P4Tester
Yu Zhou,Jun Bi,Yunsenxiao Lin,Yangyang Wang,Dai Zhang,Zhaowei Xi,Jiamin Cao,Chen Sun
DOI: https://doi.org/10.1145/3326285.3329040
2019-01-01
Abstract:P4 and programmable data planes bring significant flexibility to network operation but are inevitably prone to various faults. Some faults, like P4 program bugs, can be verified statically, while some faults, like runtime rule faults, only happen to running network devices, and they are hardly possible to handle before deployment. Existing network testing systems can troubleshoot runtime rule faults via injecting probes, but are insufficient for programmable data planes due to large overheads or limited fault coverage. In this paper, we propose P4Tester, a new network testing system for troubleshooting runtime rule faults on programmable data planes. First, P4Tester proposes a new intermediate representation based on Binary Decision Diagram, which enables efficient probe generation for various P4-defined data plane functions. Second, P4Tester offers a new probe model that uses source routing to forward probes. This probe model largely reduces rule fault detection overheads, i.e. requiring only one server to generate probes for large networks and minimizing the number of probes. Moreover, this probe model can test all table rules in a network, achieving full fault coverage. Evaluation based on real-world data sets indicates that P4Tester can efficiently check all rules in programmable data planes, generate 59% fewer probes than ATPG and Pronto, be faster than ATPG by two orders of magnitude, and troubleshoot multiple rule faults within one second on BMv2 and Tofino.