网络小计2

1.对于要处理众多小连接的服务器，需要投递更多的AcceptEx

检查系统是否有足够AcceptEx的一个方法是：

将WSAEventSelect关联到FD_ACCEPT,如果有连接到来而没有足够的AcceptEx时，注册事件会被触发

AcceptEx的一个好处是：接收连接同时会接受数据，For servers whose clients send an initial request this is ideal

但，除非AcceptEx接收到至少一个字节，否则the AcceptEx operation will not complete ，所以需要循环检查所有的连接时间，

call getsockopt with SO_CONNECT_TIME， which will return regardless of whether the socket is actually connected. If it is connected, the return value is greater than zero. A value of -1 indicates it is not connected

警告：

 As a word of warning, applications should not under any circumstances close a client socket handle used in anAcceptEx call that has not been accepted because it can lead to memory leaks. For performance reasons, the kernel-mode structures associated with anAcceptEx call will not be cleaned up when the unconnected client handle is closed until a new client connection is established or until the listening socket is closed.

警告：

虽然在完成端口的线程中投递AcceptEx看起来比较简单，也符合逻辑，但你应该避免这样做。

另外，工作线程中应该避免任何复杂的运算，让工作线程更快的处理事件

原因：

创建套接字是expensive的，因为WInsock2.0的有多层结构，当服务器创建一个套接字时，要层层调用。

所以，服务器应该在一个单独的线程里面创建套接字和投递AcceptEx,When an overlapped AcceptEx completes in the worker thread, an event can be used to signal the accept issuing thread.

Data Transfers:

默认情况下，每一个套接字都有一个发送和接收缓冲区

发送：

如果协议发送缓冲区满了，程序的发送缓冲区将被锁定，send调用将返回WSA_IO_PENDING

在处理完发送缓冲区的数据后（例如，将数据交给下层TCP处理），winsock将直接处理锁定的缓冲区，也就是说，应用程序将从缓冲区直接把数据交给TCP，绕过套接字的发送缓冲区

接收：

如果套接字的接收缓冲区是空的，当overlapped receive被调用时，程序的缓冲区将被锁定，并且失败，返回WSA_IO_PENDING,当数据来到时，将被直接复制到程序的缓冲区，绕过了socket的接收缓冲区

缓冲区模型（我猜想的）

Application Buffer

Socket Buffer

Tcp(Driver) Buffer

将SOCKET的 buffer 设为0 通常不能提供性能，因为只要有足够的overlapped send 和 receive 操作被投递，就可以避免额外的内存拷贝

禁用套接字的发送缓冲区通常比禁用接收缓冲区的影响小

因为：

Application Buffer将总是被锁定直到被传递到下层TCP处理，然而，如果接收缓冲区被设为0而没有足够的overalpped receive被调用

所有传来的数据将被缓冲在TCP层的Buffer，TCP Driver将缓存最多 receive window size -- 17K 的数据，这是极限，

通常 缓冲区要小得多，这些TCP缓冲区被分配在non-paged pool中，如果连接太多，这些non-paged pool 将被耗尽

只有几个特定情况不设定socket receive buffer才会降低性能:服务器有大量连接，而客户端只偶尔发送数据。

资源管理：

每一个overlapped send 或者 recive 操作的buffer都会被锁定，即it cannot be paged out of physical memory

可被系统锁定的内存是有限的，When this limit is reached, overlapped operations will fail with theWSAENOBUFS error

如果一个服务器要处理巨量的并发连接，服务器可以给每个连接投递长度为0的receive操作，通过这种方法

将没有内存被锁定，the per-socket receive buffer should be left intact because once the zero-byte receive operation completes, the server can simply perform a non-blocking receive to retrieve all the data buffered in the socket's receive buffer. There is no more data pending when the non-blocking receive fails with WSAEWOULDBLOCK

这种设计适用于要求最大化并发数，但可牺牲吞吐量的情况。

a non-blocking receive is performed once the zero-byte receive completes to retrieve the buffered data. If the server knows that clients send data in bursts, then once the zero-byte receive completes, it may post one or more overlapped receives in case the client sends a substantial amount of data (greater than the per-socket receive buffer that is 8 KB by default).

一个连接的套接字消耗2KB的 non-paged pool

accpet返回的套接字消耗1.5KB的non-paged pool

每一个overlapped操作消耗大约500bytes的non-paged pool

服务器策略：

服务器带该可以分为2种：

1.High Throughput:

主要考虑在少量连接上传输数据，当然这个“少量”是相对服务器可接受的连接而言的。

2.高并发

主要考虑多并发而不发送大量数据的连接