Database Reference
In-Depth Information
A major theme of this topic is to reduce costs and infrastructure needs by consid-
ering the use of cloud computing solutions whenever possible. Some companies are
building data warehousing systems completely in the cloud. Amazon's Redshift is an
example of a product that attempts to take advantage of an all-cloud architecture.
One of Redshift's key design principles is that data can be loaded from other Amazon
cloud data sources, such as DynamoDB and Elastic MapReduce. Other companies are
following suit as well, launching completely managed Hadoop Hive installations on
virtual clusters. Cloud-based data warehousing leads to a secondary advantage; as more
and more application development is done on the Web, having a data warehouse com-
pletely in the cloud makes it easier for browser-based visualization and reporting tools
to access the underlying source data.
Technology now exists to collect, process, and store data on a large scale. Data can
be processed with tools such as MapReduce, and aggregate analytics tools are get-
ting faster and faster as well. All of this can eventually take place using cloud utility
computing systems, with Web-based visualization and business productivity tools pro-
viding an interface to the data. With all of these components in place, the ability to
address many of the technical problems that result from data silos is available. Perhaps
common storage and standardized interoperability between cloud-based systems will
make transformation between data sources a much more trivial process. These tech-
nological developments, combined with a cultural change in organizational attitudes
toward empowering employees to become data experts, means that discussion about
the challenges caused by data silos will likely become less and less important.
Will Luhn's Business Intelligence System Become Reality?
The business intelligence system that Luhn envisioned was not merely a tool to ask
questions about organizational data. Luhn's ultimate goal was to create a technology to
identify patterns of interest within the stream of business data and then push relevant
information based on these models to those who needed it—at just the right time.
Luhn said that such a system would ensure that new information pertinent or useful
to certain action points would be selectively disseminated to those points without any
delay.
One might argue that some aspects of the business intelligence system that Luhn
envisioned are already available, thanks to a technology that was born near the end
of his life: the Internet. Many Web applications are able to use crowdsourced data to
present users with ratings, advice, and targeted advertising. Predictive analytics are
being used to build recommendation engines for users, pushing data to mobile devices
depending on the user's location or time of day. Some aspects of the business intelli-
gence system that Luhn envisioned are being developed today, but at a scale far beyond
what his original paper conveyed.
Search WWH ::
Custom Search