Refereed International Conference Publications

Fine-Grained Pipeline Parallelization for Network Function Programs [abstract] (IEEE Xplore, PDF)
Seungbin Song, Heelim Choi, and Hanjun Kim
Proceedings of the 2021 International Symposium on Code Generation and Optimization (CGO), March 2021.

Network programming languages enable programmers to implement new network functions on various hard- ware and software stacks in the domain of Software Defined Networking (SDN). Although the languages extend flexibility of network devices, existing compilers do not fully optimize the network programs due to their coarse-grained parallelization model. The compilers consider each packet processing table that consists of match and action operations as a unit of tasks, and parallelize the programs without decomposing match and action functions. This work proposes a new fine-grained pipeline parallelization compiler for network programming languages, named PSDN. First, the PSDN compiler decouples match and action functions from packet processing tables, and analyzes dependencies among the matches and actions. Respecting the dependencies, the compiler efficiently schedules each match and action block into a pipeline with clock cycle estimation, and fuses blocks to reduce synchronization overheads. This work implements the PSDN compiler that translates a P4 network function program to a Xilinx PX program which is synthesizable to NetFPGA-SUME hardware. The proposed compiler reduces packet processing latency by 12.1% and utilization by 3.5% compared to the previous work.