Need Info / Sources / People to be able to interview about Botnets, Clusters, and P9 CPU Servers

I am currently designing something that, functionally, you could call a botnet, however the point is to act like a WAN wide compute network, similar to how you used to connect to a mainframe back in the day. However, your client, or list of clients, that you hook up to the network, then, thusly, are part of the mainframe.

The purpose of this is twofold. However, to understand this, I need to explain something.

I have found that within the last few months I can happily use a pentium 3 laptop, running Slitaz, and actually get things done. 800mhz, 256mb ram, 40GB HDD, but I was super happy because the system used 48mb ram, looked great on the display, and the entire system kept up.

Separately, we have a problem in nature where wildlife is disappearing. We're running out of fish, deer, we've hunted them out of existence.

I see a similar issue with buying CPU's. People buying too much capacity, creating more demand for crazier chips, causing more environmental issues. Hell we're having random forest fires because the oxygen, nitrogen, carbon, and other airid resources are all screwed up from industry swirling around everywhere.

What I propose is that people get a computer they like, and keep it. If we use something like Mainframe type networks, and P9 like CPU servers, we can spread the load. This will push for the US Gov to put fibre out (like they were supposed to), it gives a use to older electronics, it allowes people in areas where there AREN'T upgrades to continue doing things on their computers, and on top of it all, it keeps manufacturing within reasonable limits.

But, what the leading problem is with my network design so far is if a bad actor COULD control all of the Nodes (theres a lot of explanation to go with nodes, if you want to help me I will explain later), then you basically have Zeus all over again.

My solution so far to this is basing everything off ring limits and Node limits. If your CPU physically can't do the job, it tosses it. If you're a user, you can only request data. If you're a processor, you can sift data. If you are a controller, you control data. Everything is tracked through 2 or 3 different types of blockchains, all run separate, and management is done based off requests from other users in the mesh network. Theres a lot more detail to this, but thats the gist of it.

Ironically I went over the design with GPT the other day and it thought it'd be great if everything could stay balanced, and it'd be a ridiculous up tick in quality for the environment and users. It even asked if it could help, saying AI might be a better way to do node management. I thought that was actually really cute.

In any case, if anyone is interested, feel free to ask Q's. I can only go so in depth in a public post like this due to the demands for such technology today. I don't want to give up the goose yet. I just need to know what I need to look into to AVOID another ZeusNet.

​

Thank you. :)