MapReduce

[hooshdar3]

Hello.
Can Any one please help me to get one MaapReduce program run(using Hadoop)?

This is the link:
http://cs.smith.edu/dftwiki/index.php/Hadoop_Tutorial_2.2_--_Running_C%2B%2B_Programs_on_Hadoop#Running_the_WordCount_program_in_C.2B.2B

Thanks in advance

[rapidcoder]

WTF would someone create *new* Hadoop jobs in C++? LOL.
Not only it will be much slower than the native Java version, but also much bigger pain to run.

[hooshdar3]

I have mostly coded in C++ for the past 10 years. For now I am forced to do it.

And Java is more slow probably. since it is an interpretive language.

EDIT:I didn't say I want to create "new" programs. I just want to run this example. I am to give a presentation in front of a class and want to have implemented that particular program as an example.

[LB]

@rapidcoder: show me the benchmarks and I may defend you.

[JLBorges]

> And Java is more slow probably. since it is an interpretive language.

Java is slower primarily because it produces as inordinate amount of garbage, and ends up using up more and more resources for janitation.

If the amonut of data is not huge, anything goes; otherwise this is typical of C++ vs Java performance (atop Hadoop HDFS). http://hypertable.com/why_hypertable/hypertable_vs_hbase_2/


> I just want to run this example.

The section on "Running the WordCount program in C++" seems to have complete step by step instructions.

What problems did you encounter?

[hooshdar3]

I simply don't have the Hadoop library.

[LB]

http://hadoop.apache.org/