US 11,943,284 B2
Overload protection for edge cluster using two tier reinforcement learning models
Zhongwen Zhu, Saint-Laurent (CA); Hatem Abou-zeid, Calgary (CA); Anirudha Jitani, Montreal (CA); and Aditya Mahajan, Montreal (CA)
Assigned to TELEFONAKTIEBOLAGET LM ERICSSON (PUBL), Stockholm (SE)
Appl. No. 18/027,801
Filed by Telefonaktiebolaget LM Ericsson (publ), Stockholm (SE)
PCT Filed Oct. 9, 2020, PCT No. PCT/IB2020/059523
§ 371(c)(1), (2) Date Mar. 22, 2023,
PCT Pub. No. WO2022/074439, PCT Pub. Date Apr. 14, 2022.
Prior Publication US 2023/0344899 A1, Oct. 26, 2023
Int. Cl. H04L 67/1008 (2022.01); G06N 20/00 (2019.01)
CPC H04L 67/1008 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method of overload protection implemented by a network node in an edge cluster comprising two or more network nodes, the method comprising:
receiving, from a client device, a client request for an application or service provided by the network node;
determining, based on a node-level routing policy for the network node, to redirect or proxy the client request; and
selecting, based on a cluster-level redirection policy or cluster-level proxy policy applicable to the network nodes in the edge cluster, a target network node in the edge cluster to handle the client request.