September 29, 2014
There are many good reasons, both business and technical, for
modernizing a data warehouse (DW). To sort it all out, it’s best to
put business reasons first. After all, in IT we provide technology that
supports business goals. When it comes to DW modernization, most
business priorities align with the following:
- Big data. The organization wants to capture new data sources
and leverage them for business advantage through business
intelligence (BI) and analytics
- Advanced analytics. The business is under pressure to compete
and grow, based on analytic insights, both short- and long-term
- Real time. Operations need high-performance and real-time
technologies so they can close sales, serve customers, and react
to market events sooner and more frequently
Technology goals are important, too, especially when they support
the business goals of data warehouse modernization:
- Assure capacity for growth. BI/DW professionals can expect
increases in data volumes, concurrent users, reports, analytics
applications, sandboxes, and complex workloads for analytics,
data integration, data quality, real time, and so on. In an effort
to “future proof” warehouse capacity, users are increasingly
depending on easily expanded data platforms based on racks,
clusters, grids, elastic clouds, and Hadoop clusters.
- Diversify the types of data platforms in the data warehouse
environment. A modern organization will practice several
distinct BI/DW disciplines, such as reporting, visualization,
OLAP, and many forms of advanced analytics. Because each
discipline has unique workload characteristics, each may need
its own standalone tool or platform within the extended data
warehouse environment. This fact is driving many BI/DW teams
to complement and extend the core DW with analytic databases,
appliances, Hadoop, and streaming data tools.
- Turn on new functionality. Sometimes business and technology
goals can be met with tools you already have by turning on
functions that you haven’t used before, such as data federation
services, in-memory functions, and in-database analytics.
- Rip and replace. When appropriate, migrate to new platforms
and tools that can handle a broader range of data types and are
faster, more scalable, tuned for analytics, and so on.
This Checklist Report explores the leading business reasons for
modernizing a data warehouse, plus the common technical measures
taken today for data warehouse modernization.