
Software updates typically require system reboots, leading to service downtimes. We aim to solve this problem for network components allowing updates while avoiding service degradation. In this paper, we explore the integration of eBPF into the P4 pipeline for efficient packet processing. This way, we combine the flexibility and dynamic adaptability of eBPF with the efficiency of P4. The integration enhances the power of applications and enables the network operator to provide customizable data paths as a service. Our solution allows updating the data path at runtime and without downtime. We implement the approach for the P4 target T4P4S and discuss different performance models. The evaluation focuses on the overhead in terms of throughput and the costs of code updates expressed in the latency of the related packets. We show that eBPF execution is possible with reasonable costs, promising dynamic network functions within P4.
ddc:004, SDN, P4, Dynamic Network Function, eBPF, 004 Datenverarbeitung; Informatik
ddc:004, SDN, P4, Dynamic Network Function, eBPF, 004 Datenverarbeitung; Informatik
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
