Future Of Data : Big Data

Evolution of Big Data

Transcript

1. Future of Data : Big Data Shankar Radhakrishnan Cognizant
2. Topics  How did we get here ?  Data Explosion  Big Data  Big Data in an Enterprise  Big Data Platform - Hadoop  Big Data Adoption  Q & A
3. How did we get here? Familiar World  EDW  Datamarts  Familiar Problems New World  Newer type of data to integrate  Increase in volume  Newer analytical requirements Data warehouse Data Integration Problems Data Processing Problems Storage Management Performance Problems Limitations out of Complexity
4. Data Explosion
5. Newer Interests  Social Intelligence  DBIM, Sentiment Analysis, Social Customer Care  Predictive Analytics  Propensity, Price Elasticity, Anti-Fraud Analytics  Segmentation Insights  Funnel Analysis, Behavioral Patterns, Cohort Analysis  Mobile Analytics  Ad-Targeting, Geo-spatial Analytics
6. Categories  Structured Data  Enterprise Data (CRM, ERP, Data Stores, Reference Data)  Semi-structured Data  Machine Generated Data (Sensor Data, RFIDs)  Unstructured Data  Social Data (Comments, Tweets), Blog posts
7. Big Data Big Data Volume Velocity Variety Complexity “Big Data” refers to high volume, velocity, variety and complex information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making
8. Big Data Platforms • Data Integration o Informatica, Infosphere o talenD, Pentaho, Karmasphere, Apache Sqoop, Apache Flume • Database Framework o Hadoop (Distributions: Cloudera, Hortonworks, MapR) o Hbase o Hive • NoSQL Databases o MongoDB, CouchDB • Machine Data Processing o Splunk, Mahout • Text Analytics o Clarabridge, Lexanalytics
9. Big Data in an Enterprise Data warehouse Data Sources ETL Big Data Sources ETL Big Data Platform ETL Datamarts Datamarts Datamarts Analytical Applications
10. Hadoop - Ecosystem
11. Big Data : Adoption Drivers Platform Cluster Storage Availability Process Distributed Scalable Performance Possibilities Data Integration Data Processing Actionable Insights Ecosystem Augmented TCO ROI
12. Big Data – Adoption Scenarios  Replatforming to Big Data (Hadoop, MapR)  Archival Solution (Hadoop)  Offloading Data warehouse, EDW (Hadoop, Hive)  Social Media Integration  Machine Data Analysis (Splunk, Mahout)  Complex Analytical Requirements (Hbase)
13. Q & A