Outline
/// 进程 线程 多线程 计算机概念
/// 进程：一个程序运行时，占用的全部计算资源的总和
/// 线程：程序执行流的最小单位；任何操作都是由线程完成的；
/// 线程是依托于进程存在的，一个进程可以包含多个线程；
/// 线程也可以有自己的计算资源
/// 多线程：多个执行流同时运行
/// 1 CPU太快了，分时间片--上下文切换(加载环境--计算--保存环境)
/// 微观角度，一个核同一时刻只能执行一个线程；宏观的来说是多线程并发
/// 2 多CPU多核 可以独立工作
/// 4核8线程--核是物理的核 线程是指虚拟核
///
/// Thread是C#语言对线程对象的封装
///
/// 是对方法执行的描述
/// 同步：完成计算之后，再进入下一行
/// 异步：不会等待方法的完成，会直接进入下一行 非阻塞
///
/// C# 异步和多线程有什么差别
/// 多线程就是多个thread并发；
/// 异步是硬件式的异步
/// 异步多线程--thread pool task
/// 异步方法
/// 1 同步方法卡界面，主(UI)线程忙于计算；
/// 异步多线程方法不卡界面，主线程完事儿了，计算任务交给子线程在做；
/// winform提升用户体验；web一个业务操作后要发邮件，异步发送邮件
///
/// 2 同步方法慢，只有一个线程干活；
/// 异步多线程方法快，因为多个线程并发运算；
/// 并不是线性增长，a资源换时间，可能资源不够 b 多线程也有管理成本
/// 但并不是越多越好
/// 多个独立任务可以同时运行；
///
/// 3 异步多线程无序：启动无序 执行时间不确定 结束也无序
/// 一定不要通过等几毫秒的形式来控制启动/执行时间/结束
///
/// 回调/状态等待/信号量
Callback
callback callback is executed after executing the action, the third parameter "hao" by Console.WriteLine (ia.AsyncState); displayed.
Action<string> action = this.DoSomethingLong;
//IAsyncResult asyncResult = null;
//AsyncCallback callback = ia =>
//{
// //Console.WriteLine(object.ReferenceEquals(asyncResult, ia));
// //Console.WriteLine(ia.AsyncState);
// //Console.WriteLine($"到这里计算已经完成了。{Thread.CurrentThread.ManagedThreadId.ToString("00")}。");
//};
//asyncResult = action.BeginInvoke("btnAsyncAdvanced_Click", callback, "hao");
c # multithreading development of three major upheaval, respectively, thread, threadpool, task
at present is the main task, the first two do not record the
Task Profile
When using multiple threads? Tasks can run concurrently; enhance the speed; optimization experience
Example:
//// a business query multiple data sources Home - multithreading - to return WaitAll After getting all the data
//// how a product search elements operating data sources, commodity search - multiple data source - multithreading - need only one result can -WaitAny
Parallel tasks: parallel
Multithreading Exception Handling
In the last try inside plus a Task.WaitAll (taskList.ToArray ()); in order to capture abnormal, type AggregateException,
or the thread inside the anomaly will be swallowed up, as it has been out of range of the try catch
Thread Cancellation
Multiple threads concurrently, after a failure, wishes to inform the other threads are stopped
task is outside can not be aborted, Thread.Abort do not fly, because the thread is OS resources, can not control Shashi cancel
threads themselves stop myself - Public access variable - modify it - threads constantly detect it (and ultimately delayed)
////CancellationTokenSource去标志任务是否取消 Cancel取消 IsCancellationRequested 是否已经取消了
////Token 启动Task的时候传入，那么如果Cancel了，这个任务会放弃启动，抛出一个异常
//CancellationTokenSource cts = new CancellationTokenSource();//bool值 //bool flag = true;
//for (int i = 0; i < 40; i++)
//{
// string name = string.Format("btnThreadCore_Click{0}", i);
// Action<object> act = t =>
// {
// try
// {
// //if (cts.IsCancellationRequested)
// //{
// // Console.WriteLine("{0} 取消一个任务的执行", t);
// //}
// Thread.Sleep(2000);
// if (t.ToString().Equals("btnThreadCore_Click11"))
// {
// throw new Exception(string.Format("{0} 执行失败", t));
// }
// if (t.ToString().Equals("btnThreadCore_Click12"))
// {
// throw new Exception(string.Format("{0} 执行失败", t));
// }
// if (cts.IsCancellationRequested)//检查信号量
// {
// Console.WriteLine("{0} 放弃执行", t);
// return;
// }
// else
// {
// Console.WriteLine("{0} 执行成功", t);
// }
// }
// catch (Exception ex)
// {
// cts.Cancel();
// Console.WriteLine(ex.Message);
// }
// };
// taskList.Add(taskFactory.StartNew(act, name, cts.Token));
//}
//Task.WaitAll(taskList.ToArray());
Multithreading temporary variables
//i最后是5 全程就只有一个i 等着打印的时候，i==5
//k 全程有5个k 分别是0 1 2 3 4
//k在外面声明 全程就只有一个k 等着打印的时候，k==4
//int k = 0;
//for (int i = 0; i < 5; i++)
//{
// k = i;
// new Action(() =>
// {
// Thread.Sleep(100);
// Console.WriteLine($"k={k} i={i}");
// }).BeginInvoke(null, null);
//}
And asynchronous execution threads to stop 100, i have read the final run has since increased to 5
Thread Safety
// lock resolved, because only one thread can go in, there is no concurrency, so to solve the problem but at the expense of performance, so try to narrow the scope of the lock
// do not conflict - split the data, to avoid conflict
// secure a thread to queue ConcurrentQueue complete the operation
https://blog.csdn.net/jiangxinyu/article/details/5380969
