sPIN: High-performance streaming Processing in the Network

TL;DR

sPIN introduces a portable programming model that offloads packet processing to network cards, significantly improving communication performance and application speedups in large-scale computing systems.

Contribution

The paper presents sPIN, a novel model for offloading packet processing to network cards, enabling flexible acceleration of communication and system services.

Findings

Full-application speedups demonstrated with sPIN

Accelerates in-memory filesystems and communication operations

Enables a new ecosystem for network acceleration

Abstract

Optimizing communication performance is imperative for large-scale computing because communication overheads limit the strong scalability of parallel applications. Today's network cards contain rather powerful processors optimized for data movement. However, these devices are limited to fixed functions, such as remote direct memory access. We develop sPIN, a portable programming model to offload simple packet processing functions to the network card. To demonstrate the potential of the model, we design a cycle-accurate simulation environment by combining the network simulator LogGOPSim and the CPU simulator gem5. We implement offloaded message matching, datatype processing, and collective communications and demonstrate transparent full-application speedups. Furthermore, we show how sPIN can be used to accelerate redundant in-memory filesystems and several other use cases. Our work investigates a portable packet-processing network acceleration model similar to compute acceleration with CUDA or OpenCL. We show how such network acceleration enables an eco-system that can significantly speed up applications and system services.