"No one is harder on a talented person than the person themselves" - Linda Wilkinson ; "Trust your guts and don't follow the herd" ; "Validate direction not destination" ;

March 24, 2012

Hadoop VS Vertica - Big Data Webinar

[You may also like - Big Data - Basics - Getting Started]

Vertica Ranked #1 based on Big Data Revenues. Please check webinar on Hadoop VS RDBMS for Big Data Analytics. Captured are notes from the Webinar.
  • Big Data volumes Projected growth by 2020 is 35 ZB
Hadoop Overview
  • Support unstructured data processing
  • Hadoop / Map reduce programming
  • Different type of computations possible 
  • Low cost Infrastructure
  • Batch Processing based not real time

Survey of Big Data Tools based on Parameters (Deep Analytics not just aggregating Data, Real Time Analytics, Volume of Data)
Veritca seems to succeed in terms of all above parameters. Vertica features listed below
  • RDBMS  compliant
  • Elastic, Scalable
  • Auto tuning feature
  • Built in Analytics
  • Based on MPP Architecture
  • Real time Analytics Support

Vertica Product Details
  • Column based Storage Overview (Key differentiator is I/O savings compared to traditional Row based access)
  • Support for Column Based Compression
  • High Availability Support
  • Less storage space required to store data in Vertica compared for Hadoop for same amount of Data
Summary
  • Hadoop is good for Exploratory Analysis
  • Vertica recommended for Real time /Interactive Analytics. Vertica also provides Hadoop Connector to pull data from Hadoop

Happy Learning!!! 

No comments: