Oracle’s Giant Leap for Democratization of Big Data
Gartner’s magic quadrant
analysis for Business Intelligence and Analytics platforms have mainly three
criteria to assess platforms which I think are spot-on. They focus on enable
(data with minimal technical know-how), produce (efficiency of data analytics
and reports building) and consume (from various platforms, etc.). The theme is
clear as businesses do not have the patience to wait for large implementation
projects or development cycles. They want to know key drivers, critical
insights, hidden opportunities and decision-making data points on various
devices right away. Learn more about Oracle
Cloud Services.
With the emergence and popularity of Big Data
platforms, businesses are able to gain insights from data that was useless and
too large to process in past. Platforms like Hadoop and Spark enable
organizations to process structured and unstructured data at amazing speed and
scale to gain quick insights. But major drawback of Big Data platforms is that
organizations need to rely on specialized resources to build custom process to
analyze data sets for their unique business needs. In most cases people who are
most frequent consumers of data like Business Analysts, Data Scientists,
Testers, Managers, Executives and business users, are not technical enough to
work with Big Data platform directly. Some organizations take route of training
they staff on higher level Big Data scripting languages like PIG and Hive with
minimal success. Today’s users are accustomed to rich user interface with drag
& drop and widget like applications on mobile devices.
Oracle’s Big Data Discovery platform addresses
this problem with a nice twist to it. With partnership with Cloudera and
offerings like No-Sql In-Memory database, Oracle have made it clear that they
continue to be innovative data company. With release of Big Data Discovery
platform, they are bringing data “enable, produce and consume” to non-technical
business focused people. Oracle BDD enables organizations to have Big Data
experts focus on exponential challenges without having to worry about mundane
tasks of loading and extracting data for business users. Also BDD non-technical
staff to perform quick analysis, transformation, charting, graphing and
reporting on Big Data through a web-based interface that works on any platform.
Business
Case for Oracle Big Data Discovery – Allows Hadoop experts to focus on complex data processing
rather than mundane data export tasks – Quick dashboards, charts, graphs and
maps for decision making, communication and operations – Power of Big Data in
hands of business analysts, QA, business users, data scientists and executives
– High-speed data transformations without
programming – Blazing fast search and sampling from Big Data stored in Hadoop
Cluster – Big Data is not always “Batch Data”. BDD uses pre-indexed queries and
Spark transformations providing real-time analytics capabilities unlike batch
processing methods like MapReduce. – Self-service data upload to Hadoop, Hive
and BDD – Command Line Interface for advanced users Technology
Three major components make up Oracle BDD.
BDD
Studio: BDD Studio is a web
application that needs to be deployed on Weblogic server. Studio provides end
user interface on any
browser for analysis of data in Hadoop
cluster. Studio also comes fitted with HDFS and Hive client for advanced users
to analyze data from command line interface.
DGraph & Gateway: DGraph engine is already
famous for high-speed indexing and searching of unstructured data. DGraph
automatically indexes Hive tables from Hadoop cluster using HCatalog. BDD
implements a Gateway component to communicate with DGraph for fast access to
indexed data. Read more: Cloud
360 Assessment
Data
Processing: Data
processing components are deployed to all nodes participating in Hadoop
cluster. Data processors enables BDD studio and DGraph to communicate with
Hadoop cluster using HCatalog and Spark transformation requests.
Original Blog Source: https://www.jadeglobal.com/blog/oracles-giant-leap-democratization-big-data
Comments
Post a Comment