Allowing non-volatile memory express (NVMe) over fabric (NVMe-oF) traffic over interfaces using a scalable end point (SEP) addressing mechanism

A computing system comprising:

• a first domain object of a plurality of domain objects;
• a network comprising a plurality of central processing unit (CPU) cores;
• one or more processors;
• and
• memory coupled to the one or more processors to store instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, the operations comprising:
• determine a set of non-volatile memory express (NVMe) queue pairs of CPU core connections of CPU cores in the network;
• generate a set of transmit and receive contexts for the first domain object aligned with the CPU core connections;
• generate a scalable end point (SEP) for the first domain object using an address vector for the domain object, and wherein the address vector for the domain object stores remote addresses for the CPU cores in the network and the domain object;
• map, using the SEP, the set of transmit and receive contexts to the set of NVMe queue pairs of the CPU core connections to the domain object;
• using the mapping between the set of transmit and receive contexts to the set of NVMe queue pairs of the CPU core connections and the domain object, transfer data through the network;
• and
• perform an operation between a host device of the domain object and a client CPU core using the SEP, wherein the operation comprises:
• generating a first completion event for a first completion queue associated with the client CPU core based on sending a request message, from the client CPU core to the host device, with a unique key tag using the SEP;
• and
• generating a second completion event for the first completion queue after completion of the operation.

A non-transitory storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

• determine a set of non-volatile memory express (NVMe) queue pairs of central processing unit (CPU) core connections of CPU cores in a network;
• generate a set of transmit and receive contexts for a domain object that are aligned with the set of NVMe queue pairs of CPU core connections of CPU cores;
• generate a scalable end point (SEP) for the domain object using an address vector for the domain object, and wherein the address vector for the domain object stores remote addresses for the CPU cores in the network and the domain object;
• map, using the SEP, the set of transmit and receive contexts to the set of NVMe queue pairs of the set of CPU core connections to the domain object;
• using the mapping between the set of transmit and receive contexts to the set of NVMe queue pairs of the set of CPU core connections and the domain object, transfer data through the network;
• and
• perform an operation between a host device of the domain object and a client CPU core using the SEP, wherein the operation comprises:
• generating a first completion event for a first completion queue associated with the client CPU core based on sending a request message, from the client CPU core to the host device, with a unique key tag using the SEP;
• and
• generating a second completion event for the first completion queue after completion of the operation.