In Memory OLAP on HADOOP using SPARK

Target Audience:

HADOOP and SPARK architects, Big Data Architects, Data Scientists and Product Managers

What will you learn:

How to enrich interactive analytics on Hadoop, with in-memory query engine which plugs into the Spark execution framework

Location:

FEATURE HERZLIYA, 9 HA-MANOFIM ST. HERZLIYA, www.feature.co.il (http://www.feature.co.il/)

Google Map: https://goo.gl/maps/scyTW7SNtsS2

Agenda:

17:30 - 18:00 Gathering (food and beverage)

18:00 - 20:00: Session

Presenter:

Mr. Javier Cuerva, Senior Enterprise Solution Architect, SAP Global Center of Expertise,

LinkedIn Profile: https://www.linkedin.com/in/javier-cuerva-6a031b8

Abstract:

Digital economy catalysts, such as social, mobile and the Internet of Things, require companies to make business decisions by analyzing ever-larger datasets that can be stored everywhere. Many enterprises have experimented with deploying large data lake projects on Hadoop to accommodate the scale of Big Data storage economically. However, processing and analyzing all this distributed Big Data has proven challenging, especially within the context of existing enterprise app and analytic processes.

Some of the current challenges are:

· In-efficient Data Processing: Current batch-mode tools used for processing data in Hadoop do not support real-time, interactive, drill-down analysis needed for most business-oriented analytics.

· Lack of Business Alignment: Unstructured raw data in Hadoop systems contain valuable information but lacks semantics to understand the business context.

In this session we will present the ability to enable OLAP analysis of Hadoop data through data hierarchy enhancements in SparkSQL and compiled queries for accelerated processing across nodes.

The following features will be presented:

· Hierarchy processing support for Spark e.g. Drill-down data analysis using data hierarchies for business analysis

· The ability to boost performance compared to Spark

· How to run an holistic big data solution in one product:

· How to work on top of the 3 major Hadoop distributions ( HortonWorks, Cloudera, MapR)

· How to access SQL based via any JDBC/ODBC tool

· How to Scale and distribute: Use Hadoop ecosystem

· How to manage hierarchy on Hadoop

http://photos3.meetupstatic.com/photos/event/5/5/e/4/600_447981988.jpeg