Sree's Blog

Tuesday, December 10, 2013

Thread Scheduling

Execution of multiple threads on a single CPU in some order is called Thread Scheduling. The Java Runtime Environment supports a very simple, deterministic algorithm called fixed-priority scheduling. This algorithm schedules threads on the basis of their priority relative to other Runnable threads.

When a thread is created, it inherits the priority from the thread that creates it. In addition, by using the setPriority method, you can modify a thread's priority at any time after its creation.

Thread priorities ranges between MIN_PRIORITY and MAX_PRIORITY.

Following are the main features of Thread Scheduling:

The JVM schedules using a pre-emptive, priority based algorithm.
All threads have a priority and the thread with the highest priority is scheduled to run by the JVM.
In case two threads have the same priority, FIFO ordering is followed.
A different thread is invoked to run in case one of the following events occur:

The currently running thread exits the Running state. i.e., either blocks or terminates.
A thread with a higher priority than the thread currently running enters the Runnable state. The lower priority thread is preempted and the higher priority thread is scheduled to run.

Time slicing is dependent on the algorithm implementation.
A thread can voluntarily give up its right to execute at any time by calling the yield() method. This process is called Cooperative Multitasking. Threads can yield the CPU only to other threads of the same priority. Attempts to yield to a lower-priority thread are ignored.

Monday, December 9, 2013

Thread Priorities

Every thread has a priority.
The priority of a thread is used to inform the thread scheduler how important the thread to get picked.
Threads with higher priority are executed in preference to threads with lower priority.
Below are the thread priorities that Java API provides

MAX_PRIORITY - The maximum priority a thread can have.
NORM_PRIORITY - The default priority that is assigned to a thread.
MIN_PRIORITY - The minimum priority that a thread can have.

Java API offers method to get and set the priorities of the thread.

getPriority() method:

Return the thread's priority.

setPriority() method:

Changes the priority of this thread. First checkAccess() method of this thread, to determine if the currently running thread has permission to modify the thread, is called with no arguments. This may result in throwing SecurityException. If the priority is not in the range MIN_PRIORITY and MAX_PRIORITY, an IllegalArgumentException will be thrown.

toString() method:

Returns a string representation of the thread, including thread name, priority, and thread group.

The priority of a newly created thread is set equal to the priority of the thread creating it, that is, the currently running thread. The method setPriority() may be used to change the priority to a new value.

Sunday, December 8, 2013

The Lifecycle of a Thread in Java

Understanding the lifecycle of threads is very important while programming in Threads.

A Thread can be various states during its life time. The following diagram describes all the states a thread can undergo and the method calls that cause the transition from one state to another:

Java Thread States

Runnable: A Thread starts its life from this state. Thread first enters into this state after the start() method of the thread is invoked. In this state, thread will be waiting for the scheduler to pick the thread for execution. The scheduler picks the threads based on their priorities. A thread can also re-enter this state either after running, waiting, sleeping or also while coming back from blocked state.
Running: A thread is in running state means its currently running. There are several ways for thread to go into Runnable state. But there is only one way to come into Running state: the scheduler selects the thread for execution for the runnable pool. A thread runs until its swapped out, becomes blocked, or voluntarily give up its turn by invoking the static method Thread.yield().
Dead: A thread enters this state when it completes execution. It may also enters this state when it is terminated by unrecoverable error condition. If a thread goes to this state means it can not be run again.
Sleep: A Java thread may be forced to sleep (suspend) for some predefined time. In this state, thread is still alive but is not runnable, it might be return to runnable state later. It can throw InterruptedException.
Blocked: A thread can enter blocked state because of waiting for the resources that are held by another thread.

Blocked on I/O: Thread enters this state because of waiting for I/O resources. In this case, the thread will be sent back to Runnable state after the availability of resources.
Blocked on Synchronization: Thread may enters this state while waiting for object lock. Thread will be moved to Runnable state after it acquires lock.

Waiting state: A call to Object.wait() method causes the current object to wait. The thread remains in waiting state until some other thread invokes notify() or notifyAll() method of this object.

Release 5.0 introduced the Thread.getState() method, which results in one of the following Thread.State values:

NEW
RUNNABLE
BLOCKED
WAITING
TIMED_WAITING
TERMINATED

The Thread class API also introduced a method isAlive(), which returns true if the thread has already been started and not stopped. If the method isAlive() returns false, the thread is either New, or is Dead.

Tuesday, December 3, 2013

Concurrency in Java

If a system that allows to execute multiple tasks simultaneously that system is called a Concurrent System.

The Java language and the JVM have been designed to support concurrent programming from the beginning. Since Java 5.0 version, it has also included high-level concurrency APIs in java.uti.concurrent packages. Also the later versions introduced updated versions of the existing APIs, also added several new APIs.

Processes and Threads:

In concurrent programming, there are two basic units of execution, called processes and threads. In Java, concurrent programming mainly concerned about threads.
A process has a self-contained execution environment. It generally has a complete private set of basic run-time resources, in particular, each process has its own memory space.
Threads are sometimes called as lightweight processes. Both processes and threads provide and execution environment, but creating new threads requires fewer resources than required for new processes.
Threads exist within a process. Every process has at least one thread.
Threads share the process's resources, including memory and open files.
Java Program runs in its own process and by default in one thread.

Thread Basics:

A thread is a thread of execution.
The JVM allows an application to have multiple threads of execution running concurrently.
Each thread has a priority. Threads with higher priority are executed in preference to the threads with the lower priority.
Each thread may or may not have also marked a Daemon.
When a JVM starts up, there is usually a single non-daemon thread (which typically invokes the main method).
The JVM continues to execute the threads until either of the following occurs:

The exit method of Runtime class has been invoked and the security manager has allowed the exit method to execute.
All threads that are not daemon threads have died, either by returning from the run method call or by throwing an exception.

Creating and Running Threads:

Threads can be created in two ways:

Extending Thread class and overriding run() method.
Building a class that implements Runnable interface and then creating an object of Thread class passing the Runnable object as a parameter.

Below is a simple example of creating threads using the second approach. This example creates 10 threads, each of the thread calculate and prints the multiplication tables from 1 to 10.

Steps to implement the example:

Create a class Name Calculator implementing Runnable interface.

public class Calculator implements Runnable

Declare a private int attribute named number and implement the constructor to initialize its value.

private int number;

Implement the run() method. This method calculates the multiplication table of the number.

@Override
 public void run() {
  for (int i = 1; i <= 10; i++) {
   System.out.printf("%s: %d * %d = %d\n", Thread.currentThread()
     .getName(), number, i, i * number);
  }
 }

Implement the main class of the application, which contains main() method.

Inside the main() method, create a for loop with 10 iterations. Inside for loop, create an object of Calculator class, an object of the Thread class, pass the object of Calculator class as a parameter, and invoke the start() method of the Thread object.

public class Main {
 public static void main(String[] args) {
  for(int i=0; i<10; i++){
   Thread thread = new Thread(new Calculator(i));
   thread.start();
  }
 }
}

Below is the partial output of a sample run of the above code:

Every Java Program has atleast one thread in Execution. When we run a program, JVM runs the execution thread that calls main() method.

Creating an object to the Thread class doesn't create an execution of the Thread. Or calling the run() method of the implemented Runnable interface will not create a new execution of Thread. Invoking start() method on creates a new execution thread.

When we call the start() method of Thread object, it creates another execution thread. Our program will have as many threads as calls to the start() method are made.

A Java program ends when all its threads(all non-daemon) finishes.

If the initial thread(the one that executes main() method) ends, the rest of threads continue with their execution until they finish.

If a thread is exited by calling System.exit(), all threads end their execution.

References:

http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/package-summary.html
http://docs.oracle.com/javase/7/docs/technotes/guides/concurrency/
Java 7 Concurrency Book by Javier Fernández

What is the difference between Concurrency and Parallelism?

Parallelism:
Parallelism is when multiple tasks literally executing at the same time. This arises when two threads are executing simultaneously.

Ex: Running Multiple tasks on a multi core processor.

Concurrency:

Concurrency is when multiple tasks can start, run and complete execution in overlapping time periods. It doesn't necessarily mean that they will be running at the same instant.

This occurs when at least two threads are making progress.

This is a more generalized form of parallelism that includes time-slicing as a form of virtual parallelism.

Ex: Multitasking in a single core processor.

Concurrent Programming

Concurrent Programming?

It is about the elements and mechanisms a platform offers to have multiple tasks or programs to execute at once and communicate each other exchanging data or to synchronize to each other.

When you work with a computer, you can do several things at once. You can hear music while you edit a document in a word processor and read your e-mail. This can be done because your operating system allows the concurrency of tasks.

Is Java a Concurrent Platform?

Yes. Java is a concurrent platform and offers a lot of API to execute concurrent tasks in a Java Program. Java continuously increasing the functionalities offered to facilitate the development of concurrent programs with each version.

All Modern Operating systems allows concurrent tasks execution.

For ex, you can read emails while listening to music. This is called Process-level concurrency.
But inside a process we can have multiple tasks which can be executed simultaneously. So, the concurrent tasks that run inside a process are called as threads.