Cloud computing, MapReduce and Hadoop [links]

Cloud computing - types of cloused and etc. are on this page:
http://thecloudtutorial.com/
http://thecloudtutorial.com/cloudcomputingstandards.html
Page describes concepts, etc... cloud computing types and also details about
particular models and solutions ; lists vendors (at this time was Adobe, Amazon,
AT&T Synaptic, GoGrid, Google, IBM, Microsoft, Rackspace, SalesForce, Zoho)
and of course an idea going behind.

MapReduce & Hadoop

MapReduce, an original concept from Google (2004);
idea of handling huge datasets using a large number of computers;
Two steps:

  • Map: master node that takes the input and divides it into many "subproblems" that are distributed to worker nodes
  • Reduce: master node collects the answers (for all sub-problems), combines and outputs

Some about MapReduce: http://en.wikipedia.org/wiki/MapReduce

Hadoop is an Apache open source implementation.
General introduction can be found on this website:
http://thecloudtutorial.com/hadoop-tutorial.html

A more detailed Yahoo-developer tutorial with source code:
http://developer.yahoo.com/hadoop/tutorial/

Describes Hadoop and setting up its components like Hadoop
FileSystem, setting up Hadoop VM-ware for Windows and
finally starting first project for Hadoop in your Java
environment. This description also covers conecpt of MapReduce.