Look for a general overview of Hadoop

Question

I am looking for some Hadoop performance reviews (cluster of 300-600 boxes, commercial equipment), especially on the following aspects:

• High concurrent read and write
• Web Crawl
• Mapreduce parallel computing
• Inverted index

Answer 1

This is not a specific question, maybe that's why no one has answered yet. Performance on a cluster of 3-600 nodes is best analyzed using tests.

However, I found some really interesting articles about Hadoop and its implementations in production:

• Hadoop architecture and its use on Facebook
• Rackspace MapReduce Hadoop
• Hadoop 16.25 62
• , , Hadoop
• , facebook hadoop, - Hive - Petabyte Hadoop

, .