 |
| source |
Data warehouse helps to optimize
the continuous pulling data out of transactional systems and conversion of that
data into ready to use information. Moreover, data warehouse used to process
the humongous amount of complex data and perform queries on that data very
efficiently.
Large organizations prefers data
warehousing because of is numerous improvements and positive gains after
successful implementation.
Enhanced Business Intelligence
Though the improved data access,
decision makes are able to query actual data to retrieve information based on
their needs. Due to the various sources of data, Managers and executives no
longer have to make their decision on limited data. Moreover, data warehouses
can be applied directly to business processing such as Inventory management,
marketing segmentation, sales, and financial management.
Increased System and Query Performance
Data warehouses are designed to
optimize the speed of analysis and retrieval of data. In addition, it is also
designed for storing huge volumes of data and query that data with high speed.
Efficient distribution of system load across an entire organization’s
technology infrastructure reduces the burden on operational environment.
Timely Access of Data
Data warehouses have scheduled
data integration routines known as ETL (Extraction, Transformation, and
Loading) which consolidate data from multiple various sources and transform the
data into actionable information. So that, business users can access data
easily from one interface. Therefore, the consistent use of query and
consolidated data repository tools enables business users to spend more time on
data analysis and minimize the time on gathering data.
Enhanced Data Quality and Consistency
Due to the efficient conversion
of data from various sources into to common actionable format, business units
and other departments can produce the consistent results within the
organization. Production of consistent data from each department will boost up
the confidence in the accuracy of data.
Subsequently, overall confidence in the organization’s data also
increases.
Historical Intelligence
Data warehouse can store large
volumes of historical data, so that, the organization can analyze the data
based on different time periods and trends to make the future predictions.
Advancement in reporting and analysis of multiple time-periods are the main
benefits of the data warehousing.
High Return on Investment
Data warehouse implementation and
other business intelligence systems generates higher amounts of revenue and
more cost savings. Studies states that, organizations that have implemented
data warehouses have increased revenue and decreased expenses than organization
that have not.
Problem areas when implementing Data warehousing:
Data Quality
Due to the large volumes of data
coming from various sources, when tries to combine with inconsistent data from
other sources, it raises the errors. It may encounters data quality challenges
like duplications, inconsistent data, missing data and logic conflicts. Poor
quality of data affects the analytics and reporting.
Cost
Though implementing data
warehouse is to save the expenses, it has other hidden problems with respect to
cost. According to the survey, there are very low number of highly skilled
staff to lead the non-BI technicians. So, with few experienced staff, it is not
easy to deliver effective results. In reality, these kinds of efforts are very
costly. Lastly, high maintenance cost for high maintenance systems.
Integration
Integration of data collected
from various sources is one of the difficult task. Different types of tools for
every operation of the data warehouse. In order to generate the desirable
solution, organization must spend considerable amount of time to analyze how
the various source of data can be integrated.
Suitable Approaches for Data Warehousing:
Inmon’s top-down approach
According to Inmon, data
warehouse is a centralized repository for entire organization. Dimensional data
marts are created after the complete data warehouse is created. Atomic data is
stored at the lowest level of detail in data warehouse. Inmon defines that data
warehouse as subject-oriented, time variant, Non-volatile and integrated
approach.
Kimball’s bottom-up approach
According to Kimball, the data
marts are created first. These data marts provides view about organizational
data and when needed these data marts can be combined into a larger data
warehouse. Kimball’s approach focus on ease of end-user accessibility and high
performance to the data warehouse. Kimball defines, data warehouse is nothing
more than the combination of all data marts.
Finally, while designing data
warehouse, organization must focus on long term and short term business
objectives. Analyze the sources of data and its quality as well as quantity.
Evaluate the level of resources.