---

Cosmos - Big Data Analysis

The Cosmos Big Data Analysis GE is a set of tools that help achieving the tasks of Streaming and Batch processing over context data. These tools are:

• Orion-Flink Connector (Source and Sink)
• Apache Flink Processing Engine
• Apache Spark Processing Engine (work in progress)
• Streaming processing examples using Orion Context Broker

• Documentation - Flink
• Documentation - Spark

Academy Courses

Lesson 1. Big Data Fundamentals

By following this course you will learn about Big Data fundamentals (distributed storing and computing), Map and Reduce basics, Hadoop and its ecosystem.

• Lesson 1 - An Introduction to Big Data

Lesson 2. HDFS and Map Reduce

By following this course you will be able to use the Global Instance of Cosmos in FIWARE Lab. Basically, doing I/O with Web HDFS API and submitting and managing MapReduce jobs with Tidopp API.

• Lesson 2 - HDFS and Map Reduce

Lesson 3. Big Data Analysis

This video presentation explains context information processing and advanced Big Data Analysis.

Step-by-Step Tutorials

Use of the Cosmos Flink Connector is described in the following step-by-step tutorial:

• FIWARE 505: Real-time Processing and Big Data Analysis
• FIWARE 506: Real-time Processing and Big Data Analysis

Code Examples

This GitHub repository exists containing a few examples for getting started with the cosmos flink connector is described in the following documentation:

• Cosmos Flink Examples
• Cosmos Spark Examples