Database Reference
In-Depth Information
advantage with self-service data generation through to right-sized environ-
ments, Optim TDM streamlines the test data process.
Big Data is associated with data growth and there's not question that data is
growing in every one of your enterprise systems. Data Lifecycle Management
should be a top priority to help curb unchecked Big Data growth, reducing the
cost of data while making your applications more efficient.
Privacy and Security
There are multiple privacy and security notations within an information in-
tegration and governance discussion, most of which can be applied to Big
Data. You need to protect and block unauthorized access to sensitive data no
matter where it resides. If you have to apply governance to a certain class of
data that you collect, there are privacy and security concerns whether you
store this data in a file system (such as the HDFS) or in a relational database
management system (RDBMS). For example, your security mantra (separa-
tion of duties, separation of concern, principle of least privilege, and defense
in depth) applies to data stored anywhere. You'll want to consider role-based
security, multitenancy, and reduced surface area configurations through
reverse proxies (among other security services), all of which BigInsights can
provide to Hadoop.
Of course, it's worth noting that if IBM InfoSphere Guardium (Guardium)
is an industry leader in auditing and alerts via its heterogeneous data activ-
ity monitoring (DAM) services, based on activities at the data management
level, why couldn't it do so for the HDFS? In 3Q 2012, IBM announced Guar-
dium's initial support for providing DAM services to a Hadoop environment
(NameNode, JobTracker, and DataNodes) and its subsystem projects (for
example, Oozie), giving administrators the ability to clearly understand who
did what, who touched what, and so on. As of this writing, most BigInsights
components that currently have audit logs can be monitored; for example,
HDFS name space operations, MapReduce (job queue and job operations,
refresh configuration), HBase Region Server (database activity), Hive, and
Avro. Because Guardium works with open source components, it can inte-
grate with BigInsights and/or other open source distributions of Hadoop.
Guardium also recognizes the Thrift and MySQL protocols (used by Hive).
These components all send existing BigInsights audit logs to Guardium and
 
Search WWH ::




Custom Search