利用有效的缓存机制，可以避免到数据库的网络往返时间，绕过占用很多资源的计算，并节省服务器资源，同时改善响应时间和等待时间。
Effective cache mechanism can avoid round trip time to database, avoid calculations which occupying much time, save server resource, as well as improving the response time and waiting time.
然而，对于一些高速网络，如千兆位的以太网和万兆位的以太网，预期的往返时间(因此重新传输超时)要低得多。
However, with high-speed networks, such as Gigabit Ethernet and 10 Gigabit Ethernet, the round-trip time (hence retransmission timeout) is expected to be much lower.
为了缓解TCP拥塞控制机制对于往返时间(rtt)较小流的偏向问题，现有方法在网络拥塞程度严重时由于对rtt较大流的过分保护而不能有效地工作。
To mitigate the bias of TCP congestion mechanisms to short RTT flows, the existing methods cannot work well when network congestion level is high due to the over protecting to long RTT flows.
为了缓解TCP拥塞控制机制对于往返时间(rtt)较小流的偏向问题，现有方法在网络拥塞程度严重时由于对rtt较大流的过分保护而不能有效地工作。
To mitigate the bias of TCP congestion mechanisms to short RTT flows, the existing methods cannot work well when network congestion level is high due to the over protecting to long RTT flows.
应用推荐