Page 105 - Building Big Data Applications

Chapter 5 Pharmacy industry applications and usage 101

Analy cs

Data Lake & Hub
Metrics
Volume EDW Enterprise Data Swamp

Opera onal Data
KPI OLAP / Analy cs discovery Explora on
MOLAP

Data Granularity

FIGURE 5.1 Volume versus granularity.

out as they might not be significant in the layer needed, resulting in complexity of
lineage, and often nonadoption of the data and its associated insights.
Formulation of the data across the layers of compute is very intricately complex and
in the traditional world of ETL we often have to add trace to detect if there are rejects or
extremely complex formula of calculations causing both performance and traceability
issues. We tend to either ignore the issue or it is added to a list of bug fixes, which never
happen or happen over a long period of time. This issue of complexity becomes a
nightmare when we deal with analytics.
Analytics in any enterprise whether small or large is an essential topic. The core
foundations of the enterprise are depending on its analytics, whether we talk about
earnings, losses, expenses, incomes, or stock prices. We measure and depending on the
measure outcomes, we work with the data and decide the performance of the enterprise.
In order to ensure sustained positive measures, the enterprise measures its customers’
journeys, its products’ journeys, its supply chain journeys, its research journeys, its
competitive journeys, its marketing journeys, its sales journeys, and even its operational
journeys. We have become obsessed with the measurements and metrics, that we even
measure the time it takes to place an order or receive a pick-up. These metrics and
measure aspects increase even more in the world of internet of things, the research and
innovation areas, healthcare and medicine-related research, patient treatmenterelated
research and forays into space and universes are all magnitudes of complexity to
comprehend and deliver. Analytics can provide a tipping point, but to get to that tipping
point, we need to figure out the methods and techniques. This is where we will introduce
you to a new method of defining and designing for complexity. It is a part of the new data
management strategy and architecture process.