I/O completion ports are the mechanism by which an application uses a pool of threads that was created when the application was started to process asynchronous I/O requests. These threads are created for the sole purpose of processing I/O requests. Applications that process many concurrent asynchronous I/O requests can do so more quickly and efficiently by using I/O completion ports than by using creating threads at the time of the I/O request.

I/O完成端口是一種機(jī)制，通過(guò)這個(gè)機(jī)制，應(yīng)用程序在啟動(dòng)時(shí)會(huì)首先創(chuàng)建一個(gè)線程池，然后該應(yīng)用程序使用線程池處理異步I/O請(qǐng)求。這些線程被創(chuàng)建的唯一目的就是用于處理I/O請(qǐng)求。對(duì)于處理大量并發(fā)異步I/O請(qǐng)求的應(yīng)用程序，相比在I/O請(qǐng)求發(fā)生時(shí)才創(chuàng)建線程來(lái)說(shuō)，使用完成端口(s)可以做的更快且更有效率。

The CreateIoCompletionPort function associates an I/O completion port with one or more file handles. When an asynchronous I/O operation started on a file handle associated with a completion port is completed, an I/O completion packet is queued to the port. This can be used to combine the synchronization point for multiple file handles into a single object. 

CreateIoCompletionPort函數(shù)使一個(gè)I/O完成端口與一個(gè)或更多的文件句柄發(fā)生關(guān)聯(lián)。當(dāng)在與一個(gè)完成端口相關(guān)的文件句柄上啟動(dòng)的異步I/O操作完成時(shí)，一個(gè)I/O完成包就會(huì)進(jìn)入到該完成端口的隊(duì)列中。對(duì)于多個(gè)文件句柄來(lái)說(shuō)，這種機(jī)制可以用來(lái)把多文件句柄的同步點(diǎn)放在單個(gè)對(duì)象中。（言下之意，如果我們需要對(duì)每個(gè)句柄文件進(jìn)行同步，一般而言我們需要多個(gè)對(duì)象（如：Event來(lái)同步），而我們使用IO Complete Port 來(lái)實(shí)現(xiàn)異步操作，我們可以同多個(gè)文件相關(guān)聯(lián)，每當(dāng)一個(gè)文件中的異步操作完成，就會(huì)把一個(gè)complete package放到隊(duì)列中，這樣我們就可以使用這個(gè)來(lái)完成所有文件句柄的同步）
A thread uses the GetQueuedCompletionStatus function to wait for a completion packet to be queued to the completion port, rather than waiting directly for the asynchronous I/O to complete. Threads that block their execution on a completion port are released in last-in-first-out (LIFO) order. This means that when a completion packet is queued to the completion port, the system releases the last thread to block its execution on the port. 

調(diào)用GetQueuedCompletionStatus函數(shù)，某個(gè)線程就會(huì)等待一個(gè)完成包進(jìn)入到完成端口的隊(duì)列中，而不是直接等待異步I/O請(qǐng)求完成。線程（們）就會(huì)阻塞于它們的運(yùn)行在完成端口（按照后進(jìn)先出隊(duì)列順序的被釋放）。這就意味著當(dāng)一個(gè)完成包進(jìn)入到完成端口的隊(duì)列中時(shí)，系統(tǒng)會(huì)釋放最近被阻塞在該完成端口的線程。
When a thread calls GetQueuedCompletionStatus, it is associated with the specified completion port until it exits, specifies a different completion port, or frees the completion port. A thread can be associated with at most one completion port.
調(diào)用GetQueuedCompletionStatus，線程就會(huì)將會(huì)與某個(gè)指定的完成端口建立聯(lián)系，一直延續(xù)其該線程退出，或被指定了不同的完成端口，或者釋放了與完成端口的聯(lián)系。一個(gè)線程只能與最多不超過(guò)一個(gè)的完成端口發(fā)生聯(lián)系。
The most important property of a completion port is the concurrency value. The concurrency value of a completion port is specified when the completion port is created. This value limits the number of runnable threads associated with the completion port. When the total number of runnable threads associated with the completion port reaches the concurrency value, the system blocks the execution of any subsequent threads that specify the completion port until the number of runnable threads associated with the completion port drops below the concurrency value.

完成端口最重要的特性就是并發(fā)量。完成端口的并發(fā)量可以在創(chuàng)建該完成端口時(shí)被指定。該并發(fā)量限制了與該完成端口相關(guān)聯(lián)的可運(yùn)行線程的數(shù)目。當(dāng)與該完成端口相關(guān)聯(lián)的可運(yùn)行線程的總數(shù)目達(dá)到了該并發(fā)量，系統(tǒng)就會(huì)阻塞任何與該完成端口相關(guān)聯(lián)的后續(xù)線程的執(zhí)行，直到與該完成端口相關(guān)聯(lián)的可運(yùn)行線程數(shù)目下降到小于該并發(fā)量為止。

The most efficient scenario occurs when there are completion packets waiting in the queue, but no waits can be satisfied because the port has reached its concurrency limit. In this case, when a running thread calls GetQueuedCompletionStatus, it will immediately pick up the queued completion packet. No context switches will occur, because the running thread is continually picking up completion packets and the other threads are unable to run. 

最有效的假想是發(fā)生在有完成包在隊(duì)列中等待，而沒(méi)有等待被滿足，因?yàn)榇藭r(shí)完成端口達(dá)到了其并發(fā)量的極限。此時(shí)，一個(gè)正在運(yùn)行中的線程調(diào)用GetQueuedCompletionStatus時(shí)，它就會(huì)立刻從隊(duì)列中取走該完成包。這樣就不存在著環(huán)境的切換，因?yàn)樵撎幱谶\(yùn)行中的線程就會(huì)連續(xù)不斷地從隊(duì)列中取走完成包，而其他的線程就不能運(yùn)行了。
The best value to pick for the concurrency value is the number of CPUs on the machine. If your transaction required a lengthy computation, a larger concurrency value will allow more threads to run. Each transaction will take longer to complete, but more transactions will be processed at the same time. It is easy to experiment with the concurrency value to achieve the best effect for your application. 

對(duì)于并發(fā)量最好的選值就是您計(jì)算機(jī)中CPU的數(shù)目。如果您的事務(wù)處理需要一個(gè)漫長(zhǎng)的計(jì)算時(shí)間，一個(gè)比較大的并發(fā)量可以允許更多線程來(lái)運(yùn)行。雖然完成每個(gè)事務(wù)處理需要花費(fèi)更長(zhǎng)的時(shí)間，但更多的事務(wù)可以同時(shí)被處理。對(duì)于應(yīng)用程序來(lái)說(shuō)，很容易通過(guò)測(cè)試并發(fā)量來(lái)獲得最好的效果。
The PostQueuedCompletionStatus function allows an application to queue its own special-purpose I/O completion packets to the completion port without starting an asynchronous I/O operation. This is useful for notifying worker threads of external events.

PostQueuedCompletionStatus函數(shù)允許應(yīng)用程序可以針對(duì)自定義的專用I/O完成包進(jìn)行排隊(duì)，而無(wú)需啟動(dòng)一個(gè)異步I/O操作。這點(diǎn)對(duì)于通知外部事件的工作者線程來(lái)說(shuō)很有用。

The completion port is freed when there are no more references to it. The completion port handle and every file handle associated with the completion port reference the completion port. All the handles must be closed to free the completion port. To close the port handle, call the CloseHandle function.
在沒(méi)有更多的引用針對(duì)某個(gè)完成端口時(shí)，需要釋放該完成端口。該完成端口句柄以及與該完成端口相關(guān)聯(lián)的所有文件句柄都需要被釋放。調(diào)用CloseHandle可以釋放完成端口的句柄。

本文來(lái)自CSDN博客，http://blog.csdn.net/vieri_ch/archive/2006/10/12/1332054.aspx