Server fabric adapter for I/O scaling of heterogeneous and accelerated compute systems

A method comprising:

• providing a server fabric adapter (SFA) comprising a processor, a plurality of endpoint ports, and a plurality of host ports, the SFA communicatively coupled to a plurality of controlling hosts, a plurality of endpoints, and a plurality of network ports;
• identifying, by the SFA, a first set of endpoint ports from the plurality of endpoint ports that connect the SFA to the plurality of endpoints, wherein each respective endpoint port of the first set of endpoint ports has a respective queue for an endpoint from the plurality of endpoints to submit one or more scatter-gather lists (SGLs);
• identifying, by the SFA, a second set of host ports from the plurality of host ports that connect the SFA to a controlling host of the plurality of controlling hosts, wherein each respective host port of the second set of host ports includes one or more respective header queues;
• and
• combining, by the SFA, data from the header queues with the SGLs to generate network packets.