ArmanMaesumi / java-monte-carlo

A Java library for multithreaded Monte Carlo simulations.

java-monte-carlo

A Java library for multithreaded Monte Carlo simulations. Current releases

See Tester.java for sample code

To Implement MonteCarlo interface:

// Import Parallel Monte Carlo library
import arman.edu.utexas.cs.ParallelMonteCarlo;
// Implement MonteCarlo interface:
public class Tester implements MonteCarlo {

	// Code for example in defaultSimulation():
	public AtomicInteger sum = new AtomicInteger(0);
    
    	// MonteCarlo interface method:
    	// SIMULATION_MODE_DEFAULT:
    	public void defaultSimulation() {
		// This simulation mode will preform n number of iterations of this method on set number of threads.
		// In this simulation mode, the method does not have access to the current
		// iteration of the simulation. You should use the default simulation mode
		// when you are doing n number of INDEPENDANT simulations, meaning the current
		// simulation, n_i, does not effect the output of this method.

		// Example 1:
		// Concurrently calculate the average value of random(0, 10) * random (0, 10):
		// Use AtomicInteger, and ThreadLocalRandom classes for safe concurrent use:
		int a = ThreadLocalRandom.current().nextInt(1, 10);
		int b = ThreadLocalRandom.current().nextInt(1, 10);

		// Accumulate a*b in AtomicInteger:
		// AtomicLong may be more suited depending on your number of iterations.
		sum.addAndGet(a * b);
		// See Example 1 code in initialize()
    	}
    
    	// MonteCarlo interface method:
 	// SIMULATION_MODE_DOMAIN:
    	public void domainSimulation(long i) {
    		// This method acts similarly to defaultSimulation(), the primary difference is that
		// domain simulations are provided the current iteration that is being processed.
		// This means that the value "long i" will attain every value from 0 to n, where
		// n is the total number of iterations.

		// Example 2:
		// Concurrently print all primes from 0 to n.

		if (isPrime(i))
			System.out.println(i + " is a prime.");
    	}
    
    	// MonteCarlo interface method:
    	public void initialize() {
		// Set simulation environment to this class instance:
		sim.setEnvironment(this);

		// Set number of iterations to perform:
		sim.setIterations(100000);

		// Set number of threads to use in simuation:
		sim.setThreads(2);

		// Set the simulation mode:
		sim.setMode(SIMULATION_MODE_DEFAULT);
		// or
		// sim.setMode(SIMULATION_MODE_DOMAIN);

		// Run simulation:
		sim.start();

		// Example 1 code:
		// Print ratio of total value of a*b to the total number of iterations.
		// Force double precision result with *1.0:
		System.out.println("Average value of a*b = " + sum.get() / (sim.getIterations() * 1.0));
    	}
}