The TCP Outcast Problem: Exposing Unfairness in Data Center Networks
The TCP Outcast Problem: Exposing Unfairness in Data Center Networks
复制标题
DOI：
--
复制
发表时间：
2012-04
期刊：
影响因子：
--
通讯作者：
P. Prakash;A. Dixit;Y. C. Hu;R. Kompella
中科院分区：
文献类型：
--
作者：
P. Prakash;A. Dixit;Y. C. Hu;R. Kompella
在本文中，我们观察到在许多常见的流量模式下，在以多根树拓扑组织的商用数据中心网络中，通过TCP共享带宽会导致严重的不公平，我们称之为TCP Outcast问题。当多个流和少数几个流到达目的地为一个公共输出端口的交换机的两个端口时，这一小部分流将显著损失它们的吞吐量份额(有时几乎损失一个数量级)。Outcast问题主要发生在商用交换机使用的尾部丢弃队列中。通过仔细分析，我们发现尾部丢弃队列表现出一种称为端口中断的现象，即来自一个端口的一系列数据包被丢弃。端口封锁对较少的流量影响更大，因为它们会丢失更多导致TCP超时的连续数据包。在本文中，我们使用一个数据中心网络试验台，使用真实的硬件，在不同的场景下展示了该TCP Outcast问题的存在。然后，我们评估了不同的解决方案，例如RED、SFQ、TCP调步，以及一种称为等长路由的新解决方案，以缓解Outcast问题。
In this paper, we observe that bandwidth sharing via TCP in commodity data center networks organized in multi-rooted tree topologies can lead to severe unfairness, which we term as the TCP Outcast problem, under many common traffic patterns. When many flows and a few flows arrive at two ports of a switch destined to one common output port, the small set of flows lose out on their throughput share significantly (almost by an order of magnitude sometimes). The Outcast problem occurs mainly in taildrop queues that commodity switches use. Using careful analysis, we discover that taildrop queues exhibit a phenomenon known as port blackout, where a series of packets from one port are dropped. Port blackout affects the fewer flows more significantly, as they lose more consecutive packets leading to TCP timeouts. In this paper, we show the existence of this TCP Outcast problem using a data center network testbed using real hardware under different scenarios. We then evaluate different solutions such as RED, SFQ, TCP pacing, and a new solution called equal-length routing to mitigate the Outcast problem.