README

Daemon: ghadoopd
	This is the daemon that runs GreenHadoop.

Core: ghadoop.py
	This is the core for GreenHadoop.
	It includes:
		- Energy scheduling
		- Data management
	The high level algorithm is in "algorithm.txt".

Main configuration: ghadoopcommons.py
	This is the main file for configuration.

Solar prediction: greenavailability/
	It contains the solar availability predictor.

Other important files for GreenHadoop:
	ghadooplogger.py: threads for logging of energy, jobs, etc.
	ghdfsmonitor.py: thread to monitor for HDFS.
	ghadoopmonitor.py: thread to monitor MapReduce part of Hadoop.
	ghadoopwaiting.py: thread to manage the waiting queue.
	ghadoopreplication.py: thread to handle file replication.

Auxiliary components:
	ghadoopdata.py: some data structures definitions.
	ghadoopworklow.py: heritance from GreenSlot that helps setting the deadlines.

Experiments:
	experimenter.sh: script to start all the set of experiments.
	ghadoopparser.py: parse the GreenHadoop outputs.

Patch:
	Hadoop (Namenode) needs to be modified. In particular,
	hdfs/src/java/org/apache/hadoop/hdfs/server/namenode/
	greenhadoop.patch provides the path for this.
	It applies to Hadoop 0.21.0.

More information:
	http://www.research.rutgers.edu/~goiri/doc/eurosys2012.pdf