自动应答延迟(秒)。
结果表明，在同等条件下，此算法能够达到较小的平均应答延迟，从而优于文中提到的其他算法。
The result indicates that ALB algorithm can get the shorter mean response time in the same condition, so it is better than the other ones in the paper.
延迟应答大大增加了重传队列的长度，这为攻击提供了必要条件。
Delaying acknowledgments effectively increases the size of the retransmission queue, a prerequisite for the attack.
在传输延迟的应答时客户机选择SACK选项的值。
The client selects the value of its SACK options whenever transmitting a delayed ACK.
对于非sack测试用例，我使用客户机报告的延迟应答队列的典型长度作为发送方窗口大小的近似值。
For the non-SACK test cases, I used the typical length of the delayed ACK queue reported by the client as an approximation of the lower bound of the sender's window size.
由靠近用户的缓存对请求作出应答，可以减少网络跃距和网络延迟，从而尽可能地提高响应能力。
Replying to requests by caches as close to the user as possible improves responsiveness by reducing network hops and network latency.
可是，攻击者通过在传输之前使每个应答阻塞一小段时间就可任意增加延迟。
The attacker can, however, arbitrarily add delay by stalling each acknowledgment a little while before transmission.
高延迟会导致在任何给定时刻都有大量正在传送的包在等待应答。
Because of the high delay, there are often large Numbers of packets "in flight" waiting for acknowledgment at any given time.
服务器发来的每个包在经过1750ms的延迟后得到应答。
Each packet from the server is acknowledged after a 1750ms delay.
服务器发来的每个包在经过1750ms的延迟后得到应答。
Each packet from the server is acknowledged after a 1750ms delay.
应用推荐