While it is _technically_ possible to build heavily decentralized, managed hardware deployments, no one has figured out yet how to charge for them except in the case of very large customers. Until this happens, there will be no swing back to anything decentralized.
Although I do understand the allure: just about anyone today could very economically purchase 5-10 servers with 10-18 cores each (and those will be _real_ cores, not hyperthreads). There's nothing impossible about automating software management on such a thing, even to the extent that you'd get in cloud (VMs, containers, distributed storage, automated updates, VM migration, network partitioning, etc). I believe Microsoft will lease you a fully set up shipping container with Azure in it, all that needs to be done is scale this down.
But again, how does one extract billions of dollars in profits from something like that?
I think the point would be that the central server `could` gather aggregated data / important business metrics that it could use to provide further value. I think the author also mentions something like this.
Ex: provide analytics back to the end user that could be a paid service, or use that information to provide a second set of b2b offerings to other partners. etc.
Now compare that to the current model of nickel and diming for every single thing and charging a 90% margin on traffic, and you’ll see why this is not attractive to the current players.
Well, I think intrinsic ineffeciency of storing all that data inside your central and then running huge spark jobs to crunch through them creates huge costs in the first place that probably needs to be offset by higher margins being charged from the end consumer. If you can just monitor the users engagement with your offering and charge on that without having to actually go through countless logs to do that, might be what the author is leaning to. But yeah, easier said than done. Knowing which metrics to capture back centrally itself could be a challenging thing to do, if the data is decentralized ....
Although I do understand the allure: just about anyone today could very economically purchase 5-10 servers with 10-18 cores each (and those will be _real_ cores, not hyperthreads). There's nothing impossible about automating software management on such a thing, even to the extent that you'd get in cloud (VMs, containers, distributed storage, automated updates, VM migration, network partitioning, etc). I believe Microsoft will lease you a fully set up shipping container with Azure in it, all that needs to be done is scale this down.
But again, how does one extract billions of dollars in profits from something like that?