Deep network slicing with virtualized programmable data-plane pipelines

A system configured for network slicing with programmable data-plane pipelines, comprising:

• network switch hardware;
• one or more hardware processors configured by machine-readable instructions to:
• create slice contexts, wherein creating the slice contexts comprises:
• create control-plane data for each of a plurality of network slices;
• create data-plane data for each of the plurality of network slices;
• and
• store the control-plane data and the data-plane data for each of the plurality of network slices to produce slice contexts;
• map resources of the network switch hardware to the slice contexts to create a resource mapping for the slice contexts;
• instantiate the plurality of network slices on the network switch hardware using the resource mapping;
• receive frames from one or more external sources;
• identify a network slice corresponding to each frame;
• and
• enrich a header of each of the frames, based upon the identification of the network slice for each frame, to direct each frame to a data-plane pipeline.

A non-transitory, tangible processor readable storage medium, encoded with processor readable instructions to perform a method for network slicing with programmable data-plane pipelines, the method comprising:

• creating slice contexts, wherein creating the slice contexts comprises:
• creating control-plane data for each of a plurality of network slices;
• creating data-plane data for each of the plurality of network slices;
• and
• storing the control-plane data and the data-plane data for each of the plurality of network slices to produce slice contexts;
• mapping resources of network switch hardware to the slice contexts to create a resource mapping for the slice contexts;
• instantiating the plurality of network slices on the network switch hardware using the resource mapping;
• receiving frames from one or more external sources;
• identifying a network slice corresponding to each frame;
• and
• enriching a header of each of the frames, based upon the identification of the network slice for each frame, to direct each frame to a data-plane pipeline.