EMI's enormous, complex data assets are a major challenge in terms of processing speed. "Running complex analysis on a daily basis requires a solution that delivers an appropriate amount of performance and flexibility," Giesenow stresses.
Millions of Data Records The breakthrough came when Sybase IQ was introduced. Today, the information stored in the operational database is replicated to the analytics server in near-real time. "Decision makers can now check the performance of the first half of the day as early as midday," Giesenow explains. "An analysis of 50,000 data records is something Sybase IQ handles on the fly. Thanks to data compression technology that reduces the volume of the operational data by a factor of ten, and thanks to the excellent system performance, we no longer depend on pre-aggregated data. Any query can access our full data assets. This means that historical data from 10 years, comprising 250 million records, is available for ad-hoc, long-term analysis at the touch of a button. Even those kinds of searches take no more than a few seconds." Another important benefit is the scalability of Sybase IQ, Giesenow emphasizes, "What applies to other things in life also goes for BI: The more you have, the more you want. As the data volume increases steadily, so do the analyses we run, both in terms of variety and number. This is why we need technology flexible enough to meet new requirements with ease while keeping our total cost of ownership down." What helps Giesenow achieve these goals is the combination of database software and hardware. Two IBM System p servers running AIX (p560Q) have been configured as a high-availability solution, comprising a High Availability Cluster Multi-Processing (HACMP) cluster with an underlying, fibre channel-based Storage Area Network (SAN). "This configuration, composed of perfectly matched hardware components from a single source, has proved to be exceptionally powerful. In addition, it features an excellent high-availability concept, as well as high storage and CPU scalability," Giesenow emphasizes. "What is more, we have found that as few as two CPUs and a mere 3.4 GB of memory are entirely sufficient for Sybase IQ. These modest hardware requirements contribute significantly to our low TCO."
Individualized Querying To support individualized searches, the front end does not provide any predefined reports. "Instead, there are many options that allow each user to build his or her own view. Users can limit the time frame for a query as desired, combine any number of products or regions, and define multiple criteria in a single query," Giesenow explains. "The SQL code is not generated until the query has been fully defined. All this would not be possible without the exceptional performance of Sybase IQ, which it owes to the indexing principle." Managers in any department can thus generate analysis as needed at the given moment.
The future of EMI Music will bring along new requirements in terms of scalability, performance, and flexibility. The volume of data available for BI analysis has grown substantially in recent years, boosted in part by the integration of other national subsidiaries: Austria and Switzerland were incorporated into the system in 2004, followed by the Benelux countries in 2006. Meanwhile, new marketing activities are in the planning stages. EMI has been one of the first enterprises to set up a premium service on the Web jointly with Apple iTunes to allow users to download music without copy protection (DAM). This gives the buyer better audio quality while enhancing the listening experience. Giesenow says, "Decisions such as this one are the result of detailed, pre-emptive analysis. By offering highly flexible, efficient evaluations based on exceptionally voluminous data assets, our IT department enables the enterprise to plan its business activities much better." Page 2 of 2 « previous page PBDJ LATEST STORIES . . .
SUBSCRIBE TO THE WORLD'S MOST POWERFUL NEWSLETTERS SUBSCRIBE TO OUR RSS FEEDS & GET YOUR SYS-CON NEWS LIVE!
|
SYS-CON FEATURED WHITEPAPERS MOST READ THIS WEEK BREAKING POWERBUILDER / SYBASE NEWS
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||