Through calculating the average queue size, Random Early Detection (RED) algorithm can drop packets with relevant probability, so congestion resulted from TCP flows can be avoided.
随机早检测算法通过计算TCP流的平均队列长度，进行适当的概率丢弃分组，从而有效地避免了由TCP流导致的网络拥塞。
It has the following advantages: (1) RED losses packets when the average queue size exceeds the preset threshold that can control the average queue size efficiently and also confine the mean delay.
它具有下述优点：(1)RED在平均队长超过了事先设定的阈值后就丢包，有效地控制了平均队长，限制了平均时延的大小。
When choosing values, you need to determine the average size and average number of messages that will be on the queue.
选择值时，您需要确定队列上的平均消息大小以及队列上的平均消息数量。
The average size of the retransmit queue can be calculated from the packets scanned counter if the number of SACK options sent is also tracked at the client.
如果在客户机可以跟踪发出的SACK选项的数目，则重传队列的大小可从扫描包计数器中计算得到。
RED can absorb the bursty flow by keeping lower average queue size and weaken the bias against bursty flow.
RED算法能够通过保持较低的平均队长吸收突发流，削减了对突发流的偏见。
Linear RED is simple and easy to calculate; however, when average queue size is near to the minimum and maximum threshold, the loss rate is unreasonable.
线性RED算法虽然简单且容易计算，但队列位于最小阈值和最大阈值附近时的丢包概率都不太合理。
Linear RED is simple and easy to calculate; however, when average queue size is near to the minimum and maximum threshold, the loss rate is unreasonable.
线性RED算法虽然简单且容易计算，但队列位于最小阈值和最大阈值附近时的丢包概率都不太合理。
应用推荐