I am working on a small microfinance application that processes financial transactions, the frequency of these transaction are quite high, which is why I am planning to make it a multi-threaded application that can process multiple transactions in parallel.
I have already designed all the workers that are thread safe,
what I need help for is how to manage these threads. here are some of my options
1.make a specified number of thread pool threads at startup and keep them running like in a infinite loop where they could keep looking for new transactions and if any are found start processing
example code:
void Start_Job(){
for (int l_ThreadId = 0; l_ThreadId < PaymentNoOfWorkerThread; l_ThreadId++)
{
ThreadPool.QueueUserWorkItem(Execute, (object)l_TrackingId);
}
}
void Execute(object l_TrackingId)
{
while(true)
{
var new_txns = Get_New_Txns(); //get new txns if any returns a queue
while(new_txns.count > 0 ){
process_txn(new_txns.Dequeue())
}
Thread.Sleep(some_time);
}
}
2.look for new transactions and assign a thread pool thread for each transaction (my understanding that these threads would be reused after their execution is complete for new txns)
example code:
void Start_Job(){
while(true){
var new_txns = Get_New_Txns(); //get new txns if any returns a queue
for (int l_ThreadId = 0; l_ThreadId < new_txns.count; l_ThreadId++)
{
ThreadPool.QueueUserWorkItem(Execute, (object)new_txn.Dequeue());
}
}
Thread.Sleep(some_time);
}
void Execute(object Txn)
{
process_txn(txn);
}
3.do the above but with tasks.
which option would be most efficient and well suited for my application,
thanks in advance :)
ThreadPool.QueueUserWorkItem is an older API and you shouldn't be using it directly
anymore. Tasks is the way to go and Thread pool is managed automatically for you.
What may suite your application would depend on what happens in process_txn and is subjective, so this is very generic guideline:
If process_txn is a compute bound operation: for example it performs only CPU bound calculations, then you may look at the Task Parallel Library. It will help you use the CPU cores more efficiently.
If process_txn is less of CPU and more IO bound operations: meaning if it may read/write from files/database or connects to some other remote service, then what you should look at is asynchronous programming and make sure your IO operations are all asynchronous which means your threads are never blocked on IO. This will help your service to be more scalable. Also depending on what your queue is, see if you can await on the queue asynchronously, so that none of your application threads are blocked just waiting on the queue.
In my code I have a loop, inside this loop I send several requests to a remote webservice. WS providers said: "The webservice can host at most n threads", so i need to cap my code since I can't send n+1 threads.
If I've to send m threads I would that first n threads will be executed immediately and as soon one of these is completed a new thread (one of the remaining m-n threads) will be executed and so on, until all m threads are executed.
I have thinked of a Thread Pool and explicit setting of the max thread number to n. Is this enough?
For this I would avoid the use of multiple threads. Instead, wrapping the entire loop up which can be run on a single thread. However, if you do want to launch multiple threads using the/a thread pool then I would use the Semaphore class to facilitate the required thread limit; here's how...
A semaphore is like a mean night club bouncer, it has been provide a club capacity and is not allowed to exceed this limit. Once the club is full, no one else can enter... A queue builds up outside. Then as one person leaves another can enter (analogy thanks to J. Albahari).
A Semaphore with a value of one is equivalent to a Mutex or Lock except that the Semaphore has no owner so that it is thread ignorant. Any thread can call Release on a Semaphore whereas with a Mutex/Lock only the thread that obtained the Mutex/Lock can release it.
Now, for your case we are able to use Semaphores to limit concurrency and prevent too many threads from executing a particular piece of code at once. In the following example five threads try to enter a night club that only allows entry to three...
class BadAssClub
{
static SemaphoreSlim sem = new SemaphoreSlim(3);
static void Main()
{
for (int i = 1; i <= 5; i++)
new Thread(Enter).Start(i);
}
// Enfore only three threads running this method at once.
static void Enter(int i)
{
try
{
Console.WriteLine(i + " wants to enter.");
sem.Wait();
Console.WriteLine(i + " is in!");
Thread.Sleep(1000 * (int)i);
Console.WriteLine(i + " is leaving...");
}
finally
{
sem.Release();
}
}
}
I hope this helps.
Edit. You can also use the ThreadPool.SetMaxThreads Method. This method restricts the number of threads allowed to run in the thread pool. But it does this 'globally' for the thread pool itself. This means that if you are running SQL queries or other methods in libraries that you application uses then new threads will not be spun-up due to this blocking. This may not be relevant to you, in which case use the SetMaxThreads method. If you want to block for a particular method however, it is safer to use Semphores.
I am trying to run a sample code (a very basic one) involving threading and critical regions.
Here's my code:
public static void DoCriticalWork(object o)
{
SomeClass instance = o as SomeClass;
Thread.BeginCriticalRegion();
instance.IsValid = true;
Thread.Sleep(2);
instance.IsComplete = true;
Thread.EndCriticalRegion();
instance.Print();
}
And I am calling it as follows:
private static void CriticalHandled()
{
SomeClass instance = new SomeClass();
ParameterizedThreadStart operation = new ParameterizedThreadStart(CriticalRegion.DoCriticalWork);
Thread t = new Thread(operation);
Console.WriteLine("Start thread");
t.Start(instance);
Thread.Sleep(1);
Console.WriteLine("Abort thread");
t.Abort();
Console.WriteLine("In main");
instance.Print();
}
However, the output I get is:
**
Start thread
Abort thread
In main
IsValid: True
IsComplete: False
**
Since the critical region is defined, IsComplete should be true and not false.
Can someone please explain why it is not working?
Here is SomeClass for reference:
public class SomeClass
{
private bool _isValid;
public bool IsValid
{
get { return _isValid; }
set { _isValid = value; }
}
private bool _isComplete;
public bool IsComplete
{
get { return _isComplete; }
set { _isComplete = value; }
}
public void Print()
{
Console.WriteLine("IsValid: {0}", IsValid);
Console.WriteLine("IsComplete: {0}", IsComplete);
Console.WriteLine();
}
}
Edit
Expln from MCTS notes:
The idea behind a critical region is to provide a region of code that must be executed as if it were a single line. Any attempt to abort a thread while it is within a critical region will have to wait until after the critical region is complete. At that point, the thread will be aborted, throwing the ThreadAbortException. The difference between a thread with and without a critical region is illustrated in the following figure:
Critical Region http://www.freeimagehosting.net/uploads/9dd3bb5445.gif
Thread.BeginCriticalRegion does not prevent a Thread from being aborted. I believe it is used to notify the runtime that if the Thread is aborted in the critical section, it is not necessarily safe to continue running the application/AppDomain.
The MSDN docs have a more complete explanation: http://msdn.microsoft.com/en-us/library/system.threading.thread.begincriticalregion.aspx
Its a sleep timing problem. Just widen the gap between the two sleeps and you will get the answer.
There are two threads: The main thread and the thread that does the critical work. Now when abort is called the thread 't' will abort instantly even if it has not completed the critical region.
Now as you have send the main thread to sleep for 2ms and thread t for 1ms, sometimes t will complete the critical section and sometimes it will not. So that's why the value for IsComplete is sometimes false and sometimes true.
Now just send the main thread to sleep for 100ms and you will find the IsComplete is always true.
Vice versa send the thread "t" to sleep for 100ms and you will find the IsComplete is always false.
EDIT
FROM MSDN
Notifies a host that execution is about to enter a region of code in which the effects of a thread abort or unhandled exception might jeopardize other tasks in the application domain.
For example, consider a task that attempts to allocate memory while holding a lock. If the memory allocation fails, aborting the current task is not sufficient to ensure stability of the AppDomain, because there can be other tasks in the domain waiting for the same lock. If the current task is terminated, other tasks could be deadlocked.
When a failure occurs in a critical region, the host might decide to unload the entire AppDomain rather than take the risk of continuing execution in a potentially unstable state. To inform the host that your code is entering a critical region, call BeginCriticalRegion. Call EndCriticalRegion when execution returns to a non-critical region of code.
From CLR Inside Out: Writing Reliable Code
State Corruption
There are three buckets that state corruption may fall into. The first is local state, which includes local variables and heap objects that are only used by a particular thread. The second is shared state, which includes anything shared across threads within the AppDomain, such as objects stored in static variables. Caches often fall into this category. The third is process-wide, machine-wide, and cross-machine shared state—files, sockets, shared memory, and distributed lock managers fall into this camp.
The amount of state that can be corrupted by an async exception is the maximum amount of state a thread is currently modifying. If a thread allocates a few temporary objects and doesn't expose them to other threads, only those temporary objects can be corrupted. But if a thread is writing to shared state, that shared resource may be corrupted, and other threads may potentially encounter this corrupted state. You must not let that happen. In this case, you abort all the other threads in the AppDomain and then unload the AppDomain. In this way, an asynchronous exception escalates to an AppDomain, causing it to unload and ensuring that any potentially corrupted state is thrown away. Given a transacted store like a database, this AppDomain recycling provides resiliency to corruption of local and shared state.
Critical Regions allow you to handle situations where some piece of code might corrupt other Application Domains and causing irreparable damage to the system.
A good solution is to encapsulate the code from DoCriticalWork() with
try { ... } catch(ThreadAbortedException) {...}
where you should act as you wish (maybe set IsComplete = true?)
You can read more about ThreadAbortException, and be sure to check Thread.ResetAbort and the finally block usage for this case.
As Andy mentioned and quoting from Thread.BeginCriticalRegion:
Notifies a host that execution is about to enter a region of code in which the effects of a thread abort or unhandled exception might jeopardize other tasks in the application domain.
A mutex is a programming concept that is frequently used to solve multi-threading problems. My question to the community:
What is a mutex and how do you use it?
When I am having a big heated discussion at work, I use a rubber chicken which I keep in my desk for just such occasions. The person holding the chicken is the only person who is allowed to talk. If you don't hold the chicken you cannot speak. You can only indicate that you want the chicken and wait until you get it before you speak. Once you have finished speaking, you can hand the chicken back to the moderator who will hand it to the next person to speak. This ensures that people do not speak over each other, and also have their own space to talk.
Replace Chicken with Mutex and person with thread and you basically have the concept of a mutex.
Of course, there is no such thing as a rubber mutex. Only rubber chicken. My cats once had a rubber mouse, but they ate it.
Of course, before you use the rubber chicken, you need to ask yourself whether you actually need 5 people in one room and would it not just be easier with one person in the room on their own doing all the work. Actually, this is just extending the analogy, but you get the idea.
A Mutex is a Mutually exclusive flag. It acts as a gate keeper to a section of code allowing one thread in and blocking access to all others. This ensures that the code being controlled will only be hit by a single thread at a time. Just be sure to release the mutex when you are done. :)
Mutual Exclusion. Here's the Wikipedia entry on it.
The point of a mutex is to synchronize two threads. When you have two threads attempting to access a single resource, the general pattern is to have the first block of code attempting access to set the mutex before entering the code. When the second code block attempts access, it sees that the mutex is set and waits until the first block of code is complete (and unsets the mutex), then continues.
Specific details of how this is accomplished obviously varies greatly by programming language.
When you have a multi-threaded application, the different threads sometimes share a common resource, such as a variable or similar. This shared source often cannot be accessed at the same time, so a construct is needed to ensure that only one thread is using that resource at a time.
The concept is called "mutual exclusion" (short Mutex), and is a way to ensure that only one thread is allowed inside that area, using that resource etc.
How to use them is language specific, but is often (if not always) based on a operating system mutex.
Some languages doesn't need this construct, due to the paradigm, for example functional programming (Haskell, ML are good examples).
What is a Mutex?
The mutex (In fact, the term mutex is short for mutual exclusion) also known as spinlock is the simplest synchronization tool that is used to protect critical regions and thus prevent race conditions. That is a thread must acquire a lock before entering into a critical section (In critical section multi threads share a common variable, updating a table, writing a file and so on), it releases the lock when it leaves critical section.
What is a Race Condition?
A race condition occurs when two or more threads can access shared data and they try to change it at the same time. Because the thread scheduling algorithm can swap between threads at any time, you don't know the order in which the threads will attempt to access the shared data. Therefore, the result of the change in data is dependent on the thread scheduling algorithm, i.e. both threads are "racing" to access/change the data.
Real life example:
When I am having a big heated discussion at work, I use a rubber
chicken which I keep in my desk for just such occasions. The person
holding the chicken is the only person who is allowed to talk. If you
don't hold the chicken you cannot speak. You can only indicate that
you want the chicken and wait until you get it before you speak. Once
you have finished speaking, you can hand the chicken back to the
moderator who will hand it to the next person to speak. This ensures
that people do not speak over each other, and also have their own
space to talk.
Replace Chicken with Mutex and person with thread and you basically have the concept of a mutex.
#Xetius
Usage in C#:
This example shows how a local Mutex object is used to synchronize access to a protected resource. Because each calling thread is blocked until it acquires ownership of the mutex, it must call the ReleaseMutex method to release ownership of the thread.
using System;
using System.Threading;
class Example
{
// Create a new Mutex. The creating thread does not own the mutex.
private static Mutex mut = new Mutex();
private const int numIterations = 1;
private const int numThreads = 3;
static void Main()
{
// Create the threads that will use the protected resource.
for(int i = 0; i < numThreads; i++)
{
Thread newThread = new Thread(new ThreadStart(ThreadProc));
newThread.Name = String.Format("Thread{0}", i + 1);
newThread.Start();
}
// The main thread exits, but the application continues to
// run until all foreground threads have exited.
}
private static void ThreadProc()
{
for(int i = 0; i < numIterations; i++)
{
UseResource();
}
}
// This method represents a resource that must be synchronized
// so that only one thread at a time can enter.
private static void UseResource()
{
// Wait until it is safe to enter.
Console.WriteLine("{0} is requesting the mutex",
Thread.CurrentThread.Name);
mut.WaitOne();
Console.WriteLine("{0} has entered the protected area",
Thread.CurrentThread.Name);
// Place code to access non-reentrant resources here.
// Simulate some work.
Thread.Sleep(500);
Console.WriteLine("{0} is leaving the protected area",
Thread.CurrentThread.Name);
// Release the Mutex.
mut.ReleaseMutex();
Console.WriteLine("{0} has released the mutex",
Thread.CurrentThread.Name);
}
}
// The example displays output like the following:
// Thread1 is requesting the mutex
// Thread2 is requesting the mutex
// Thread1 has entered the protected area
// Thread3 is requesting the mutex
// Thread1 is leaving the protected area
// Thread1 has released the mutex
// Thread3 has entered the protected area
// Thread3 is leaving the protected area
// Thread3 has released the mutex
// Thread2 has entered the protected area
// Thread2 is leaving the protected area
// Thread2 has released the mutex
MSDN Reference Mutex
There are some great answers here, here is another great analogy for explaining what mutex is:
Consider single toilet with a key. When someone enters, they take the key and the toilet is occupied. If someone else needs to use the toilet, they need to wait in a queue. When the person in the toilet is done, they pass the key to the next person in queue. Make sense, right?
Convert the toilet in the story to a shared resource, and the key to a mutex. Taking the key to the toilet (acquire a lock) permits you to use it. If there is no key (the lock is locked) you have to wait. When the key is returned by the person (release the lock) you're free to acquire it now.
In C#, the common mutex used is the Monitor. The type is 'System.Threading.Monitor'. It may also be used implicitly via the 'lock(Object)' statement. One example of its use is when constructing a Singleton class.
private static readonly Object instanceLock = new Object();
private static MySingleton instance;
public static MySingleton Instance
{
lock(instanceLock)
{
if(instance == null)
{
instance = new MySingleton();
}
return instance;
}
}
The lock statement using the private lock object creates a critical section. Requiring each thread to wait until the previous is finished. The first thread will enter the section and initialize the instance. The second thread will wait, get into the section, and get the initialized instance.
Any sort of synchronization of a static member may use the lock statement similarly.
To understand MUTEX at first you need to know what is "race condition" and then only you will understand why MUTEX is needed. Suppose you have a multi-threading program and you have two threads. Now, you have one job in the job queue. The first thread will check the job queue and after finding the job it will start executing it. The second thread will also check the job queue and find that there is one job in the queue. So, it will also assign the same job pointer. So, now what happens, both the threads are executing the same job. This will cause a segmentation fault. This is the example of a race condition.
The solution to this problem is MUTEX. MUTEX is a kind of lock which locks one thread at a time. If another thread wants to lock it, the thread simply gets blocked.
The MUTEX topic in this pdf file link is really worth reading.
Mutexes are useful in situations where you need to enforce exclusive access to a resource accross multiple processes, where a regular lock won't help since it only works accross threads.
Mutex: Mutex stands for Mutual Exclusion. It means only one process/thread can enter into critical section at a given time. In concurrent programming multiple threads/process updating the shared resource (any variable, shared memory etc.) may lead to some unexpected result. ( As the result depends upon the which thread/process gets the first access).
In order to avoid such an unexpected result we need some synchronization mechanism, which ensures that only one thread/process gets access to such a resource at a time.
pthread library provides support for Mutex.
typedef union
{
struct __pthread_mutex_s
{
***int __lock;***
unsigned int __count;
int __owner;
#ifdef __x86_64__
unsigned int __nusers;
#endif
int __kind;
#ifdef __x86_64__
short __spins;
short __elision;
__pthread_list_t __list;
# define __PTHREAD_MUTEX_HAVE_PREV 1
# define __PTHREAD_SPINS 0, 0
#else
unsigned int __nusers;
__extension__ union
{
struct
{
short __espins;
short __elision;
# define __spins __elision_data.__espins
# define __elision __elision_data.__elision
# define __PTHREAD_SPINS { 0, 0 }
} __elision_data;
__pthread_slist_t __list;
};
#endif
This is the structure for mutex data type i.e pthread_mutex_t.
When mutex is locked, __lock set to 1. When it is unlocked __lock set to 0.
This ensure that no two processes/threads can access the critical section at same time.
A semaphore is a programming concept that is frequently used to solve multi-threading problems. My question to the community:
What is a semaphore and how do you use it?
Think of semaphores as bouncers at a nightclub. There are a dedicated number of people that are allowed in the club at once. If the club is full no one is allowed to enter, but as soon as one person leaves another person might enter.
It's simply a way to limit the number of consumers for a specific resource. For example, to limit the number of simultaneous calls to a database in an application.
Here is a very pedagogic example in C# :-)
using System;
using System.Collections.Generic;
using System.Text;
using System.Threading;
namespace TheNightclub
{
public class Program
{
public static Semaphore Bouncer { get; set; }
public static void Main(string[] args)
{
// Create the semaphore with 3 slots, where 3 are available.
Bouncer = new Semaphore(3, 3);
// Open the nightclub.
OpenNightclub();
}
public static void OpenNightclub()
{
for (int i = 1; i <= 50; i++)
{
// Let each guest enter on an own thread.
Thread thread = new Thread(new ParameterizedThreadStart(Guest));
thread.Start(i);
}
}
public static void Guest(object args)
{
// Wait to enter the nightclub (a semaphore to be released).
Console.WriteLine("Guest {0} is waiting to entering nightclub.", args);
Bouncer.WaitOne();
// Do some dancing.
Console.WriteLine("Guest {0} is doing some dancing.", args);
Thread.Sleep(500);
// Let one guest out (release one semaphore).
Console.WriteLine("Guest {0} is leaving the nightclub.", args);
Bouncer.Release(1);
}
}
}
The article Mutexes and Semaphores Demystified by Michael Barr is a great short introduction into what makes mutexes and semaphores different, and when they should and should not be used. I've excerpted several key paragraphs here.
The key point is that mutexes should be used to protect shared resources, while semaphores should be used for signaling. You should generally not use semaphores to protect shared resources, nor mutexes for signaling. There are issues, for instance, with the bouncer analogy in terms of using semaphores to protect shared resources - you can use them that way, but it may cause hard to diagnose bugs.
While mutexes and semaphores have some similarities in their implementation, they should always be used differently.
The most common (but nonetheless incorrect) answer to the question posed at the top is that mutexes and semaphores are very similar, with the only significant difference being that semaphores can count higher than one. Nearly all engineers seem to properly understand that a mutex is a binary flag used to protect a shared resource by ensuring mutual exclusion inside critical sections of code. But when asked to expand on how to use a "counting semaphore," most engineers—varying only in their degree of confidence—express some flavor of the textbook opinion that these are used to protect several equivalent resources.
...
At this point an interesting analogy is made using the idea of bathroom keys as protecting shared resources - the bathroom. If a shop has a single bathroom, then a single key will be sufficient to protect that resource and prevent multiple people from using it simultaneously.
If there are multiple bathrooms, one might be tempted to key them alike and make multiple keys - this is similar to a semaphore being mis-used. Once you have a key you don't actually know which bathroom is available, and if you go down this path you're probably going to end up using mutexes to provide that information and make sure you don't take a bathroom that's already occupied.
A semaphore is the wrong tool to protect several of the essentially same resource, but this is how many people think of it and use it. The bouncer analogy is distinctly different - there aren't several of the same type of resource, instead there is one resource which can accept multiple simultaneous users. I suppose a semaphore can be used in such situations, but rarely are there real-world situations where the analogy actually holds - it's more often that there are several of the same type, but still individual resources, like the bathrooms, which cannot be used this way.
...
The correct use of a semaphore is for signaling from one task to another. A mutex is meant to be taken and released, always in that order, by each task that uses the shared resource it protects. By contrast, tasks that use semaphores either signal or wait—not both. For example, Task 1 may contain code to post (i.e., signal or increment) a particular semaphore when the "power" button is pressed and Task 2, which wakes the display, pends on that same semaphore. In this scenario, one task is the producer of the event signal; the other the consumer.
...
Here an important point is made that mutexes interfere with real time operating systems in a bad way, causing priority inversion where a less important task may be executed before a more important task because of resource sharing. In short, this happens when a lower priority task uses a mutex to grab a resource, A, then tries to grab B, but is paused because B is unavailable. While it's waiting, a higher priority task comes along and needs A, but it's already tied up, and by a process that isn't even running because it's waiting for B. There are many ways to resolve this, but it most often is fixed by altering the mutex and task manager. The mutex is much more complex in these cases than a binary semaphore, and using a semaphore in such an instance will cause priority inversions because the task manager is unaware of the priority inversion and cannot act to correct it.
...
The cause of the widespread modern confusion between mutexes and semaphores is historical, as it dates all the way back to the 1974 invention of the Semaphore (capital "S", in this article) by Djikstra. Prior to that date, none of the interrupt-safe task synchronization and signaling mechanisms known to computer scientists was efficiently scalable for use by more than two tasks. Dijkstra's revolutionary, safe-and-scalable Semaphore was applied in both critical section protection and signaling. And thus the confusion began.
However, it later became obvious to operating system developers, after the appearance of the priority-based preemptive RTOS (e.g., VRTX, ca. 1980), publication of academic papers establishing RMA and the problems caused by priority inversion, and a paper on priority inheritance protocols in 1990, 3 it became apparent that mutexes must be more than just semaphores with a binary counter.
Mutex: resource sharing
Semaphore: signaling
Don't use one for the other without careful consideration of the side effects.
Mutex: exclusive-member access to a resource
Semaphore: n-member access to a resource
That is, a mutex can be used to syncronize access to a counter, file, database, etc.
A sempahore can do the same thing but supports a fixed number of simultaneous callers. For example, I can wrap my database calls in a semaphore(3) so that my multithreaded app will hit the database with at most 3 simultaneous connections. All attempts will block until one of the three slots opens up. They make things like doing naive throttling really, really easy.
Consider, a taxi that can accommodate a total of 3(rear)+2(front) persons including the driver. So, a semaphore allows only 5 persons inside a car at a time.
And a mutex allows only 1 person on a single seat of the car.
Therefore, Mutex is to allow exclusive access for a resource (like an OS thread) while a Semaphore is to allow access for n number of resources at a time.
#Craig:
A semaphore is a way to lock a
resource so that it is guaranteed that
while a piece of code is executed,
only this piece of code has access to
that resource. This keeps two threads
from concurrently accesing a resource,
which can cause problems.
This is not restricted to only one thread. A semaphore can be configured to allow a fixed number of threads to access a resource.
Semaphore can also be used as a ... semaphore.
For example if you have multiple process enqueuing data to a queue, and only one task consuming data from the queue. If you don't want your consuming task to constantly poll the queue for available data, you can use semaphore.
Here the semaphore is not used as an exclusion mechanism, but as a signaling mechanism.
The consuming task is waiting on the semaphore
The producing task are posting on the semaphore.
This way the consuming task is running when and only when there is data to be dequeued
There are two essential concepts to building concurrent programs - synchronization and mutual exclusion. We will see how these two types of locks (semaphores are more generally a kind of locking mechanism) help us achieve synchronization and mutual exclusion.
A semaphore is a programming construct that helps us achieve concurrency, by implementing both synchronization and mutual exclusion. Semaphores are of two types, Binary and Counting.
A semaphore has two parts : a counter, and a list of tasks waiting to access a particular resource. A semaphore performs two operations : wait (P) [this is like acquiring a lock], and release (V)[ similar to releasing a lock] - these are the only two operations that one can perform on a semaphore. In a binary semaphore, the counter logically goes between 0 and 1. You can think of it as being similar to a lock with two values : open/closed. A counting semaphore has multiple values for count.
What is important to understand is that the semaphore counter keeps track of the number of tasks that do not have to block, i.e., they can make progress. Tasks block, and add themselves to the semaphore's list only when the counter is zero. Therefore, a task gets added to the list in the P() routine if it cannot progress, and "freed" using the V() routine.
Now, it is fairly obvious to see how binary semaphores can be used to solve synchronization and mutual exclusion - they are essentially locks.
ex. Synchronization:
thread A{
semaphore &s; //locks/semaphores are passed by reference! think about why this is so.
A(semaphore &s): s(s){} //constructor
foo(){
...
s.P();
;// some block of code B2
...
}
//thread B{
semaphore &s;
B(semaphore &s): s(s){} //constructor
foo(){
...
...
// some block of code B1
s.V();
..
}
main(){
semaphore s(0); // we start the semaphore at 0 (closed)
A a(s);
B b(s);
}
In the above example, B2 can only execute after B1 has finished execution. Let's say thread A comes executes first - gets to sem.P(), and waits, since the counter is 0 (closed). Thread B comes along, finishes B1, and then frees thread A - which then completes B2. So we achieve synchronization.
Now let's look at mutual exclusion with a binary semaphore:
thread mutual_ex{
semaphore &s;
mutual_ex(semaphore &s): s(s){} //constructor
foo(){
...
s.P();
//critical section
s.V();
...
...
s.P();
//critical section
s.V();
...
}
main(){
semaphore s(1);
mutual_ex m1(s);
mutual_ex m2(s);
}
The mutual exclusion is quite simple as well - m1 and m2 cannot enter the critical section at the same time. So each thread is using the same semaphore to provide mutual exclusion for its two critical sections. Now, is it possible to have greater concurrency? Depends on the critical sections. (Think about how else one could use semaphores to achieve mutual exclusion.. hint hint : do i necessarily only need to use one semaphore?)
Counting semaphore: A semaphore with more than one value. Let's look at what this is implying - a lock with more than one value?? So open, closed, and ...hmm. Of what use is a multi-stage-lock in mutual exclusion or synchronization?
Let's take the easier of the two:
Synchronization using a counting semaphore: Let's say you have 3 tasks - #1 and 2 you want executed after 3. How would you design your synchronization?
thread t1{
...
s.P();
//block of code B1
thread t2{
...
s.P();
//block of code B2
thread t3{
...
//block of code B3
s.V();
s.V();
}
So if your semaphore starts off closed, you ensure that t1 and t2 block, get added to the semaphore's list. Then along comes all important t3, finishes its business and frees t1 and t2. What order are they freed in? Depends on the implementation of the semaphore's list. Could be FIFO, could be based some particular priority,etc. (Note : think about how you would arrange your P's and V;s if you wanted t1 and t2 to be executed in some particular order, and if you weren't aware of the implementation of the semaphore)
(Find out : What happens if the number of V's is greater than the number of P's?)
Mutual Exclusion Using counting semaphores: I'd like you to construct your own pseudocode for this (makes you understand things better!) - but the fundamental concept is this : a counting semaphore of counter = N allows N tasks to enter the critical section freely. What this means is you have N tasks (or threads, if you like) enter the critical section, but the N+1th task gets blocked (goes on our favorite blocked-task list), and only is let through when somebody V's the semaphore at least once. So the semaphore counter, instead of swinging between 0 and 1, now goes between 0 and N, allowing N tasks to freely enter and exit, blocking nobody!
Now gosh, why would you need such a stupid thing? Isn't the whole point of mutual exclusion to not let more than one guy access a resource?? (Hint Hint...You don't always only have one drive in your computer, do you...?)
To think about : Is mutual exclusion achieved by having a counting semaphore alone? What if you have 10 instances of a resource, and 10 threads come in (through the counting semaphore) and try to use the first instance?
I've created the visualization which should help to understand the idea. Semaphore controls access to a common resource in a multithreading environment.
ExecutorService executor = Executors.newFixedThreadPool(7);
Semaphore semaphore = new Semaphore(4);
Runnable longRunningTask = () -> {
boolean permit = false;
try {
permit = semaphore.tryAcquire(1, TimeUnit.SECONDS);
if (permit) {
System.out.println("Semaphore acquired");
Thread.sleep(5);
} else {
System.out.println("Could not acquire semaphore");
}
} catch (InterruptedException e) {
throw new IllegalStateException(e);
} finally {
if (permit) {
semaphore.release();
}
}
};
// execute tasks
for (int j = 0; j < 10; j++) {
executor.submit(longRunningTask);
}
executor.shutdown();
Output
Semaphore acquired
Semaphore acquired
Semaphore acquired
Semaphore acquired
Could not acquire semaphore
Could not acquire semaphore
Could not acquire semaphore
Sample code from the article
A semaphore is an object containing a natural number (i.e. a integer greater or equal to zero) on which two modifying operations are defined. One operation, V, adds 1 to the natural. The other operation, P, decreases the natural number by 1. Both activities are atomic (i.e. no other operation can be executed at the same time as a V or a P).
Because the natural number 0 cannot be decreased, calling P on a semaphore containing a 0 will block the execution of the calling process(/thread) until some moment at which the number is no longer 0 and P can be successfully (and atomically) executed.
As mentioned in other answers, semaphores can be used to restrict access to a certain resource to a maximum (but variable) number of processes.
A hardware or software flag. In multi tasking systems , a semaphore is as variable with a value that indicates the status of a common resource.A process needing the resource checks the semaphore to determine the resources status and then decides how to proceed.
Semaphores are act like thread limiters.
Example: If you have a pool of 100 threads and you want to perform some DB operation. If 100 threads access the DB at a given time, then there may be locking issue in DB so we can use semaphore which allow only limited thread at a time.Below Example allow only one thread at a time. When a thread call the acquire() method, it will then get the access and after calling the release() method, it will release the acccess so that next thread will get the access.
package practice;
import java.util.concurrent.Semaphore;
public class SemaphoreExample {
public static void main(String[] args) {
Semaphore s = new Semaphore(1);
semaphoreTask s1 = new semaphoreTask(s);
semaphoreTask s2 = new semaphoreTask(s);
semaphoreTask s3 = new semaphoreTask(s);
semaphoreTask s4 = new semaphoreTask(s);
semaphoreTask s5 = new semaphoreTask(s);
s1.start();
s2.start();
s3.start();
s4.start();
s5.start();
}
}
class semaphoreTask extends Thread {
Semaphore s;
public semaphoreTask(Semaphore s) {
this.s = s;
}
#Override
public void run() {
try {
s.acquire();
Thread.sleep(1000);
System.out.println(Thread.currentThread().getName()+" Going to perform some operation");
s.release();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
So imagine everyone is trying to go to the bathroom and there's only a certain number of keys to the bathroom. Now if there's not enough keys left, that person needs to wait. So think of semaphore as representing those set of keys available for bathrooms (the system resources) that different processes (bathroom goers) can request access to.
Now imagine two processes trying to go to the bathroom at the same time. That's not a good situation and semaphores are used to prevent this. Unfortunately, the semaphore is a voluntary mechanism and processes (our bathroom goers) can ignore it (i.e. even if there are keys, someone can still just kick the door open).
There are also differences between binary/mutex & counting semaphores.
Check out the lecture notes at http://www.cs.columbia.edu/~jae/4118/lect/L05-ipc.html.
This is an old question but one of the most interesting uses of semaphore is a read/write lock and it has not been explicitly mentioned.
The r/w locks works in simple fashion: consume one permit for a reader and all permits for writers.
Indeed, a trivial implementation of a r/w lock but requires metadata modification on read (actually twice) that can become a bottle neck, still significantly better than a mutex or lock.
Another downside is that writers can be started rather easily as well unless the semaphore is a fair one or the writes acquire permits in multiple requests, in such case they need an explicit mutex between themselves.
Further read:
Mutex is just a boolean while semaphore is a counter.
Both are used to lock part of code so it's not accessed by too many threads.
Example
lock.set()
a += 1
lock.unset()
Now if lock was a mutex, it means that it will always be locked or unlocked (a boolean under the surface) regardless how many threads try access the protected snippet of code. While locked, any other thread would just wait until it's unlocked/unset by the previous thread.
Now imagine if instead lock was under the hood a counter with a predefined MAX value (say 2 for our example). Then if 2 threads try to access the resource, then lock would get its value increased to 2. If a 3rd thread then tried to access it, it would simply wait for the counter to go below 2 and so on.
If lock as a semaphore had a max of 1, then it would be acting exactly as a mutex.
A semaphore is a way to lock a resource so that it is guaranteed that while a piece of code is executed, only this piece of code has access to that resource. This keeps two threads from concurrently accesing a resource, which can cause problems.