Programmable transport protocol architecture

At least one non-transitory computer-readable medium, comprising instructions stored thereon, that if executed by one or more processors, cause the one or more processors to:

• configure a plurality of programmable event processors of a network interface device programmed to perform one or more transport layer protocols, wherein:
• the network interface device comprises a bus and a compute complex comprising a plurality of cores,
• the plurality of programmable event processors are connected in a pipeline manner,
• the plurality of programmable event processors are configurable to perform event recirculation through at least one of the plurality of programmable event processors and bypass of a programmable event processor,
• the plurality of programmable event processors are reprogrammable,
• the plurality of programmable event processors are to perform memory accesses separate from compute operations,
• the plurality of programmable event processors are to group one or more events into at least one group,
• the plurality of programmable event processors are to enforce atomic processing of other events within a group of the at least one group, wherein the atomic processing comprises propagation of state changes to among events of the group,
• the plurality of programmable event processors are to perform parallel processing of events belonging to different groups, and
• based on transmission of packets, the plurality of programmable event processors are programmed to perform at least one transport protocol processing operation associated with the packets based on an event graph description with defined nodes.

A method comprising:

• in a datacenter:
• a plurality of programmable event processors of a network interface device programmed to perform one or more transport layer protocols, wherein:
• the network interface device comprises a compute complex comprising a bus and a plurality of cores,
• the plurality of programmable event processors are connected in a pipeline manner,
• the plurality of programmable event processors are configurable to perform event recirculation through at least one of the plurality of programmable event processors and bypass of a programmable event processor,
• the plurality of programmable event processors are to perform memory accesses separate from compute operations,
• the plurality of programmable event processors are to group one or more events into at least one group,
• the plurality of programmable event processors are to enforce atomic processing of other events within a group of the at least one group, wherein the atomic processing comprises propagation of state changes to among events of the group,
• the plurality of programmable event processors are to perform parallel processing of events belonging to different groups, and
• based on transmission of packets, the plurality of programmable event processors programmed for performing at least one transport protocol processing operation associated with the packets based on an event graph description with defined nodes.