Prime Numbers Generating Threads in Java printing twice - java

I am trying to print prime numbers between one point to another, lets say from 1 to 1000 in one thread and 1000 to 2000 in another thread but when I print each thread using foreach loop it gives me an unordered Arraylist which is printed twice.
I am trying to print 1, 2, 3, 5, 7... using two concurrent threads. Please help me out so that I can better understand threading.
public class PrimeNumberGenerator implements Runnable{
protected long from, to;
static ArrayList<Long> primeList = new ArrayList<Long>();
public PrimeNumberGenerator(long from,long to)
{
this.from = from;
this.to = to;
}
public long count = 0;
public void run() {
for(long n=from; n<=to; n++){
boolean isPrime = true;
for(long i = 2; i<n; i++) {
if(n % i==0) {
isPrime = false;
break;
}
}
if(isPrime) {
count++;
primeList.add(n);
}
}
}
public ArrayList<Long> getPrimes() {
return primeList;
}
public static void main(String[] args) throws Exception {
// TODO Auto-generated method stub
PrimeNumberGenerator gen1 = new PrimeNumberGenerator(1L,1000L);
PrimeNumberGenerator gen2 = new PrimeNumberGenerator(1001L,2000L);
Thread t1 = new Thread(gen1);
Thread t2 = new Thread(gen2);
t1.start();
t2.start();
t1.join();
t2.join();
gen1.getPrimes().forEach(primeList -> System.out.println(primeList));
gen2.getPrimes().forEach(primeList -> System.out.println(primeList));
}
}

The problem is that you have two threads filling the same ArrayList at the same time, because your ArrayList is static (meaning there will only be one instand shared throughout the whole application)
The first thread may add three numbers, then the second thread adds three numbers and then the first again, resulting in an ArrayList that contains
[1, 2, 3, 1009, 1013, 1019, 5, 7, 11]
Then in the end you (correctly) wait for the Threads to finish and print the same (incorrectly ordered) ArrayList twice!
Just make your ArrayList non static and it will work, that way both PrimeNumberGenerator will have their own ArrayList!

I recommend using a TreeSet to keep the primes ordered. The TreeSet will need to be properly synchronized for multiple thread access.
public class PrimeNumberGenerator implements Runnable {
protected long from, to;
static Set<Long> primeList = new TreeSet<Long>();
public PrimeNumberGenerator(long from, long to)
{
this.from = from;
this.to = to;
}
public long count=0;
public void run() {
for(long n=from;n<=to;n++) {
boolean isPrime = true;
for(long i = 2; i<n; i++) {
if(n % i==0) {
isPrime = false;
break;
}
}
if(isPrime) {
count++;
synchronized(primeList) {
primList.add(n);
}
}
}
}
public static ArrayList<Long> getPrimes(){
//Make a copy so we don't need to synchronize outside of this class
return new ArrayList<>(primeList);
}
public static void main(String[] args) throws Exception {
// TODO Auto-generated method stub
PrimeNumberGenerator gen1 = new PrimeNumberGenerator(1L,1000L);
PrimeNumberGenerator gen2 = new PrimeNumberGenerator(1001L,2000L);
Thread t1 = new Thread(gen1);
Thread t2 = new Thread(gen2);
t1.start();
t2.start();
t1.join();
t2.join();
PrimeNumberGenerator.getPrimes().forEach(primeList -> System.out.println(primeList));
}
}

Related

How can I assign a value to the threadArray variable in Task1 , Task2, Task3 class run() method?

I have to insert the elements using three threads by creating three classes, namely Task1,Task2 and Task3. The values to be inserted into the array are 0,1,2,....299.
Override the run method in the threads. Three integer i,j, and k representing the number of elements each thread should append inside the given array.
Thread one should append 0 to i-1 inside the array,thread two should append i to i+j-1 inside the array,and the third thread should append i+j to 299 inide the array.
Threads one and two must run simultaneously, and the values of the threads one and two must be inserted inside the indices of the array from 0 to i+j-1 randomly.The third thread should start only after the first two threads have been executed completely.
In these code three task are given.
first task and second task start executing the thread at the same time and after completion of first two task then only third task start. If these situation getting correct then test() method return true.
public static final int[] threadArray = new int[300]; how I add random number into these array using Task1 Task2 and Task3 class.
Input :
80
130
90
Output :
True
import java.util.Scanner;
class Task1 extends Thread
{
static int a = 0;
static int beg = 0;
public void run()
{
for(int i=a;i<=beg;i++)
{
Solution.threadArray[i] = i;
try {
Thread.sleep(500);
} catch (InterruptedException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
}
}
class Task2 extends Thread
{
static int a = 0;
static int beg = 0;
#Override
public void run()
{
// TODO Auto-generated method stub
for(int i=a;i<=beg;i++)
{
Solution.threadArray[i] = i;
}
}
}
class Task3 extends Thread
{
static int a = 0;
static int beg = 0;
public void run()
{
// TODO Auto-generated method stub
for(int i=a;i<=beg;i++)
{
Solution.threadArray[i] = i;
}
}
}
public class Solution
{
public static final int[] threadArray = new int[300];
public static volatile String i = 0+"";
public boolean test() throws InterruptedException
{
Task1 task1 = new Task1();
Task2 task2 = new Task2();
Task3 task3 = new Task3();
Thread task2Thread = new Thread(task2);
Thread task3Thread = new Thread(task3);
task1.start();
task2Thread.start();
task1.join();
task2Thread.join();
task3Thread.start();
int first = Task1.a+Task2.a;
int containsSecondThread = Task1.a;
String oneAndTwo = "";
String sizeOfTask1 = "";
for(int i=0;i<first;i++)
{
oneAndTwo += threadArray[i]+" ";
}
for(int i=0;i<containsSecondThread;i++)
{
sizeOfTask1 += threadArray[i]+" ";
}
int begOfTask3 = Task3.beg;
String checkingString = "";
for(int i=begOfTask3;i<threadArray.length;i++)
{
checkingString += i + " ";
}
String task3String = "";
for(int j = begOfTask3;j<threadArray.length;j++)
{
task3String += threadArray[j]+" ";
}
if((!oneAndTwo.contains(begOfTask3+"") && sizeOfTask1.contains(Task2.beg+"")) || task3String.equals(checkingString))
{
return true;
}
return false;
}
public static void main(String[] args) throws InterruptedException
{
Scanner sc= new Scanner(System.in);
Solution solution = new Solution();
int one = sc.nextInt();
Task1.a = one;
Task1.beg = 0;
int two = sc.nextInt();
Task2.a = two;
Task2.beg = one;
int three = sc.nextInt();
Task3.a = three;
Task3.beg = one+two;
System.out.print(solution.test());
}
}
First, some observations regarding your code: Instead of using static variables in the classes (i.e., Task1, Task2, and Task3) that extend the class Thread (to understand why have a look at Why are static variables considered evil?):
static int a = 0;
static int beg = 0;
use non-static final fields, and initialize them via the constructor:
class Task1 extends Thread
{
private final int begin;
private final int end;
Task1(int begin, int end){
this.begin = begin;
this.end = end;
}
public void run(){
for(int i=begin; i<= end; i++)
....
}
}
adapt the main method accordingly:
public static void main(String[] args){
...
Task1 task1 = new Task1(begin, end);
}
and then pass the tasks-related objects as parameters of to the test method:
public boolean test(Task1 task1, Task2 task2, Task3 task3){
...
}
For the concatenation of the strings use StringBuilder:
StringBuilder oneAndTwo = new StringBuilder();
for(int i=0;i<first;i++)
{
oneAndTwo.append(threadArray[i]).append(" ");
}
This looks wrong:
Task1.a = one;
Task1.beg = 0;
by looking at the loop of the run method from Task1, this means that, if Task1.a is not a negative number, then Task1 will not do any work.
To use the threads to generate the random values of the array:
int[] threadArray = new int[300];
you can start by extracting a method to generate those random values, based on formula:
r.nextInt(high-low) + low;
this formula generates a random value between low and high.
Adapt the tasks, accordingly:
class Task1 extends Thread
{
private final Random random_values = new Random();
private final int low;
private final int high;
...
public int generate_random(){
return r.nextInt(high-low) + low;
}
public void run()
{
for(....)
{
Solution.threadArray[i] = generate_random();
...
}
}
}
Make sure to pass to the threads the information about the range of the random values to be generated (i.e., the low and high parameters), and the reference to the array that will be filled up with those random values (i.e., array int[] threadArray) . Also make sure that you split the iterations int[] threadArray among the threads. Therefore, each thread should generate a chunk of the random values. An example of such distribution would be:
Thread 1 : 0 to 100;
Thread 2 : 100 to 200;
Thread 3 : 200 to 300;
You can make this more robust and divide the array length by the number to threads and assign the work among threads, accordingly.
I could have provided you with the entire solution, but I feel that is better instead if I give you the pointers so that you can do it in your own.
EDIT: Based on the new edit of your question:
You just need to adapt the Task classes as follows:
class Task1 extends Thread {
static int a = 0;
static int beg = 0;
public void run(){
for(int i=beg;i < a;i++)
Solution.threadArray[i] = i;
}
}
class Task2 extends Thread {
static int a = 0;
static int beg = 0;
public void run(){
for(int i=beg; i< beg + a;i++)
Solution.threadArray[i] = i;
}
}
class Task3 extends Thread{
static int a = 0;
static int beg = 0;
public void run(){
for(int i=beg;i< a + beg;i++)
Solution.threadArray[i] = i;
}
}
Thread1 and Thread2 are supposed to access Common Resource in threadArray[0... Task1.a+Task2+a]. So we have to make use of static volatile variable i declared in Solution Class.
class Task1 extends Thread
{
static int a=0,beg=0;
public void run()
{
int k=Task1.beg;
int i1=0;
while(i1<Task1.a)
{
Solution.threadArray[Integer.parseInt(Solution.i)]=k++;
int a1=Integer.parseInt(Solution.i);
a1++;i1++;
Solution.i=a1+"";
try{
Thread.sleep(1);
}
catch(InterruptedException e){}
}
}
}
class Task2 extends Thread
{
static int a=0,beg=0;
public void run()
{
int y=0;
int k=Task2.beg;
while(y<Task2.a)
{
Solution.threadArray[Integer.parseInt(Solution.i)]=k++;
int a1=Integer.parseInt(Solution.i);
a1++;y++;
Solution.i=a1+"";
try{
Thread.sleep(1);
}
catch(InterruptedException e){}
}
}
}
Thread3 work independently after First 2 threads complete.
class Task3 extends Thread
{
static int beg=0,a=0;
public void run()
{
for(int i=Task3.beg;i<Task3.beg+Task3.a;i++)
{
Solution.threadArray[i]=i;
}
}
}

How to synchronise threads and preserve their execution order with CyclingBarrier?

I want to write a multithread app that prints characters from Strings one by one and after first "round" it would preserve order for the other rounds. It should work somehting like this:
For Strings:
private String[] strings = {"aaaa", "bb", "ccccccccccccc", "dddddd"};
It would print:
abcd abcd acd acd cd cd c c c c c c c
or maybe
dbac dbac dac dac dc dc c c c c c c c
depending on which proccess started first in the very first round
My solution so far looks like this
import java.util.concurrent.BrokenBarrierException;
import java.util.concurrent.CyclicBarrier;
public class Printer {
private CyclicBarrier cyclicBarrier;
private final static String one = "aaa";
private final static String two = "bbbb";
private final static String three = "c";
private final static String four = "dddddd";
public static void main(String[] args) {
Printer printer = new Printer();
printer.runSimulation(4);
}
private void runSimulation(int numberOfStrings) {
cyclicBarrier = new CyclicBarrier(numberOfStrings, new AggregatorThread());
Thread thread = new Thread(new PrintingThread(padSpaces(one, 10)));
Thread thread1 = new Thread(new PrintingThread(padSpaces(two, 10)));
Thread thread3 = new Thread(new PrintingThread(padSpaces(three, 10)));
Thread thread4 = new Thread(new PrintingThread(padSpaces(four, 10)));
thread.start();
thread1.start();
thread3.start();
thread4.start();
}
class AggregatorThread implements Runnable{
#Override
public void run() {
System.out.print(" ");
}
}
class PrintingThread implements Runnable{
private String toPrint;
private int iterator;
public PrintingThread(String toPrint) {
this.toPrint = toPrint;
this.iterator = 0;
}
#Override
public void run() {
while(iterator < toPrint.length()) {
System.out.print(toPrint.charAt(iterator));
iterator++;
try {
cyclicBarrier.await();
} catch (InterruptedException | BrokenBarrierException e) {
e.printStackTrace();
}
}
}
}
private String padSpaces(String inputString, int length) {
if (inputString.length() >= length) {
return inputString;
}
StringBuilder sb = new StringBuilder();
while (sb.length() < length - inputString.length()) {
sb.append(' ');
}
StringBuilder sb1 = new StringBuilder(inputString);
sb1.append(sb);
return sb1.toString();
}
}
But it doesn't preserve the order of letters written to the console and also i fill the Strings to the some hardcoded value right now, but i would want to make it work properly without equal strings.
Any suggestions on this?
Since you're asking for a solution with CyclicBarrier, here's a way you could do this with one... It definitely wouldn't be my first thought for how to solve the issue though (assuming the issue isn't 'do this with a CyclicBarrier'...).
Create a CyclicBarrier of length 4.
Assign each Thread a number (0 to 3) when it starts (using an AtomicInteger or otherwise).
Have each Thread do something like:
while (barrier.getNumberWaiting() != this.threadNumber) {
}
// Do your adding to the StringBuilder here...
barrier.await();
I.e. each Thread spins until the number of waiting parties is equal to that Thread's number.
Whichever is assigned 0 will always go through first, while all the others are stuck spinning. Once that Thread has done its StringBuilder thing, it will then await, which in turn frees the Thread assigned 1 to go through. The order will stay consistent after the number assignments.
To get the unique id per process, a simple AtomicInteger can be used.
private final AtomicInteger idCounter = new AtomicInteger();
private final CyclicBarrier barrier = new CyclicBarrier(4);
private final AtomicInteger doneCounter = new AtomicInteger();
public Runnable createRunnable() {
return () -> {
final int threadId = this.idCounter.getAndIncrement();
boolean threadDone = false;
boolean moreCharacters = true;
while (true) {
while (this.barrier.getNumberWaiting() != threadId) {
}
// Add to StringBuilder here...
// Set the 'moreCharacters' flag as false once this thread
// has handled its String.
// They will still need to spin though, to make sure the
// parties waiting keep adding up as appropriate.
if (!moreCharacters && !threadDone) {
// 'threadDone' used so that each thread only
// increments the 'doneCounter' once.
this.doneCounter.incrementAndGet();
threadDone = true;
}
barrier.await();
if (this.doneCounter.get() == 4) {
// Exit out of the loop once all Threads are done.
break;
}
}
};
}

Erastotenes Sieve in paraller programm Java

I would like to make a programm which count prime numbers using Erastotenes Sieve. In this issue I want to use semaphore to communicate between thread to make calculations on table with numbers.
So far I have written code like that.
public static void main( String[] args ) throws InterruptedException {
System.out.println("Podaj gorny zakres\n");
Scanner scanner = new Scanner(System.in);
Erastotenes erastotenes = new Erastotenes(Integer.parseInt(scanner.nextLine()));
erastotenes.initializeTable();
long start = System.nanoTime();
List<SingleProcess.MyThread> list = new ArrayList<>();
List<Integer> numbers = Dollar.$(2,erastotenes.getMaximumNumber()+1).toList();
for(int i=0;i<2;i++)
{
list.add(new SingleProcess.MyThread(erastotenes,numbers.subList((numbers.size()/2)*i,(numbers.size()/2)*i+numbers.size()/2)));
list.get(list.size()-1).start();
list.get(list.size()-1).join();
}
System.out.println(System.nanoTime() - start);
//System.out.println("Liczba elementów: "+erastotenes.countPrimeElements());
}
Erastotenes class.
public class Erastotenes {
private int upperRange;
private int maximumNumber;
private int table[];
public Erastotenes(int upperRange) {
this.upperRange = upperRange;
this.maximumNumber = (int)(Math.floor(Math.sqrt(upperRange)));
this.table = new int[upperRange+1];
}
public int getMaximumNumber() {
return maximumNumber;
}
public int getUpperRange() {
return upperRange;
}
public void initializeTable()
{
for(int i=1;i<=upperRange;i++) {
table[i] = i;
}
}
public void makeSelectionOfGivenNumber(int number)
{
if (table[number] != 0) {
int multiple;
multiple = number+number;
while (multiple<=upperRange) {
table[multiple] = 0;
multiple += number;
}
}
}
public List<Integer> getList()
{
List<Integer> list = Ints.asList(table);
return list.stream().filter(item->item.intValue()!=0 && item.intValue()!=1).collect(Collectors.toList());
}
}
The class describing single Thread to make calculations with static Semaphore looks like this.
public class SingleProcess {
static Semaphore semaphore = new Semaphore(1);
static class MyThread extends Thread {
Erastotenes erastotenes;
List<Integer> numbers;
MyThread(Erastotenes erastotenes,List<Integer> numbers) {
this.erastotenes = erastotenes;
this.numbers=numbers;
}
public void run() {
for(int number:numbers) {
try {
semaphore.acquire();
//1System.out.println(number + " : got the permit!");
erastotenes.makeSelectionOfGivenNumber(number);
} catch (InterruptedException e) {
e.printStackTrace();
} finally {
semaphore.release();
}
}
}
}
}
I thought that splitting on half table with numbers from 2 to maximum numbers as square root like in erastotrenes algorithm for these two Threads will boost calculations, but with upperRange to 100000000 the difference between paraller and sequence is not so big. How can I in another realize this problem of paraller programming Erastotenes Sieve?
I think your main problem is this:
for(int i=0;i<2;i++)
{
list.add(new SingleProcess.MyThread(erastotenes,numbers.subList((numbers.size()/2)*i,(numbers.size()/2)*i+numbers.size()/2)));
list.get(list.size()-1).start();
list.get(list.size()-1).join();
}
You start a thread and then immediately wait for it to finish; that kills the parallelism entirely. You can start and wait in the end:
for(int i=0;i<2;i++)
{
list.add(new SingleProcess.MyThread(erastotenes,numbers.subList((numbers.size()/2)*i,(numbers.size()/2)*i+numbers.size()/2)));
list.get(list.size()-1).start();
}
for (Thread t : list) {
t.join();
}
But, there's also a problem with your semaphore tbh. Each thread blocks all other threads from doing anything as long as it's working on a number; that means that again, all parallelism is gone.
You can do away with the semaphore altogether IMO; there's not really a lot of danger in setting the same index to 0 several times, which is all that happens in this "critical section" - but it's not critical at all because no one ever reads the array value in question before all threads are finished.

Return Value from Java Thread Class

So below is a simple Java class using multithreading, and my question is, is there a way for me to store the randomNumber from each thread (maybe in a variable called randomNumberOne or randomNumberTwo), so that I can use those to possibly get the sum of both and return it?
I know this example sounds stupid but basically with my real code I am returning a value from each of my threads and want to get the average of them. I haven't found any solution for returning values in threads in java (also I am new to multithreading completely).
public class Example {
public static void main(String[] args){
MathThread one = new MathThread();
MathThread two = new MathThread();
one.start();
two.start();
}
}
class MathThread extends Thread{
public MathThread(){
}
public void run(){
Random rand = new Random();
int randomNumber = rand.nextInt((100 - 1) + 1) + 1;
System.out.println(randomNumber);
}
Output
5
33
Add a result variable to your MathThread class and get the value after you join the thread:
class MathThread extends Thread
{
private int result;
public int getResult()
{
this.join();
return result;
}
public void run()
{
// ...
result = randomNumber;
}
}
one.start();
two.start();
double average = (one.getResult() + two.getResult()) / 2.0;
In Java 8 you can do
IntStream.of(0, 2).parallel()
.map(i -> new Random().nextInt(100)+1)
.forEach(System.out::println);
Without using the Stream API you can do
List<Future> futures = new ArrayList<>();
for (int i = 0; i < 2; i++)
futures.add(ForkJoinPool.commonPool()
.submit(new Callable<Integer>() {
public Integer call() {
return new Random().nextInt(100)+1;
}));
for(Future<Integer> future : futures)
System.out.printl(future.get());
Here is the simple snippet to customize:
// 1. Create and fill callables to execute
List<Callable<Integer>> callables = new LinkedList<>();
// callabels.addAll(makeMeCallables());
// 2. Run using Executor of your choice
ExecutorService service = Executors.newCachedThreadPool();
List<Future<Integer>> results = service.invokeAll(callables);
// 3. Get the results
if (results.get(i).isDone()) {
Future f = result.get(i);
// process f.get()
}

how to spwan recursive calls and for loop iterations in multiple threads in java (if possible)?

I was wondering if it is possible to spawn each loop iteration (each iteration into a thread by itself) and finally collect the result. consider this example, nothing fancy in it at all. Just a simple for-loop, but the idea is to extend on it. The body of for loop does not matter, I just filled up with some code. but basically assume it has some expensive calculation which will take few minutes to complete for each iteration. so I want to do each loop calculation in a separate thread.
public class Threadspawns {
private double[] arr = new double[4];
public void calculations(){
for (int i =2; i < 6; i++){
//expensive calculation
arr[i-2]=Math.pow(i,500);
}
}
public static void main(String[] args){
Threadspawns t = new Threadspawns();
long start = System.currentTimeMillis();
t.calculations();
long end = System.currentTimeMillis();
System.out.println(Arrays.toString(t.arr));
System.out.println("time taken "+ (end-start));
}
}
on the same note,
if it is possible to actually split the recursive calls into multiple threads and collect them as they return.
Example is that of fibonacci
public static int fibonacci(int n){
if (n==0)
return 0;
if (n==1)
return 1;
return fibonacci(n-2)+fibonacci(n-1);
}
It is possible that it cannot be done for fibonacci recurive method. but any other example of paralleling recursive calls between threads IF POSSIBLE would be nice to know.
P.S: I have basic knowledge of Thread and Runnable, but wondering if the above is doable
Solution to your first requirement that is moving the expensive calculation into a Callable task. Hope it find it useful:
import java.util.Arrays;
import java.util.concurrent.*;
public class Threadspawns {
private final int THREAD_COUNT = 8;
private final int CALCULATION_COUNT = 60000;
private double[] arr = new double[CALCULATION_COUNT];
public void calculations() {
ExecutorService executorService = Executors.newFixedThreadPool(THREAD_COUNT);
ExecutorCompletionService<Double> completionService = new ExecutorCompletionService<Double>(executorService);
for (int i = 2; i < CALCULATION_COUNT; i++) {
completionService.submit(new Calculation(i));
}
//Get from all Future tasks till all tasks completed
for (int i = 2; i < CALCULATION_COUNT; i++) {
try {
arr[i] = completionService.take().get();
} catch (InterruptedException e) {
e.printStackTrace(); //do something
} catch (ExecutionException e) {
e.printStackTrace(); //do something
}
}
}
private static class Calculation implements Callable<Double> {
private final int num;
private Calculation(int num) {
this.num = num;
}
#Override
public Double call() throws Exception {
return Math.pow(num, 500);
}
}
public static void main(String[] args) {
Threadspawns t = new Threadspawns();
long start = System.currentTimeMillis();
t.calculations();
long end = System.currentTimeMillis();
System.out.println(Arrays.toString(t.arr));
System.out.println("time taken " + (end - start));
}
}

Categories