We Keep Coding

run threads according to the time limit

I want to start max 40 http requests each second and after 1 second, I want it to run another 40 from its own queue(like threadpooltaskexecutor's blocking queue). I am looking for an executor or thread pool implementation for this requirement.
Any recommendations?
Thx
Ali
EDIT: Fix rate is not efficient for the obvious reasons. As the queue items start one by one, the ones on the back of the queue will be just started but ones that has been started for a while may be finished.
Extra EDIT: The problem is to call only 40 request in a second, not have max 40 active. It can be 80 at other second but in 1 second there should only 40 newly created connections.

One way to do this is to use another architecture, it will make the process that much easiser.
1) Create a Thread class that implements the runnable.
2) It takes as parameters a list<>() of http requests that you want to make
3) Make the run() function loop the entire list (size 40)
4) Let the thread live for one second.
Here is a sample example:
class MyClass extends Thread
private ArrayList<...> theList;
public MyClass(ArrayList<..> theList){
this.theList = theList;
}
public void run(){
//Here, you simply want to loop for the entier list (max 40)
for(Req r: theList){
r.sendRequest()
)
}
public statc void main(String args[]){
//Create an instance of your thread:
MyClass t = new MyClass(ReqList<..>());
//Now that you have your thread, simply do the following:
while(true){
t = new MyClass( (insert your new list));
t.start();
try{
Thread.sleep(1000);
}catch(Exception e){
}
)
}
}
And there you have it

First define a class that implements Callable which will do your thread's treatment :
class MyClass implements Callable<String>
{
/**
* Consider this as a new Thread.
*/
#Override
public String call()
{
//Treatment...
return "OK"; //Return whatever the thread's result you want to be then you can access it and do the desired treatment.
}
}
Next step is to create an ExecutorService in my example, a Thread pool and throw in some tasks.
int nbThreadToStart = 40;
ExecutorService executor = Executors.newFixedThreadPool(/* Your thread pool limit */);
List<Future<String>> allTasks = new ArrayList<Future<String>>(/* Specify a number here if you want to limit your thread pool */);
for(int i = 0; i < 10; i++)//Number of iteration you want
{
for(int i = 0; i < nbThreadToStart; i++)
{
try
{
allTasks.add(executor.submit(new MyClass()));
}
catch(Exception e)
{
e.printStackTrace();
}
}
try
{
Thread.sleep(1000);
}
catch(Exception e)
{
e.printStackTrace();
}
}
executor.shutdown();
//You can then access all your thread(Tasks) and see if they terminated and even add a timeout :
try
{
for(Future<String> task : allTasks)
task.get(60, TimeUnit.SECONDS);//Timeout of 1 seconds. The get will return what you specified in the call method.
}
catch (TimeOutException te)
{
...
}
catch(InterruptedException ie)
{
...
}
catch(ExecutionException ee)
{
...
}
I'm not sure what you really want, but I think you should handle multi-threading with a thread pool specially if you're planning on receiving a lot of requests to avoid any undesired memory leak etc.
If my example is not clear enough, note that there is many other methods offered by ExexutorService,Future etc. that are very usefull when dealing with Thread.
Check this out :
http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/Executors.html
http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/Executor.html
http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/Future.html
That's it for my recommandations.

How should I interrupt and dispose of a group of threads in java?

I'm writing a game engine which performs alhpa-beta search on at a game state, and I'm trying to parallelize it. What I have so far is working at first, and then it seems to slow to a halt. I suspect that this is because I'm not correctly disposing of my threads.
When playing against the computer, the game calls on the getMove() function of a MultiThreadedComputerPlayer object. Here is the code for that method:
public void getMove(){
int n = board.legalMoves.size();
threadList = new ArrayList<WeightedMultiThread>();
moveEvals = new HashMap<Tuple, Integer>();
// Whenever a thread finishes its work at a given depth, it awaits() the other threads
// When all threads are finished, the move evaluations are updated and the threads continue their work.
CyclicBarrier barrier = new CyclicBarrier(n, new Runnable(){
public void run() {
for(WeightedMultiThread t : threadList){
moveEvals.put(t.move, t.eval);
}
}
});
// Prepare and start the threads
for (Tuple move : board.legalMoves) {
MCBoard nextBoard = board.clone();
nextBoard.move(move);
threadList.add(new WeightedMultiThread(nextBoard, weights, barrier));
moveEvals.put(move, 0);
}
for (WeightedMultiThread t : threadList) {t.start();}
// Let the threads run for the maximum amount of time per move
try {
Thread.sleep(timePerMove);
} catch (InterruptedException e) {System.out.println(e);}
for (WeightedMultiThread t : threadList) {
t.stop();
}
// Play the best move
Integer best = infHolder.MIN;
Tuple nextMove = board.legalMoves.get(0);
for (Tuple m : board.legalMoves) {
if (moveEvals.get(m) > best) {
best = moveEvals.get(m);
nextMove = m;
}
}
System.out.println(nextMove + " is the choice of " + name + " given evals:");
for (WeightedMultiThread t : threadList) {
System.out.println(t);
}
board.move(nextMove);
}
And here run() method of the threads in question:
public void run() {
startTime = System.currentTimeMillis();
while(true) {
int nextEval = alphabeta(0, infHolder.MIN, infHolder.MAX);
try{barrier.await();} catch (Exception e) {}
eval = nextEval;
depth += 1;
}
}
I need to be able to interrupt all the threads when time is up-- how am I supposed to implement this? As of now I'm constantly catching (and ignoring) InterruptedExceptions.

Thread.stop was deprecated for a reason. When you interrupt a thread in the middle, the thread doesn't have the chance to properly release resources it was using, and doesn't notify other threads of its completion...something that's very important in multi-threaded apps. I'm not surprised your performance tanks; I would be willing to bet your memory usage shoots through the roof. You also don't recycle the threads, you start and stop them without creating new objects, which means whatever broken state the variables were left in is probably still plaguing them.
A better way is to set a flag that tells the thread it should return. So include in your WeightedMultiThread class a boolean named something like shouldQuit, and set it to false every time start() is called. Then, instead of while (true) do while (!shouldQuit), and instead of t.stop(), use t.shouldQuit = true. After you do that to every thread, have another loop that checks each thread for t.isAlive(), and once every thread has returned, go about your business. You should have much better results that way.

This looks like an ideal place to use an ExecutorService. You can create Callable instances that implement the parallel tasks, submit them to the ExecutorService, then use awaitTermination to enforce a timeout.
For example:
public void getMove() {
ExecutorService service = Executors.newFixedThreadPool(board.legalMoves.size());
List<Future<Something>> futures = new ArrayList<Future<Something>>(board.legalMoves.size());
for (Tuple move : board.legalMoves) {
futures.add(service.submit(new WeightedMultiThread(...)));
}
service.awaitTermination(timePerMove, TimeUnit.MILLISECONDS);
service.shutdownNow(); // Terminate all still-running jobs
for (Future<Something> future : futures) {
if (future.isDone()) {
Something something = future.get();
// Add best move logic here
}
}
...
}
Replace Something with something that encapsulates information about the move that has been evaluated. I'd suggest Something be a class that holds the Tuple and its associated score. Your WeightedMultiThread class can do something like this:
class WeightedMultiThread implements Callable<Something> {
public Something call() {
// Compute score
...
// Return an appropriate data structure
return new Something(tuple, score);
}
}
Even better would be to create the ExecutorService once and re-use it for each call to getMove. Creating threads is expensive, so best to only do it once if you can. If you take this approach then you should not call shutdownNow, but instead use the Future.cancel method to terminate jobs that have not completed in time. Make sure your WeightedMultiThread implementation checks for thread interruption and throws an InterruptedException. That's usually a good way to write a long-running task that needs to be interruptible.
EDIT:
Since you're doing a level-by-level exploration of the game space, I'd suggest that you encode that in the getMove function rather than in the Tuple evaluation code, e.g.
public Tuple getMove() {
ExecutorService service = ...
Tuple best = null;
long timeRemaining = MAX_TIME;
for (int depth = 0; depth < MAX_DEPTH && timeRemaining > 0; ++depth) {
long start = System.currentTimeMillis();
best = evaluateMoves(depth, service, timeRemaining);
long end = System.currentTimeMillis();
timeRemaining -= (end - start);
}
return best;
}
private Tuple evaluateMoves(int depth, ExecutorService service, long timeRemaining) {
List<Future<Whatever>> futures = service.submit(...); // Create all jobs at this depth
service.awaitTermination(timeRemaining, TimeUnit.MILLISECONDS);
// Find best move
...
return best;
}
That could probably be cleaner, but you get the idea.

The most sensitive way is to use interruption mechanism. Thread.interrupt() and Thread.isInterrupted() methods. This ensures your message will be delivered to a thread even if it sits inside a blocking call (remember some methods declare throwing InterruptedException?)
P.S. It would be useful to read Brian Goetz's "Java Concurrency in Practice" Chapter 7: Cancellation and Shutdown.

Stop thread anytime i want

I have a method, wich supposed to interrupt a thread, but it's not. Do I need to always check the thread interrupted in the while method to stop the thread? How can I just terminate the thread at anytime?
solverTh = new Thread(new Runnable() {
#Override
public void run() {
while(somethingistrue){
//do lot of stuff here for long time
}
}
});
solverTh.start();
}
public void terminate(){
if(solverTh != null){
solverTh.interrupt();
}
}
okay than I thought the "lot of stuff" is irrelevant, but I will post it than. It makes openGL operations, I added the boolean variable "terminated" to the code it works now, I just wanted to find a nicer solution:
(glc is a GLCanvas, and the rotmultiplecube method rotates 3 objects)
Anyways I've solved the problem now, thanks for the answers.
terminated = false;
try {
Thread.sleep(2000);
} catch (InterruptedException ex) {
Logger.getLogger(BruteForce.class.getName()).log(Level.SEVERE, null, ex);
}
int colorToBeSolved = Statics.RED_BLUE_TABLE[stateToBeSolved];
System.out.println(stateToBeSolved + "," + colorToBeSolved);
if(entities[0].getColor() != colorToBeSolved){
if(terminated) return;
fullRotate(Statics.FIRST_ROW, Statics.DOWN);
}
if(entities[1].getColor() != colorToBeSolved){
if(terminated) return;
fullRotate(Statics.SECOND_COL, Statics.RIGHT);
}
if(entities[2].getColor() != colorToBeSolved){
if(terminated) return;
fullRotate(Statics.THIRD_COL, Statics.RIGHT);
}
if(entities[3].getColor() != colorToBeSolved){
if(terminated) return;
fullRotate(Statics.SECOND_ROW, Statics.DOWN);
}
if(entities[6].getColor() != colorToBeSolved){
if(terminated) return;
fullRotate(Statics.THIDR_ROW, Statics.DOWN);
}
for(int i = 0; i < 9; ++i){
int col = i % 3;
int row = 3 + i/3;
while(entities[i].getState() != stateToBeSolved){
for(int j = 0;j < 2; ++j){
if(entities[i].getState() != stateToBeSolved){
if(terminated) return;
fullRotate(col, Statics.LEFT);
if(terminated) return;
fullRotate(row, Statics.UP);
if(terminated) return;
fullRotate(col, Statics.RIGHT);
if(terminated) return;
fullRotate(row, Statics.DOWN);
}
}
for(int j = 0;j < 2; ++j){
if(entities[i].getState() != stateToBeSolved){
if(terminated) return;
fullRotate(col, Statics.RIGHT);
if(terminated) return;
fullRotate(row, Statics.UP);
if(terminated) return;
fullRotate(col, Statics.LEFT);
if(terminated) return;
fullRotate(row, Statics.DOWN);
}
}
}
}
}
and the fullrotate method:
private void fullRotate(int selectionIndex, int direction){
for(int i = 0; i < 9; ++i){
glc.rotMultipleCubeSlow(selectionIndex, direction);
try {
Thread.sleep(20);
} catch (InterruptedException ex) {
terminate();
}
}
glc.setMovesText(selectionIndex, direction);
glc.setMultipleStateAndColorsByTable(selectionIndex, direction);
glc.isEntitiesRight();
}

while(somethingistrue !Thread.currentThread().isInterrupted()){
//do lot of stuff here for long time
}
Does not have to work for blocking IO. Use dirty tricks: override Thread.interrupt() close IO object, cause IOException that if properly handled may end thread run method.

The elegant solution is to modify your fullRotate() method to throw InterruptedException.
private void fullRotate(int selectionIndex, int direction)
throws InterruptedException{
for(int i = 0; i < 9; ++i){
glc.rotMultipleCubeSlow(selectionIndex, direction);
Thread.yield();
}
glc.setMovesText(selectionIndex, direction);
glc.setMultipleStateAndColorsByTable(selectionIndex, direction);
glc.isEntitiesRight();
}
When you call Thread.interrupt() you cause InterruptedException when any of the methods that throw it is invoked, in your case the Thread.sleep() or Thread.yield(). This means that the best approach is to use it to actually interrupt the calculation.
You still need to check Thread.currentThread().isInterrupted() if you want immediate response to your Thread.interrupt()
You can ether remove if(terminated) return; or substitute it with Thread.currentThread().isInterrupted() check. Removing will be fine because the Thread.sleep(20)/Thread.yield() from fullRotate() will throw the InterruptedException. Also code will be cleaner without all these if(terminated) all over the place.
Use Thread.yield() instead for Thread.sleep(20). Obviously you don't want to sleep, because you put 20 millis. 20 milis is very close to the context switch time quantum. The thread will ether sleep more, or less. You don't want it to sleep more without any reason, so use yield().
Your thread run() then becomes:
solverTh = new Thread(new Runnable() {
#Override
public void run() {
while(somethingistrue &&
!Thread.currentThread().isInterrupted()) {
try {
//do lot of stuff here for long time
} catch (InterruptedException ex) {
// handle stop processing
}
}
}
});
solverTh.start();
Also you have to remove the try catch from the following:
try {
Thread.sleep(2000);
} catch (InterruptedException ex) {
Logger.getLogger(BruteForce.class.getName()).log(Level.SEVERE, null, ex);
}

The only way to interrupt thread is to make it exit itself. Strightforward interruption is not even implemented in Java because of deadlocks possibility. So your thread code must look like:
solverTh = new Thread(new Runnable() {
#Override
public void run() {
while(somethingistrue)
// Do a little stuff here
}
}
});
solverTh.start();
And somethingistrue is a kind of a signal for thread to interrupt.

When a thread is running ( consuming CPU cycles ) , then it will not by default ( automatically ) respond to Thread.interrupt(). You will have to write the code to do this explicitly.
Break up //do lot of stuff here for long time into 2 or more steps , and insert between these steps checks for the Thread.currentThread().isInterrupted() - if true - break out , else continue. This is only safe way to achieve what you want.
It depends on what the long running stuff is, you will have to design the steps and decide when its best to check for interruption and breakout.

The only thing that can reliably stop the execution of one thread from another is the OS. So, there are not many choices:
1) Signal the thread to stop itself. This scheme kinda depends on what the thread is doing. If it's running on another processor or stuck on a blocking call you cannot unblock, (note-many blocking calls can be persuaded to return early), there can be problems.
What is 'lot of stuff' doing?
2) Use an OS call to terminate the thread. This can be a viable option, depending on what the thread does. If there is any possibility of terminating the thread while it holds a public lock on a vital resource, (eg. it's in the middle of a malloc() and has the memory-manager locked), then you can get into trouble. You have to be sure of what thread is doing to safely abort it in this way.
3) Use a separate process to run the 'stuff'. This will obviously work OK, but usually involves slow and painful inter-process comms to pass data and return results.
4) Design the app so that it does not need to terminate the thread. Some apps never need to terminate any threads except at app shutdown, so there's no problem - the OS can stop anything. In those cases where a thread must be 'stopped' during an app run and is running a lengthy CPU-intensive operation or is blocked for a long and possibly indeterminate period, 'orphaning' a thread by setting its priority to minimum/idle and just leaving it to eventually die off is another common approach.
The worst possible scenario is a thread running a lot of stuff for long time that uses the memory-manager or other public locks, possibly in a library where you don't know exactly what it's doing, can't be hooked and reads/writes data in such a way that 'orphaning' it off means that another thread cannot be started to use the data. You're really stuft then and you may have to terminate the app and restart. It's just best to avoid designs where a thread can get into such a state :)
5) Forgot one - if the thread is using data you can get at, setting something to NULL, 0, MaxInt or some other like bodge can cause an exception to be raised in the thread running the long stuff. When execution bubbles out of long stuff, the thread can check the Interrupted state in the exception handler and exit if set.

Allowing only two thread to operate on a function

I have an unusual problem.
I have a function, operation in this function can be done by two threads at a time.
static int iCount = 1;
public synchronized void myFunct(){
while(iCount >= 3)
{
try {
wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
iCount++;
//Do Stuffs
//After operation decrement count
iCount --;
notifyAll();
}
What i am trying to do is, i want to allow only two threads to do some operation, and other threads must wait.
But here first two threads increment the count and does the operation and other threads go for an wait state but do not get the notification.
I guess i am overlooking something.

Sounds like you want to use a Semaphore, you always call acquire() before doing your operation, and then release() in a finally block.
private static final Semphore semaphore = new Semaphore(2);
public static void myFunct() throws InterruptedException {
semaphore.aquire();
try {
// do stuff
} finally {
semaphore.release();
}
}

Your function is synchronized, so only one thread at a time can be in it.
I'm not sure I understand your question... But if you want to allow two threads to go somewhere at once, have a look at Semaphore.

Is this a singleton class?
If not then it's a problem because many concurrent instances may change the value of icounter and in addition they will block on it forever because no thread will be able to call notify on their instance object.
Anyway you should move the sync inside the function and lock iCount and not the instance, also make it volatile.
public void myFunct(){
synchronized(iCount) {
while(iCount >= 3)
{
try {
wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
iCount++;
}
//Do Stuffs
//After operation decrement count
synchronized(iCount) {
iCount--;
}
notifyAll();

Why aren't you just using a Semaphore?

An alternative might be to use a ThreadPoolExecutor with a maximum of two threads.

You need java.util.concurrent.Semaphore, initialized with 2 permits.
As for your current code - threads may cache values of variables. Try adding the volatile keyword.

There are many problems with this code. Among them:
You have no real control on the number of threads running myFunct, since the method is synchronized on the instance level, while the counter is static. So N different threads operating on N different instances may run the same method concurrently.
Manipulating the counter by multiple threads is not thread safe. Consider synchronizing it or using AtomicInteger.
Regarding the limit on the number of threads, consider using the Semaphore class.

How to make the main thread wait for the other threads to complete in ThreadPoolExecutor

I am using the ThreadPoolExecutor to implement threading in my Java Application.
I have a XML which I need to parse and add each node of it to a thread to execute the completion. My implementation is like this:
parse_tp is a threadpool object created & ParseQuotesXML is the class with the run method.
try {
List children = root.getChildren();
Iterator iter = children.iterator();
//Parsing the XML
while(iter.hasNext()) {
Element child = (Element) iter.next();
ParseQuotesXML quote = new ParseQuotesXML(child, this);
parse_tp.execute(quote);
}
System.out.println("Print it after all the threads have completed");
catch(Exception ex) {
ex.printStackTrace();
}
finally {
System.out.println("Print it in the end.");
if(!parse_tp.isShutdown()) {
if(parse_tp.getActiveCount() == 0 && parse_tp.getQueue().size() == 0 ) {
parse_tp.shutdown();
} else {
try {
parse_tp.awaitTermination(30, TimeUnit.SECONDS);
} catch (InterruptedException ex) {
log.info("Exception while terminating the threadpool "+ex.getMessage());
ex.printStackTrace();
}
}
}
parse_tp.shutdown();
}
The problem is, the two print out statements are printed before the other threads exit. I want to make the main thread wait for all other threads to complete.
In normal Thread implementation I can do it using join() function but not getting a way to achieve the same in ThreadPool Executor. Also would like to ask if the code written in finally block to close the threadpool proper ?
Thanks,
Amit

A CountDownLatch is designed for this very purpose. Examples may be found here and here. When the number of threads is not known in advance, consider a Phaser, new in Java 1.7, or an UpDownLatch.

To answer your second question, I think you are doing a reasonable job trying to clean up your thread pool.
With respect to your first question, I think the method that you want to use is submit rather than execute. Rather than try to explain it all in text, here's an edited fragment from a unit test that I wrote that makes many tasks, has each of them do a fragment of the total work and then meets back at the starting point to add the results:
final AtomicInteger messagesReceived = new AtomicInteger(0);
// ThreadedListenerAdapter is the class that I'm testing
// It's not germane to the question other than as a target for a thread pool.
final ThreadedListenerAdapter<Integer> adapter =
new ThreadedListenerAdapter<Integer>(listener);
int taskCount = 10;
List<FutureTask<Integer>> taskList = new ArrayList<FutureTask<Integer>>();
for (int whichTask = 0; whichTask < taskCount; whichTask++) {
FutureTask<Integer> futureTask =
new FutureTask<Integer>(new Callable<Integer>() {
#Override
public Integer call() throws Exception {
// Does useful work that affects messagesSent
return messagesSent;
}
});
taskList.add(futureTask);
}
for (FutureTask<Integer> task : taskList) {
LocalExecutorService.getExecutorService().submit(task);
}
for (FutureTask<Integer> task : taskList) {
int result = 0;
try {
result = task.get();
} catch (InterruptedException ex) {
Thread.currentThread().interrupt();
} catch (ExecutionException ex) {
throw new RuntimeException("ExecutionException in task " + task, ex);
}
assertEquals(maxMessages, result);
}
int messagesSent = taskCount * maxMessages;
assertEquals(messagesSent, messagesReceived.intValue());
I think this fragment is similar to what you're trying to do. The key components were the submit and get methods.

First of all you can use ThreadPoolExecutor.submit() method, which returns Future instance, then after you submitted all your work items you can iterate trough those futures and call Future.get() on each of them.
Alternatively, you can prepare your runnable work items and submit them all at once using ThreadPoolExecutor.invokeAll(), which will wait until all work items completed and then you can get the execution results or exception calling the same Future.get() method.