US 12,328,221 B2
Scalable mechanisms for workload management in edge computing
Braulio Gabriel Dumba, White Plains, NY (US); Gloire Rubambiza, Mountain View, CA (US); and Andrew John Anderson, Stamford, CT (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Jun. 15, 2023, as Appl. No. 18/335,749.
Prior Publication US 2024/0422052 A1, Dec. 19, 2024
Int. Cl. H04L 41/0668 (2022.01); G16Y 10/75 (2020.01); G16Y 30/00 (2020.01); H04L 67/568 (2022.01)
CPC H04L 41/0668 (2013.01) [G16Y 10/75 (2020.01); G16Y 30/00 (2020.01); H04L 67/568 (2022.05)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
maintaining, in an edge site, a plurality of rendezvous nodes to which a plurality of Internet of Things (IoT) devices are communicatively coupled, wherein the plurality of rendezvous nodes are also communicatively coupled to an edge hub in the edge site; and
maintaining a cached model at the edge hub, wherein the cached model enables the edge site to stay operational in an event of a failure of a cloud hub to which the edge hub is communicatively coupled; and
deploying a rendezvous node selection algorithm that takes into consideration cost and resilience trade-offs, wherein the rendezvous node selection algorithm minimizes a number of intermediate nodes, wherein the rendezvous node selection algorithm is scalable in time and space complexity and number of edge endpoints, wherein the plurality of IoT devices are distributed among rendezvous nodes as leaves of a balanced tree, wherein the edge hub maintains a list of virtual identifiers for all reachable rendezvous nodes, wherein one or more of the leaves of the balanced tree are maintained as empty slots for assignment of new IoT devices, and wherein processing load is balanced among the plurality of rendezvous nodes.