TCP incast avoidance based on connection serialization in data center networks

In distributed file systems, a well-known congestion collapse called TCP Incast occurs because many servers send data to the same client and then many packets overflow the port buffer of the client link. Incast leads to throughput degradation in the network. In this paper, we propose two methods to avoid Incast based on the fact that only one connection can almost fully use the network bandwidth because the bandwidth-delay product is not so large in current data center networks. The first method is a method which completely serializes connection establishments. By the serialization, the number of packets in the port buffer becomes very small, which leads to Incast avoidance. The second method is a method which overlaps the slow start period of the next connection to the current established connection to improve throughput in the first method. Numerical results from extensive simulation runs show the effectiveness of our two proposed methods.