TCP Servers: Offloading TCP Processing in Internet Servers. Design, Implementation, and Performance

TCP Server is a system architecture aiming to offload network processing from the host(s) running an Internet server. The TCP Server can be executed on a dedicated processor, node, or intelligent network interface using lowoverhead, non-intrusive communication between it and the host(s) running the server application. In this paper, we present and evaluate two implementations of the TCP Server architecture: (1) using dedicated network processors on a symmetric multiprocessor (SMP) server and (2) using dedicated nodes on a cluster-based server built around a memory-mapped communication interconnect. We have quantified the impact of offloading on the performance of network servers for these two TCP Server implementations, using server applications with realistic workloads. We were able to achieve performance gains of up to 30% with our SMP-based as well as cluster-based implementations for the scenarios we studied. Based on our experience and results, we conclude that offloading the network processing from the host processor using a TCP Server architecture is beneficial to server performance when the server is overloaded. A complete offloading of the TCP/IP processing requires substantial computing resources on the TCP server. Depending on the application workload, either the host processor or the TCP server can become the bottleneck stressing the need for an adaptive scheme to balance the load between the host and the TCP server.