Distributed Framework For Running Jobs In Javascript

INSTALL
============================================================================
Requirements: Tomcat, java6 , mysql server.
Optional: Ant

Install (Do Steps 1,2,3,6,7 (ant installed) OR 1,4,5,6,7 (ant not installed)): 
1. Setup a tomcat and mysql server.
2. In the webapps folder of tomcat create a new folder called distSystem
3. In the root folder of the project run ./install.sh param1
	Where param1 is the path to tomcat e.g. /usr/share/tomcat7/
	(Warning: The install script will stop/start tomcat)
4. Download the zip file containing a compiled version (build_XX.zip)
5. Extract the zip file into a folder in webapps
6. In the root project folder run the following commands:
	6.1 mysql -u root -p < database.sql
	6.2 mysql -u root -p < jobs/input_jobOne.sql
	6.3 mysql -u root -p < jobs/input_jobTwo.sql
	
	(6.2/6.3 will take some time to complete)
	
7. In your browser visit localhost:8080/<folderNameInWebapps>

Notes: the command ant package will recompile the project
Tested in google-chrome &  chromium

Running Demo:
Inside the college network the project is running on https://149.157.243.77:8080/concepts/ (Self signed cert)


DESCRIPTION
============================================================================
This distributed processing system is implemented in java/jsp.
The server is event driven and responds to requests coming in from the interface layer (jsp)
This interacts with a mysql to return jobs/data while recieving results.

Each new job is added into the JOB table.
	After this a DATA_<JOBID> table and a RESULT_<JOBID> is created for the job. These must contain certail fields.
	These fields can be seen in sample.sql
	
When a server first loads it creates a list of jobs in memory.
	When a request comes in for a job a counter is incremented (this counter can wrap around)
	So each request sends a different job. This ensures that each job gets a fair amount of processing.
	
	Each job stores a list of data to be sent and when a request comes in a data element is send & removed from the list.
	Storing the data in a list reduces the amount of database activity and allows the app to run multiple instances.
	
	
	When this list becomes empty a request is sent to the database for X more jobs
	and a process is also run to ensure conistancy of the database (data that was sent but no result recieved are reset)
	These calls are synchronisied to prevent concurrency issues.
	
	When a result is returned a check is performed to see if it is already in the database.
	If it is not then the result is added to the DB, if it is already in the database then the result is checked.
		If the results are equal then the data is marked as solved, otherwise it is reset to be sent again.
		
		
	As well as being a distributed javascript processor framework (solving data in JS and then sending the results back to the server)
	
	The server can run multiple instances at the same time and from different places.
	
	If the database is stored on a cloud platform then the server side can scale effectively as the number of
	requests increase.