Network-on-chip computing systems with wireless interconnects

1. A computing device, comprising:

a substrate;

a plurality of computing cores on the substrate, the individual computing cores including one or more computing processors, wherein the computing cores are organized into a plurality of subnets individually associated with:

a set of the computing cores;

a communications hub associated with the set of the computing cores; and

a plurality of conductive or semi-conductive connectors connecting the set of the computing cores into a small world network and connecting the communications hub to the set of the computing cores; and

a plurality of wireless transceivers on the substrate, the wireless transceivers being individually associated with one of the communications hubs of a corresponding subnet.

2. The computing device of claim 1 wherein the plurality of conductive or semi-conductive interconnects connect the set of the computing cores and the communications hub in a star ring structure.

3. The computing device of claim 1 , further comprising a plurality of conductive or semi-conductive interconnects connecting adjacent pairs of the subnets in a mesh structure.

4. The computing device of claim 1 , further comprising a plurality of conductive or semi-conductive interconnects connecting the communications hubs of adjacent pairs of the subnets in a mesh structure.

5. The computing device of claim 1 , further comprising a conductive or semi-conductive interconnect connecting the communications hub of a first one of the subnets and the communications hub of a second one of the subnets, the first one of the subnets being associated with one of the wireless transceivers and the second one of the subnets not being associated with one of the wireless transceivers.

6. The computing device of claim 1 wherein a propagation length between a pair of the wireless transceivers is greater than a threshold, the propagation length representing a number of hops between the subnets associated with the pair of the wireless transceivers via the conductive or semi-conductive interconnects.

7. The computing device of claim 1 wherein a number of the wireless transceivers is less than a number of the subnets, and wherein a propagation length between a pair of the wireless transceivers is greater than two, the propagation length representing a number of hops between the subnets associated with the pair of the wireless transceivers via the conductive or semi-conductive interconnects.

8. A method of routing communications messages on a computing device having a plurality of computing cores on a common substrate, the computing cores being interconnected via conductive or semi-conductor connectors into a plurality of subnets individually associated with a set of the computing cores and a communications hub, at least some of the subnets having a wireless transceiver at the communications hub, the method comprising:

determining a pair of the subnets each having a wireless transceiver at the corresponding communications hub, each of the subnets being closest to a source computing core and a destination computing core, respectively;

determining if a propagation length between the source and destination computing cores is shorter via the first and second subnet with corresponding wireless transceivers or via the conductive or semi-conductor connectors; and

in response to determining that the propagation length between the source and destination computing cores is shorter via the first and second subnet with corresponding wireless transceivers, routing communications between the source and destination computing cores via the first and second subnets with the corresponding wireless transceivers.

9. The method of claim 8 wherein the propagation length is represented by a number of hops between the source and destination computing cores, each hop corresponding to a communications distance between a pair of adjacent computing cores.

10. The method of claim 8 wherein in response to determining that the propagation length between the source and destination computing cores is shorter via the conductive or semi-conductive connectors, routing communications between the source and destination computing cores via the conductive or semi-conductive connectors.

11. The method of claim 8 , further comprising:

determining if a wireless channel is available between the first and second subnets; and

in response to determining that a wireless channel is available between the first and second subnets, routing communications between the source and destination computing cores via the first and second subnets with the corresponding wireless transceivers.

12. The method of claim 8 , further comprising:

determining if a wireless channel is available between the first and second subnets; and

in response to determining that a wireless channel is not available between the first and second subnets, assigning a computing core adjacent to the source computing core as a new source computing core and repeating the determining operations with the new source computing core.

13. The method of claim 8 , further comprising:

determining if the source and destination computing cores are in the same subnet; and

in response to determining that the source and destination computing cores are in the same subnet, routing communications between the source and destination computing cores via the conductive or semi-conductive connectors in the same subnet.

14. The method of claim 8 , further comprising:

determining if the source and destination computing cores are in the same subnet;

in response to determining that the source and destination computing cores are in the same subnet, routing communications between the source and destination computing cores via the conductive or semi-conductive connectors in the same subnet; and

in response to determining that the source and destination computing cores are not in the same subnet, performing the determining operations.

15. A computing device having a plurality of computing cores on a common substrate, the computing cores being interconnected via conductive or semi-conductor connectors into a plurality of subnets individually associated with a set of the computing cores and a communications hub, at least some of the subnets having a wireless transceiver at the communications hub, the computing device further includes a memory containing instructions executable by one or more of the computing cores to cause the computing device to:

determine a pair of the subnets each having a wireless transceiver at the corresponding communications hub, each of the subnets being closest to a source computing core and a destination computing core, respectively;

determine if a propagation length between the source and destination computing cores is shorter via the first and second subnet with corresponding wireless transceivers or via the conductive or semi-conductor connectors; and

in response to determining that the propagation length between the source and destination computing cores is shorter via the first and second subnet with corresponding wireless transceivers, route communications between the source and destination computing cores via the first and second subnets with the corresponding wireless transceivers.

16. The computing device of claim 15 wherein the propagation length is represented by a number of hops between the source and destination computing cores, each hop corresponding to a communications distance between a pair of adjacent computing cores.

17. The computing device of claim 15 wherein the memory contains additional instructions executable by one or more of the computing cores to cause the computing device to route communications between the source and destination computing cores via the conductive or semi-conductive connectors in response to determining that the propagation length between the source and destination computing cores is shorter via the conductive or semi-conductive connectors.

18. The computing device of claim 15 wherein the memory containing additional instructions executable by one or more of the computing cores to cause the computing device to:

determine whether a wireless channel is available between the first and second subnets; and

in response to determining that a wireless channel is available between the first and second subnets, route communications between the source and destination computing cores via the first and second subnets with the corresponding wireless transceivers.

19. The computing device of claim 15 wherein the memory containing additional instructions executable by one or more of the computing cores to cause the computing device to:

determine whether a wireless channel is available between the first and second subnets; and

in response to determining that a wireless channel is not available between the first and second subnets, assign a computing core adjacent to the source computing core as a new source computing core and repeating the determining operations with the new source computing core.

20. The computing device of claim 15 wherein the memory containing additional instructions executable by one or more of the computing cores to cause the computing device to:

determine whether the source and destination computing cores are in the same subnet; and

in response to determining that the source and destination computing cores are in the same subnet, route communications between the source and destination computing cores via the conductive or semi-conductive connectors in the same subnet.