1. Pattern Programming with the Seeds Framework © 2013 B. Wilkinson/Clayton Ferner SIGCSE 2013 Workshop 31 intro.ppt Modification date: Feb 17, 2013 1

2. Basic User Programmer Interface To create and execute parallel programs, programmer selects a pattern and implements three principal Java methods with a module class: Diffuse method – to distribute pieces of data. Compute method – the actual computation Gather method – used to gather the results Programmer also has to fill in details in a “bootstrap” class to deploy and start the framework. Diffuse Compute Gather “Run module” bootstrap class The framework self-deploys on a geographically distributed platform and executes pattern. 2 “Module” class

3. public Data DiffuseData (int segment) { DataMap d =new DataMap (); input Data = …. d.put(“name_of_inputdata", inputData); return d; } public Data Compute (Data data) { DataMap input = (DataMap )data; //data produced by DiffuseData() DataMap output = new DataMap (); //output returned to gatherdata inputData = input.get(“name_of_inputdata”); … // computation output.put("name_of _results", results); // to return to GatherData() return output; } public void GatherData (int segment, Data dat) { DataMap out = (DataMap ) dat; outdata = out.get (“name_of_results”); result … // aggregate outdata from all the worker nodes. result a private variable } By framework segment used by Framework to keep track of where to put results GatherData gives back Data object with a segment number Data cast into a DataMap 3 Module class

4. Seeds Workpool DiffuseData, Compute, and GatherData Methods DiffuseData DataMap d Returns d to each slave DataMap d created in diffuse DataMap output created in compute DataMap output GatherData Private variable total (answer) Compute DataMap input Master Slaves Note DiffuseData, Compute and GatherData methods start with a capital letter although method names should not! 4

5. DataMap methods put (String, data) – puts data into DataMap identified by string get (String) – gets stored data identified by string DataMap extends Java HashMap which implement a Map, see http://doc.java.sun.com/DocWeb/api/java.util.HashMap Data and DataMap classes For implementation convenience two classes: Data class used to pass data between master and slaves Uses a “segment” number to keep track of packets as they go from one method to another. DataMap class inside compute method DataMap is a subclass of Data and so allows casting. 5

6. Bootstrap class package edu.uncc.grid.example.workpool; import java.io.IOException; import net.jxta.pipe.PipeID; import edu.uncc.grid.pgaf.Anchor; import edu.uncc.grid.pgaf.Operand; import edu.uncc.grid.pgaf.Seeds; import edu.uncc.grid.pgaf.p2p.Types; public class RunMonteCarloPiModule { public static void main(String[] args) { try { MyModule pi = new MyModule(); Seeds.start( "/path/to/seeds/seed/folder", false); PipeID id = Seeds.startPattern(new Operand( (String[])null, new Anchor( "hostname", Types.DataFlowRoll.SINK_SOURCE), pi ) ); System.out.println(id.toString() ); Seeds.waitOnPattern(id); System.out.println( "The result is: " + pi.getPi() ) ; Seeds.stop(); } catch (SecurityException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } catch (Exception e) { e.printStackTrace(); } 6 This code deploys framework and starts execution of pattern Different patterns have similar code Name of module class

7. Framework methods used in Bootstrapping class Seeds methods start // starts framework, deploy nodes // on list of servers startPattern // starts seeds pattern waitOnPattern// waits for pattern to complete stop//stops framework 7

8. User methods used in Bootstrap class Additional methods can be specified by programmer in the Workpool class and can be invoked in the Bootstrap class. Typically a method is invoked that produces the final result. Example public double getPi() { // returns value of pi based all workers double pi = (total / (random_samples * DoubleDataSize)) * 4; return pi; } 8

9. Seeds Implementations JXTA P2P networking version suitable for a fully distributed network of computers and requiring an Internet connection even in just running on a single computer, “No Network” JXTA P2P version for running on a single computer, not requiring an Internet connection Multicore version implemented with threads for more efficient execution on single multicore computer or shared memory multiprocessor system -- does not require an Internet connection. Two JXTA versions use the same application code and run in a similar fashion Multicore version use same Module code but slightly Bootstrap Run Module source code (see next) 9

10. Multicore version of Seeds Thread-based version using shared memory Faster than JXTA P2P version on a multicore platform Bootstrap class does not need to start and stop JXTA P2P. Seeds.start() and Seeds.stop() not needed. Module class unchanged 10 public class RunMonteCarloPiModule { public static void main(String[] args) { try { MyModule pi=new MyiModule(); Thread id = Seeds.startPatternMulticore( new Operand( (String[])null, new Anchor( args[0], Types.DataFlowRole.SINK_SOURCE), pi ), 4 ); id.join(); System.out.println( "The result is: " + pi.getPi() ) ; } catch (SecurityException e) { … }

11. Workpool Pattern 1.Embarrassing Parallel Computation Monte Carlo  11

12. Monte Carlo Methods A so-called “embarrassingly parallel” computation as it decomposes into obviously independent tasks that can be done in parallel without any inter-task communications during the computation. Monte Carlo methods use random selections. (For parallelizing Monte Carlo code, must address best way to generate random numbers in parallel.) 12

13. Circle formed within a 2 x 2 square. Ratio of area of circle to square given by: Points within square chosen randomly. Score kept of how many points happen to lie within circle. Fraction of points within circle will be approximately  /4 given sufficient number of randomly selected samples. 13

14. One quadrant can be described by integral: Random pairs of numbers, (xr,yr) generated, each between 0 and 1. Counted as in circle if 14

15. public Data DiffuseData (int segment) { DataMap d =new DataMap (); d.put("seed", R.nextLong()); return d; // returns a random seed for each job unit } DiffuseData Method (Required to be implemented) Seeds Monte Carlo code MonteCarloPiModule.java 15

16. public Data Compute (Data data) { DataMap input = (DataMap )data; DataMap output = new DataMap (); Long seed = (Long) input.get("seed"); // get random seed Random r = new Random(); r.setSeed(seed); Long inside = 0L; for (int i = 0; i < DoubleDataSize ; i++) { double x = r.nextDouble(); double y = r.nextDouble(); double dist = x * x + y * y; if (dist <= 1.0) { ++inside; } output.put("inside", inside); // to return to GatherData() return output; } Compute Method (Required to be implemented) 16

17. public void GatherData (int segment, Data dat) { DataMap out = (DataMap ) dat; Long inside = (Long) out.get("inside"); total += inside; // aggregate answer from all the worker nodes. } GatherData Method (Required to be implemented) 17

18. getDataCount Method (Required to be implemented) public int getDataCount() { return random_samples; } 18

19. Method to compute  result (used in bootstrap module) public double getPi() { // returns value of pi based on all workers double pi = (total / (random_samples * DoubleDataSize)) * 4; return pi; } 19

20. package edu.uncc.grid.example.workpool; import java.util.Random; import java.util.logging.Level; import edu.uncc.grid.pgaf.datamodules.Data; import edu.uncc.grid.pgaf.datamodules.DataMap; import edu.uncc.grid.pgaf.interfaces.basic.Workpool; import edu.uncc.grid.pgaf.p2p.Node; public class MonteCarloPiModule extends Workpool { private static final long serialVersionUID = 1L; private static final int DoubleDataSize = 1000; double total; int random_samples; Random R; public MonteCarloPiModule() { R = new Random(); } @Override public void initializeModule(String[] args) { total = 0; Node.getLog().setLevel(Level.WARNING); // reduce verbosity for logging random_samples = 3000; // set number of random samples } 20 public Data Compute (Data data) { // input gets the data produced by DiffuseData() DataMap input = (DataMap )data; // output will emit the partial answers done by this method DataMap output = new DataMap (); Long seed = (Long) input.get("seed"); // get random seed Random r = new Random(); r.setSeed(seed); Long inside = 0L; for (int i = 0; i < DoubleDataSize ; i++) { double x = r.nextDouble(); double y = r.nextDouble(); double dist = x * x + y * y; if (dist <= 1.0) { ++inside; } output.put("inside", inside);// store partial answer to return to GatherData() return output; } public Data DiffuseData (int segment) { DataMap d =new DataMap (); d.put("seed", R.nextLong()); return d; // returns a random seed for each job unit } public void GatherData (int segment, Data dat) { DataMap out = (DataMap ) dat; Long inside = (Long) out.get("inside"); total += inside; // aggregate answer from all the worker nodes. } public double getPi() { // returns value of pi based on the job done by all the workers double pi = (total / (random_samples * DoubleDataSize)) * 4; return pi; } public int getDataCount() { return random_samples; } Complete Monte Carlo pi code Note: No explicit message passing Computation

21. Bootstrap class RunMonteCarloPiModule.java package edu.uncc.grid.example.workpool; import java.io.IOException; import net.jxta.pipe.PipeID; import edu.uncc.grid.pgaf.Anchor; import edu.uncc.grid.pgaf.Operand; import edu.uncc.grid.pgaf.Seeds; import edu.uncc.grid.pgaf.p2p.Types; public class RunMonteCarloPiModule { public static void main(String[] args) { try { MonteCarloPiModule pi = new MonteCarloPiModule(); Seeds.start( args[0], false); PipeID id = Seeds.startPattern( new Operand( (String[])null, new Anchor( args[1], Types.DataFlowRoll.SINK_SOURCE), pi ) ); System.out.println(id.toString() ); Seeds.waitOnPattern(id); System.out.println( "The result is: " + pi.getPi() ) ; Seeds.stop(); } catch (SecurityException e) { … } Deploys framework and runs code (JXTA P2P version) 21

22. Multicore version of Seeds 22 public class RunMonteCarloPiModule { public static void main(String[] args) { try { MonteCarloPiModule pi=new MonteCarloPiModule(); Thread id = Seeds.startPatternMulticore( new Operand( (String[])null, new Anchor( args[0], Types.DataFlowRole.SINK_SOURCE), pi ), 4 ); id.join(); System.out.println( "The result is: " + pi.getPi() ) ; } catch (SecurityException e) { … }

23. Measuring Time Can instrument code in the bootstrap class: public class RunMyModule { public static void main (String [] args ) { try{ long start = System.currentTimeMillis(); MyModule m = new MyModule(); Seeds.start(. ); PipeID id = ( … ); Seeds.waitOnPattern(id); Seeds.stop(); long stop = System.currentTimeMillis(); double time = (double) (stop - start) / 1000.0; System.out.println(“Execution time = " + time); } catch (SecurityException e) { … … 23

24. Compiling/executing Can be done on the command line (ant script provided) or through an IDE (Eclipse) 24

25. Now to try the code. Turn to “Session 1 Hands-on notes:” Need first to transfer Seeds software and projects from a flash drive to your PC. 25