常用演算法帖(C#): 並發集合

文章目錄

• IProducerConsumerCollection
• ConcurrentQueue
• ConcurrentStack
• ConcurrentDictionary
• ConcurrentBag
• BlockingCollection

    微軟對C#(4.0)的架構添加了全新的並發編程架構，現在我們也能用C#開發支援並發概念的程式的。在並發編程中最讓人煩惱的應該就是如何資料同步：避免髒讀和髒寫，當然我們可以通過Lock技術來實現，也可以使用微軟提供給我們的並發集合，這些集合都提供了TryDo方法。用它們對資料的讀/寫操作能在TryDo返回True的情況下執行。我們來看看它們吧：

 

IProducerConsumerCollection

    所有的並發集合都實現了這個介面，TryAdd和TryTake分別在讀和寫的時候判斷是否能正常進行，不行則返回false。

    public interface IProducerConsumerCollection<T> : IEnumerable<T>, ICollection, IEnumerable    {        void CopyTo(T[] array, int index);        T[] ToArray();        bool TryAdd(T item);        bool TryTake(out T item);    }

 

ConcurrentQueue

    並發隊列，隊列類型的資料結構。

 public static void ConcurrentQueueTest()        {            Parallel.For<Queue<string>>(1, 10000,                () =>                {                    while (strNornalQueue.Count != 26)                    {                        if (strNornalQueue.Count == 0)                        {                            for (int i = 65; i <= 90; i++)                            {                                strNornalQueue.Enqueue(Convert.ToChar(i).ToString());                            }                        }                    }                    Console.WriteLine("-------------start------------");                    return strNornalQueue;                },                (index, state, head) =>                {                    string tmp = string.Empty;                    tmp = head.Dequeue();                    Console.WriteLine(string.Format("The element '{0}' was set by thread {1}", tmp, System.Threading.Thread.CurrentThread.ManagedThreadId));                    head.Enqueue(tmp);                                        return strNornalQueue;                },                (result) =>                {                    Console.WriteLine("-------------end------------------");                });            Console.WriteLine(string.Format("current collection has {0} elements, Has duplicate data:{1}", strNornalQueue.Count, IsDuplicate<string>(strNornalQueue.GetEnumerator())));            foreach (string item in strNornalQueue)            {                Console.WriteLine(item);            }        }

 

 

ConcurrentStack

並發棧，棧類型的資料結構。

     public static void ConcurrentStackTest()        {            Parallel.For<ConcurrentStack<string>>(1, 10000,                () =>                {                    while (strStack.Count != 26)                    {                        if (strStack.Count == 0)                        {                            for (int i = 65; i <= 90; i++)                            {                                strStack.Push(Convert.ToChar(i).ToString());                            }                        }                    }                    Console.WriteLine("-------------start------------");                    return strStack;                },                (index, state, head) =>                {                    string tmp = string.Empty;                    if (head.TryPop(out tmp))                    {                        Console.WriteLine(string.Format("The element '{0}' was set by thread {1}", tmp, System.Threading.Thread.CurrentThread.ManagedThreadId));                        head.Push(tmp);                    }                    else                    {                        Console.WriteLine("queue is buzy now");                    }                    return strStack;                },                (result) =>                {                    Console.WriteLine("-------------end------------------");                });            Console.WriteLine(string.Format("current collection has {0} elements, Has duplicate data:{1}", strStack.Count, IsDuplicate<string>(strStack.GetEnumerator())));            foreach (string item in strStack)            {                Console.WriteLine(item);            }        }

 

ConcurrentDictionary

    並發字典，字典類型的資料結構。

 public static void ConcurrentDictionary()        {            for (int i = 65; i <= 90; i++)            {                strDictionary.TryAdd(Convert.ToChar(i).ToString(), Convert.ToChar(i).ToString());            }            Parallel.For<ConcurrentDictionary<string,string>>(1, 10000,               () =>               {                                     Console.WriteLine("-------------start------------");                   return strDictionary;               },               (index, state, head) =>               {                   string tmp = string.Empty;                   if (head.TryRemove(Convert.ToChar(new Random().Next(65,90)).ToString(), out tmp))                   {                                              Console.WriteLine(string.Format("The element '{0}' was set by thread {1}", tmp, System.Threading.Thread.CurrentThread.ManagedThreadId));                       head.TryAdd(tmp, tmp);                   }                   else                   {                       Console.WriteLine("queue is buzy now");                   }                   return strDictionary;               },               (result) =>               {                   Console.WriteLine("-------------end------------------");               });        }

 

ConcurrentBag

    類似堆棧的資料結構。

public static void ConcurrentBag()        {            for (int i = 65; i <= 90; i++)            {                strBag.Add(Convert.ToChar(i).ToString());            }            Parallel.For<ConcurrentBag<string>>(1, 10000,                () =>                {                    return strBag;                },                (index, state, head) =>                {                    string tmp = string.Empty;                    if (strBag.TryTake(out tmp))                    {                        Console.WriteLine(string.Format("The element '{0}' was set by thread {1}", tmp, System.Threading.Thread.CurrentThread.ManagedThreadId));                        strBag.Add(tmp);                    }                    else                    {                        Console.WriteLine("queue is buzy now");                    }                                       return strBag;                },                (result) =>                {                    Console.WriteLine("-------------end------------------");                });            Console.WriteLine(string.Format("current collection has {0} elements, Has duplicate data:{1}", strBag.Count, IsDuplicate<string>(strBag.GetEnumerator())));            foreach (string item in strBag)            {                Console.WriteLine(item);            }        }

 

 

BlockingCollection

    並發集合，在程式操作完之前會一直阻塞其他程式對其進行操作。

 public static void BlockingCollectionTest()        {            Parallel.For<BlockingCollection<string>>(1, 10000,                    () =>                    {                        while (strBlockCollection.Count != 26)                        {                            if (strBlockCollection.Count == 0)                            {                                for (int i = 65; i <= 90; i++)                                {                                    strBlockCollection.Add(Convert.ToChar(i).ToString());                                }                            }                        }                        Console.WriteLine("-------------start------------");                        return strBlockCollection;                    },                    (index, state, head) =>                    {                        string tmp = string.Empty;                        tmp=head.Take();                                                Console.WriteLine(string.Format("The element '{0}' was set by thread {1}", tmp, System.Threading.Thread.CurrentThread.ManagedThreadId));                        head.Add(tmp);                                                return strBlockCollection;                    },                    (result) =>                    {                        Console.WriteLine("-------------end------------------");                    });            Console.WriteLine(string.Format("current collection has {0} elements, Has duplicate data:{1}", strBlockCollection.Count, IsDuplicate<string>(strBlockCollection.AsEnumerable().GetEnumerator())));            foreach (string item in strBlockCollection)            {                Console.WriteLine(item);            }        }