Dataplaners

Dataplaners are the operational elements that perform the forwarding and processing of user traffic in a network device or software-defined environment. They implement the data plane, handling tasks such as packet forwarding, header parsing, ACL checks, quality of service and shaping, NAT, tunneling, and encryption as configured by higher layers. The dataplane executes the rules that have been installed by the control plane, and it is designed to operate at line rate with minimal latency.

Dataplaners are distinct from the control plane, which runs routing and signaling protocols, makes decisions about where traffic should go, and programs the dataplane with forwarding rules. In traditional devices, dataplanes are implemented in hardware ASICs or NPUs; in modern networks they can also be software-based, running in general-purpose CPUs or in specialized runtimes such as DPDK or VPP. Virtualized dataplanes run within hypervisors or cloud platforms to enable software-defined or network-function-virtualization environments.

Programmability is a growing focus for dataplaners. Languages like P4 are used to define how packets are processed, while Linux eBPF allows customizable, in-kernel dataplane logic. The combination of programmable dataplanes and fast-path implementations enables flexible, scalable, and cost-effective networks, particularly in data centers and cloud environments.

Security considerations for dataplaners include access control, isolation between tenants, and integrity of forwarding tables, as well as the ability to enforce encryption, NAT, and tamper-resistant configurations. Ongoing trends include increased use of programmable NICs, edge dataplaners, and distributed control planes that manage large-scale forwarding.