Page 36 - index

Data-centric security — a strategy that precisely pinpoints, protects, and continuously
inspects data at the element level — can address the risks of big data while preserving data
accessibility across a broad user base and a variety of uses in Hadoop. By adopting a data-
centric security strategy in addition to the traditional security tools, and organizations can
answer fundamentally important questions about big data: what kind of sensitive data is
present across our enterprise? and where does that sensitive data reside? After answering
those questions, the right steps can be taken to protect sensitive data, using data use cases
and enterprise privacy policies that work in tandem.

A data-centric security strategy can result in more accurate risk management, more flexible
policies for data sharing, and critical visibility and precision within data structures including
clickstreams, logs, user documents, and e-mail content that now have some of the highest
growth rates in Hadoop.

Everything Starts With Discovery

Data-centric discovery for Hadoop allows businesses to detect sensitive data at the element
level, so that organizations that are bringing Credit Cards, Social Security Numbers, Names,
Addresses, Health Records, Financial Performance Results can determine where, how much,
and how often these elements are found across the entire data store in Hadoop. Customers use
discovery in five fundamental ways: visibility, as a means to protect data, as a means to count
and realm how much sensitive data exists, as a way to risk profile and tie sensitive elements
together, and as a way to continuously monitor data for changes or new risks.

Business Benefit Hadoop Data Discovery

Visibility into risk Discovery reports all sensitive data across
entire Hadoop.

Comprehensive Can run against data stored in Hadoop
Coverage (HDFS), NoSQL (Cassandra/Datastax),
Relational Data Management Systems
(RDBMS), File Systems and Sharepoint

Real-time coverage Can run in data-in-motion through agents
for Flume, FTP, Sqoop, or Kafka (Summer
2015)

Protection of data Automated protection policies provide
options to encrypt (AES/FPE), mask, or

36 Cyber Warnings E-Magazine – June 2015 Edition
Copyright © Cyber Defense Magazine, All rights reserved worldwide