2012-03-27

Hadoop

• 一些關於 Hadoop 的學習資訊：
  • MapReduce Framework
    • Great 1 hour video introduction: ​http://nosqltapes.com/video/understanding-mapreduce-with-mike-miller
    • Read the famous 2004 paper from Google that kicked off the MapReduce revolution. This is a very readable paper that can be digested in about 2 - 3 hours: ​http://research.google.com/archive/mapreduce.html
    • Here's a 33 minute video on what kinds of simple things you can do with MapReduce: ​http://www.cloudera.com/videos/mapreduce_algorithms
    • Google's MapReduce course: ​http://code.google.com/edu/parallel/mapreduce-tutorial.html
  • Beginner Hadoop
    • Excellent beginner's video on understanding Hadoop, MapReduce and HDFS: ​http://www.cloudera.com/protected/?resource=introduction-to-apache-mapreduce-and-hdfs
    • Understanding the Hadoop ecosystem: ​http://www.cloudera.com/protected/?resource=apache-hadoop-ecosystem
  • HDFS
    • An easy 2-3 hour read about Hadoop's distributed File System: ​http://www.aosabook.org/en/hdfs.html
  • Labs
    • Install VirtualBox on your laptop, get an Ubuntu Virtual Machine going and follow this excellent tutorial to install your first Hadoop node: ​http://www.michael-noll.com/tutorials/running-hadoop-on-ubuntu-linux-single-node-cluster/
    • Then use this to scale your cluster to multiple nodes: ​http://www.michael-noll.com/tutorials/running-hadoop-on-ubuntu-linux-multi-node-cluster/
    • Run a MapReduce job in Python on your cluster: ​http://www.michael-noll.com/tutorials/writing-an-hadoop-mapreduce-program-in-python/