Edgemesh - P2P

[Hernán Humberto Álvarez Valera]

Hi All,

I hope this message finds you well. 

I've been exploring KubeEdge and its application in edge computing environments, and I've come across a few concepts that I'm finding challenging to fully comprehend. 

From my understanding, a set of end devices can connect to an edge node in KubeEdge, which in turn can communicate with other subnets using Tunnel (libP2P). I'm particularly interested in understanding how this setup influences or facilitates service discovery, load balancing, and application deployment. 

I came across an article that mentioned the use of a round-robin method for load balancing within KubeEdge. Is this the primary method used, or are there other techniques in play? 

Also, concerning application deployment, is this always managed by the control plane in the cloud, or are there instances where this is handled differently? 

I apologize if my questions seem basic. I'm in the early stages of familiarizing myself with KubeEdge and any insights or resources you could share to help me better understand these concepts would be greatly appreciated. 

Thank you in advance for your assistance. 

Best Regards,