Page 43 - BANKING FINANCE APRIL 2017

ARTICLE

tional data warehouse. For Big Data Analytics cannot do Softwares for Big Data Analysis:
interactive queries over tabular data structures, cannot
Whenever people think of Big Data Analysis, they always
support online analytical processing (OLAP), or online trans-
connect it with Hadoop. But Big Data is not just Hadoop
action processing (OLTP) applications.
because in Cyber security, stock market, traffic control, sen-
sor information, and monitoring trends in social media
But it can very well augment the existing analytic infrastruc-
Hadoop is not applicable.
ture, enables one to filter through high volumes of raw data
and combine the results with structured data stored in
And if an Organisation has many silos of information, it
DBMS or warehouse. In short Big Data doesn't work alone would be difficult to move these silos of information to an
or in silos but can be integrated with traditional data for
Hadoop Distributed File System (HDFS) where Hadoop stores
arriving at new insights.
its data. That is why Big Data Analysis is dependent on a
number of softwares which helps to analyze a diverse range
of data, including data that's loosely structured or largely
unstructured.

Here again Data means, apart from financial data, various
types of text data like financial documents, legal documents,
Hierarchical Data Data Ware Housing Big Data
marketing collateral, emails, blogs, news reports, press re-
leases, and social media websites fall into the set. The fol-
Advantages of Big Data Analytics: lowing are the list of softwares which help in Big Data
Most Organisations have the notion that Business intelli- Analytics.
gence is reporting. Because of this traditional data analysis 1. Apache Hadoop, a software framework for data-inten-
tools are taking the Organisations to the future as if a car is sive applications that exploit distributed computing en-
driven by looking into the rear mirror. vironments

2. Pig, a high-level programming language and runtime
Instead if technology is used to delve through Big Data stores
environment for Hadoop
for predictive analytics, it would enable banks to provide
3. Jaql, a high-level query language based on JavaScript
more value for their customers and help in understanding
their views and opinions. It would also act as an enabler for Object Notation (JSON), which also supports SQL.
doing business, introduce better cost structures and end up 4. Hive, a data warehouse infrastructure designed to sup-
with better revenue. port batch queries and analysis of files managed by
Hadoop
This is possible because unlike in traditional approach where 5. Hbase, a column-oriented data storage environment
IT is hell bent on finding solutions to a question raised by designed to support large, sparsely populated tables in
the business unit, Big Data Approach will ensure that ques- Hadoop
tions will come out from the iterations of Data. This will act
6. Flume, a facility for collecting and loading data into
as driving a car by following the Google map as business
Hadoop
have to be led by insights from Big Data.
7. Lucene, text search and indexing technology
The POS data, CRM data, Social media data etc will help 8. Avro, data serialization technology
Banks in realising what customers are saying about the prod-
9. ZooKeeper, a coordination service for distributed appli-
uct & services, about the competitors, their likes and their
cations
dislikes, spending habits because a lot of buying is happen-
ing today with social relationship. Hence the products and 10. Oozie,workflow/job orchestration technology
services should appeal to individual tastes and connect to
the brain at the time of buying. Also a data analysis tool is developed by BigInsights Enter-

BANKING FINANCE | APRIL | 2017 | 43

Insurance Training Centre © Call 09883398055 / 09883380339