We Keep Coding

Java BlockingQueue messages not processed not processed in full

I am trying to process few million records from a text file (i.e. reading the file sequentially using one thread, and trying to process the retrieved lines using multiple threads). A method call after 'queue.take();' is only executing for the number of times equal to initial capacity allocated to BlockingQueue (100 in this example), and then the process doesn't pickup anymore records.
Could you please help in debugging or identify the issue?
Main Method:
############
final int threadCount = 10;
BlockingQueue<String> queue = new ArrayBlockingQueue<String>(100);
ExecutorService service = Executors.newFixedThreadPool(threadCount);
for (int i = 0; i < (threadCount - 1); i++) {
service.submit(new EvaluateLine(queue));
}
service.submit(new ProcessFile(queue)).get();
service.shutdownNow();
service.awaitTermination(365, TimeUnit.DAYS);
EvaluateLine:
#############
private final BlockingQueue<String> queue;
public EvaluateLine(BlockingQueue<String> queue){
this.queue = queue;
}
#Override
public void run() {
String line;
while(true) {
try {
line = queue.take();
SyncOutput.Process(line);
} catch (InterruptedException ex) {
break;
}
}
}
ProcessFile:
############
private final BlockingQueue<String> queue;
public ProcessFile(BlockingQueue<String> queue) {
this.queue = queue;
}
#Override
public void run() {
Path path = Paths.get("C:\\Desktop\\testdata.txt");
BufferedReader br = null;
try {
br =Files.newBufferedReader(path, StandardCharsets.UTF_8);
for (String line; (line = br.readLine()) != null; ) {
queue.put(line);
}
} catch(IOException e){
e.printStackTrace();
} catch(InterruptedException e){
e.printStackTrace();
} finally{
try {
if (br != null) br.close();
}catch(IOException e){
e.printStackTrace();
}
}
Edit:
SyncOutput
##########
class SyncOutput{
public static void ProcessExpression(String inputLine) {
evalExpression(inputLine);
}
public static double evalExpression(String s){
Expression e = new ExpressionBuilder(s)
.build();
return e.evaluate();
}
}

The following code segment(using Expression evaluation library: exp4j) is blocking the multi-thread execution, not sure why. But I have replaced this code block with a different library(parsii) and now everything looks fine.
public static void ProcessExpression(String inputLine) {
evalExpression(inputLine);
}
public static double evalExpression(String s){
Expression e = new ExpressionBuilder(s)
.build();
return e.evaluate();
}

Why does multithreaded version take the same amount of time as single threaded version?

I have the following work queue implementation, which I use to limit the number of threads in use. It works by me initially adding a number of Runnable objects to the queue, and when I am ready to begin, I run "begin()". At this point I do not add any more to the queue.
public class WorkQueue {
private final int nThreads;
private final PoolWorker[] threads;
private final LinkedList queue;
Integer runCounter;
boolean hasBegun;
public WorkQueue(int nThreads) {
runCounter = 0;
this.nThreads = nThreads;
queue = new LinkedList();
threads = new PoolWorker[nThreads];
hasBegun = false;
for (int i = 0; i < nThreads; i++) {
threads[i] = new PoolWorker();
threads[i].start();
}
}
public boolean isQueueEmpty() {
synchronized (queue) {
if (queue.isEmpty() && runCounter == 0) {
return true;
} else {
return false;
}
}
}
public void begin() {
hasBegun = true;
synchronized (queue) {
queue.notify();
}
}
public void add(Runnable r) {
if (!hasBegun) {
synchronized (queue) {
queue.addLast(r);
runCounter++;
}
} else {
System.out.println("has begun executing. Cannot add more jobs ");
}
}
private class PoolWorker extends Thread {
public void run() {
Runnable r;
while (true) {
synchronized (queue) {
while (queue.isEmpty()) {
try {
queue.wait();
} catch (InterruptedException ignored) {
}
}
r = (Runnable) queue.removeFirst();
}
// If we don't catch RuntimeException,
// the pool could leak threads
try {
r.run();
synchronized (runCounter) {
runCounter--;
}
} catch (RuntimeException e) {
// You might want to log something here
}
}
}
}
}
This is a runnable I use to keep track of when all the jobs on the work queue have finished:
public class QueueWatcher implements Runnable {
private Thread t;
private String threadName;
private WorkQueue wq;
public QueueWatcher(WorkQueue wq) {
this.threadName = "QueueWatcher";
this.wq = wq;
}
#Override
public void run() {
while (true) {
if (wq.isQueueEmpty()) {
java.util.Date date = new java.util.Date();
System.out.println("Finishing and quiting at:" + date.toString());
System.exit(0);
break;
} else {
try {
Thread.sleep(1000);
} catch (InterruptedException ex) {
Logger.getLogger(PlaneGenerator.class.getName()).log(Level.SEVERE, null, ex);
}
}
}
}
public void start() {
wq.begin();
System.out.println("Starting " + threadName);
if (t == null) {
t = new Thread(this, threadName);
t.setDaemon(false);
t.start();
}
}
}
This is how I use them:
Workqueue wq = new WorkQueue(9); //Get same results regardless of 1,2,3,8,9
QueueWatcher qw = new QueueWatcher(wq);
SomeRunnable1 sm1 = new SomeRunnable1();
SomeRunnable2 sm2 = new SomeRunnable2();
SomeRunnable3 sm3 = new SomeRunnable3();
SomeRunnable4 sm4 = new SomeRunnable4();
SomeRunnable5 sm5 = new SomeRunnable5();
wq.add(sm1);
wq.add(sm2);
wq.add(sm3);
wq.add(sm4);
wq.add(sm5);
qw.start();
But regardless of how many threads I use, the result is always the same - it always takes about 1m 10seconds to complete. This is about the same as when I just did a single threaded version (when everything ran in main()).
If I set wq to (1,2,3--9) threads it is always between 1m8s-1m10s. What is the problem ? The jobs (someRunnable) have nothing to do with each other and cannot block each other.
EDIT: Each of the runnables just read some image files from the filesystems and create new files in a separate directory. The new directory eventually contains about 400 output files.
EDIT: It seems that only one thread is always doing work. I made the following changes:
I let the Woolworker store an Id
PoolWorker(int id){
this.threadId = id;
}
Before running I print the id of the worker.
System.out.println(this.threadId + " got new task");
r.run();
In WorkQueue constructor when creating the poolworkers I do:
for (int i = 0; i < nThreads; i++) {
threads[i] = new PoolWorker(i);
threads[i].start();
}
But it seems that that only thread 0 does any work, as the output is always:
0 got new task

Use queue.notifyAll() to start processing.
Currently you're using queue.notify(), which will only wake a single thread. (The big clue that pointed me to this was when you mentioned only a single thread was running.)
Also, synchronizing on Integer runCounter isn't doing what you think it's doing - runCounter++ is actually assigning a new value to the Integer each time, so you're synchronizing on a lot of different Integer objects.
On a side note, using raw threads and wait/notify paradigms is complicated and error-prone even for the best programmers - it's why Java introduced the java.util.concurrent package, which provide threadsafe BlockingQueue implementations and Executors for easily managing multithreaded apps.

Java looping Threads using CyclicBarrier

I have a program with this general structure:
init
create CyclicBarrier
initialise all threads, attaching to barrier
*start all threads*
wait for join
display stats
*start all threads*
perform calculation
await barrier
My problem is I need the threads' run() method to keep looping until a certain condition is met, but pausing after every iteration to let all threads synchronise.
I've already tried attaching a Runnable method to the barrier, but this ends up requiring the recreation and restarting of each thread, which isn't a very good solution.
I've also tried using the CyclicBarrier's reset() method, but this just seems to cause errors on the existing threads, even when executed after all threads have completed.
My question is:
-Is it possible to 'reset' a barrier and have all the barrier's threads follow the same conditions as they did before the first invocations of await()?
-Or is there another method I should be using to achieve this?
Thanks in advance

The barrier.wait() will suspend the threads. The barrier is already in the main thread, it does not need another. In your algorithm above you show the threads being restarted after displaying stats. You should not need to do this. If the recently awakened threads are in a loop they will go back into the barrier.wait() again.

Following #Totoro's answer, below is a little bit of example code which also incorporates the requirement "I need the threads' run() method to keep looping until a certain condition is met, pausing after every iteration to let all threads synchronise". That makes it complex pretty quick, but hopefully the program output will clarify the example code (or I should just make better examples).
import java.util.concurrent.CyclicBarrier;
import java.util.concurrent.atomic.AtomicBoolean;
import java.util.concurrent.atomic.AtomicInteger;
public class BarrierCalc implements Runnable {
public static final int CALC_THREADS = 3;
private static final AtomicBoolean runCondition = new AtomicBoolean();
private static final AtomicBoolean stopRunning = new AtomicBoolean();
public static void main(String[] args) {
CyclicBarrier barrier = new CyclicBarrier(CALC_THREADS + 1);
for (int i = 0; i < CALC_THREADS; i++) {
new Thread(new BarrierCalc(barrier)).start();
}
try {
runCondition.set(true);
barrier.await();
showln(0, "STATS!");
barrier.await();
showln(0, "start looping 1");
Thread.sleep(200);
runCondition.set(false);
showln(0, "stop looping 1");
barrier.await();
runCondition.set(true);
barrier.await();
showln(0, "start looping 2");
Thread.sleep(100);
runCondition.set(false);
showln(0, "stop looping 2");
barrier.await();
stopRunning.set(true);
showln(0, "finishing");
barrier.await();
} catch (Exception e) {
e.printStackTrace();
}
}
private static final AtomicInteger calcId = new AtomicInteger();
private CyclicBarrier barrier;
private int id;
public BarrierCalc(CyclicBarrier barrier) {
this.barrier = barrier;
id = calcId.incrementAndGet();
}
public void run() {
showln(id, "waiting for start");
try {
barrier.await(); // display stats
barrier.await(); // start running
int loopNumber = 0;
while (!stopRunning.get()) {
showln(id, "looping " + (++loopNumber));
while (runCondition.get()) {
Thread.sleep(10); // simulate looping
}
showln(id, "synchronizing " + loopNumber);
barrier.await();
showln(id, "synchronized " + loopNumber);
// give main thread a chance to set stopCondition and runCondition
barrier.await();
}
showln(id, "finished");
} catch (Exception e) {
e.printStackTrace();
}
}
private static final long START_TIME = System.currentTimeMillis();
public static void showln(int id, String msg) {
System.out.println((System.currentTimeMillis() - START_TIME) + "\t ID " + id + ": " + msg);
}
}
Keep in mind that program output might not be in the order expected: threads that are writing at the same time to one synchronized output (System.out) are given write-access in random order.

You can take a look at my example where I played with CyclicBarrier.Here each worker makes some calculation and at the barrier the condition is checked. If it meets the condition than all workers stop calculations, otherwise they continue:
class Solver {
private static final int REQUIRED_AMOUNT = 100;
private static final int NUMBER_OF_THREADS = 4;
AtomicInteger atomicInteger = new AtomicInteger();
AtomicBoolean continueCalculation = new AtomicBoolean(true);
final CyclicBarrier barrier;
public static void main(String[] args) {
new Solver();
}
class Worker implements Runnable {
int workerId;
Worker(int workerId) {
this.workerId = workerId;
}
public void run() {
try {
while(continueCalculation.get()) {
calculate(workerId);
barrier.await();
}
} catch (Exception ex) {
System.out.println("Finishing " + workerId);
}
}
}
public Solver() {
Runnable barrierAction = () -> {
if (done()) {
continueCalculation.set(false);
}
};
barrier = new CyclicBarrier(NUMBER_OF_THREADS, barrierAction);
List<Thread> threads = new ArrayList(NUMBER_OF_THREADS);
for (int i = 0; i < NUMBER_OF_THREADS; i++) {
Thread thread = new Thread(new Worker(i));
threads.add(thread);
thread.start();
}
}
private void calculate(int workerId) throws InterruptedException {
// Some long-running calculation
Thread.sleep(2000L);
int r = new Random().nextInt(12);
System.out.println("Worker #" + workerId + " added " + r +" = " + atomicInteger.addAndGet(r));
}
private boolean done() {
int currentResult = atomicInteger.get();
boolean collected = currentResult >= REQUIRED_AMOUNT;
System.out.println("=======================================================");
System.out.println("Checking state at the barrier: " + currentResult);
if (collected) {
System.out.println("Required result is reached");
}
System.out.println("=======================================================");
return collected;
}
}

Tracking Executing Threads

I am trying to figure out how I can track all the threads that my application is spawning. Initially, I thought I had it figured out using a CyclicBarrier, however I am seeing threads executing after my await call.
Below is the working pseudo code:
public class ThreadTesterRunner {
public static void main(String[] args) throws InterruptedException {
final CyclicBarrier cb = new CyclicBarrier(1);
ThreadRunner tr = new ThreadRunner(cb);
Thread t = new Thread(tr, "Thread Runner");
t.start();
boolean process = true;
// wait until all threads process, then print reports
while (process){
if(tr.getIsFinished()){
System.out.println("Print metrics");
process = false;
}
Thread.sleep(1000);
}
}
}
class ThreadRunner implements Runnable {
static int timeOutTime = 2;
private ExecutorService executorService = Executors.newFixedThreadPool(10);
private final CyclicBarrier barrier;
private boolean isFinished=false;
public ThreadRunner(CyclicBarrier cb) {
this.barrier = cb;
}
public void run(){
try {
boolean stillLoop = true; int i = 0;
while (stillLoop){
int size;
Future<Integer> future = null;
try {
future = executorService.submit(new Reader()); // sleeps
size = future.get();
} catch (InterruptedException | ExecutionException ex) {
// handle Errs
}
if(i == 3){
stillLoop = false;
this.barrier.await();
this.isFinished=true;
}
//System.out.println("i = "+i+" Size is: "+size+"\r");
i++;
}
} catch (InterruptedException | BrokenBarrierException e1) {
e1.printStackTrace();
}
}
public boolean getIsFinished(){
return this.isFinished;
}
}
class Reader implements Callable {
private ExecutorService executorService = Executors.newFixedThreadPool(1);
#Override
public Object call() throws Exception {
System.out.println("Reading...");
Thread.sleep(2000);
executorService.submit(new Writer());
return 1000;
}
}
class Writer implements Callable {
#Override
public Void call() throws Exception {
Thread.sleep(4000);
System.out.println("Wrote");
return null;
}
}
Can anyone suggest a way to ONLY print "print metrics" after all threads have run?

It doesn't seem like you're doing anything to coordinate with your Reader and Writer threads, which are the ones you want to wait for. If you pass your synchronization barrier through to those threads so that they can register and signal when they are done, it works just fine.
Here's a version rewritten to do so, using a Phaser instead of a CyclicBarrier. Note that each Reader and Writer registers itself upon construction, and notifies the synchronization barrier when it is done executing:
public class ThreadTesterRunner {
public static void main(String[] args) throws InterruptedException {
final Phaser cb = new Phaser();
ThreadRunner tr = new ThreadRunner(cb);
Thread t = new Thread(tr, "Thread Runner");
t.start();
boolean process = true;
// wait until all threads process, then print reports
while (process){
if(tr.getIsFinished()){
System.out.println("Print metrics");
process = false;
}
//else {
// System.out.println("Waiting: registered=" + cb.getRegisteredParties() + ", arrived=" + cb.getArrivedParties() + ", unarrived=" + cb.getUnarrivedParties());
//}
Thread.sleep(1000);
}
}
}
class ThreadRunner implements Runnable {
static int timeOutTime = 2;
private ExecutorService executorService = Executors.newFixedThreadPool(10);
private final Phaser barrier;
private boolean isFinished=false;
public ThreadRunner(Phaser phaser) {
this.barrier = phaser;
}
public void run(){
try {
boolean stillLoop = true; int i = 0;
while (stillLoop){
int size;
Future<Integer> future = null;
try {
future = executorService.submit(new Reader(this.barrier)); // sleeps
size = future.get();
} catch (InterruptedException | ExecutionException ex) {
// handle Errs
}
if(i == 3){
stillLoop = false;
this.barrier.awaitAdvance(0);
this.isFinished=true;
}
//System.out.println("i = "+i+" Size is: "+size+"\r");
i++;
}
} catch (Exception e1) {
e1.printStackTrace();
}
}
public boolean getIsFinished(){
return this.isFinished;
}
}
class Reader implements Callable {
private Phaser barrier;
private ExecutorService executorService = Executors.newFixedThreadPool(1);
public Reader(Phaser phase) {
phase.register();
this.barrier = phase;
}
#Override
public Object call() throws Exception {
System.out.println("Reading...");
Thread.sleep(2000);
executorService.submit(new Writer(this.barrier));
this.barrier.arrive();
return 1000;
}
}
class Writer implements Callable {
private Phaser barrier;
public Writer(Phaser phase) {
phase.register();
this.barrier = phase;
}
#Override
public Void call() throws Exception {
Thread.sleep(4000);
System.out.println("Wrote");
this.barrier.arrive();
return null;
}
}

From what I can see you aren't waiting for the Writer to finish in the Reader. Is that the problem you are seeing?
You are also accessing isFinished from more than one thread without synchronization (which however, merely may delay the termination of the loop in this situation).
I don't see CyclicBarrier doing anything.
Not sure what you are trying to do, but I'd think about how simpler I can make it. For example, can Reader and Writer be combined into one task? Then, waiting for them to finish would merely be:
executorService.invokeAll(tasks);
System.out.println("Print metrics");
where tasks is a collection of tasks (see also this javadoc)

Text is not getting printed once the Threads are done [duplicate]

This question already has answers here:
How to wait for all threads to finish, using ExecutorService?
(27 answers)
Closed 8 years ago.
Please have a look at the following code.
public class BigFileWholeProcessor {
private static final int NUMBER_OF_THREADS = 2;
public void processFile(String fileName) {
BlockingQueue<String> fileContent = new LinkedBlockingQueue<String>();
BigFileReader bigFileReader = new BigFileReader(fileName, fileContent);
BigFileProcessor bigFileProcessor = new BigFileProcessor(fileContent);
ExecutorService es = Executors.newFixedThreadPool(NUMBER_OF_THREADS);
es.execute(bigFileReader);
es.execute(bigFileProcessor);
es.shutdown();
if(es.isTerminated())
{
System.out.println("Completed Work");
}
}
}
public class BigFileReader implements Runnable {
private final String fileName;
int a = 0;
public static final String SENTINEL = "SENTINEL";
private final BlockingQueue<String> linesRead;
public BigFileReader(String fileName, BlockingQueue<String> linesRead) {
this.fileName = fileName;
this.linesRead = linesRead;
}
#Override
public void run() {
try {
//since it is a sample, I avoid the manage of how many lines you have read
//and that stuff, but it should not be complicated to accomplish
BufferedReader br = new BufferedReader(new FileReader(new File("E:/Amazon HashFile/Hash.txt")));
String str = "";
while((str=br.readLine())!=null)
{
linesRead.put(str);
System.out.println(a);
a++;
}
linesRead.put(SENTINEL);
} catch (Exception ex) {
ex.printStackTrace();
}
System.out.println("Completed");
}
}
public class BigFileProcessor implements Runnable {
private final BlockingQueue<String> linesToProcess;
public BigFileProcessor (BlockingQueue<String> linesToProcess) {
this.linesToProcess = linesToProcess;
}
#Override
public void run() {
String line = "";
try {
while ( (line = linesToProcess.take()) != null) {
//do what you want/need to process this line...
if(line==BigFileReader.SENTINEL)
{
break;
}
String [] pieces = line.split("(...)/g");
}
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
I want to print the text "completed work" in BigFileWholeProcessor once all the thread work is done. But instead, it is not getting printed. Why is this? How to identify that all the threads are done and need printing?

shutdown() only signal ES to shutdown, you need
awaitTermination(long timeout, TimeUnit unit)
before print message

Use submit() method instead of execute(). The get() method can be used if you want to wait for the thread to finish at any point of time. Read documentation on use of Future object for further details.
ExecutorService es = Executors.newFixedThreadPool(2);
Future<?> f = es.submit(new Thread(new TestRun()));
f.get(); // Wait for result... (i.e similar to `join()` in this case)
es.shutdown(); // Shutdown ExecutorService
System.out.println("Done.");
I have defined a TestRun class implementing Runnable, not shown here. The Future object makes more sense in other scenarios.