TCP Incast Avoidance Based on Connection Serialization in Data Center Networks

摘要

In distributed file systems, a well-known congestion collapse called TCP incast (Incast briefly) occursbecause many servers almost simultaneously send data to the same client and then many packets overflowthe port buffer of the link connecting to the client. Incast leads to throughput degradation in the network. Inthis paper, we propose three methods to avoid Incast based on the fact that the bandwidth-delay product issmall in current data center networks. The first method is a method which completely serializes connectionestablishments. By the serialization, the number of packets in the port buffer becomes very small, whichleads to Incast avoidance. The second and third methods are methods which overlap the slow start periodof the next connection with the current established connection to improve throughput in the first method.Numerical results from extensive simulation runs show the effectiveness of our three proposed methods.