Design and Implementation of a Stateful Network Packet Processing Framework for GPUs
Abstract: Graphics processing units (GPUs) are a powerful platform for building the highhigh speed network traffic processing applications using low low-cost cost hardware. The existing systems tap the massively parallel architecture of GPUs to speed up certain computationally intensive ntensive tasks, such as cryptographic operations and pattern matching. However, they still suffer from significant overheads due to critical-path path operations that are still being carried out on the CPU, and redundant inter-device device data transfers. In this pap paper, er, we present GASPP, a programmable network traffic processing framework tailored to modern graphics processors. GASPP integrates optimized GPU GPU-based based implementations of a broad range of operations commonly used in the network traffic processing applications, applicatio including the first purely GPU GPU-based based implementation of network flow tracking and TCP stream reassembly. GASPP also employs novel mechanisms for tackling the control flow irregularities across SIMT threads, and for sharing the memory context between the network interfaces and the GPU. Our evaluation shows that GASPP can achieve multigigabit traffic forwarding rates even for complex and computationally intensive network operations, such as stateful traffic classification, intrusion detection, and packet eencryption. ncryption. Especially when consolidating multiple network applications on the same system, GASPP achieves