Oh man, they are coming for our tandems too!? My area is still under NEC 2020 by ordinance, which is still more recent than a lot of surrounding counties. It’s still recent enough to need GFCI on dual-pole breakers going to outlets, which was a real pain in terms of availability and cost.
There is a lot of addressable market that can add chargers inexpensively. But I get that doesn’t mean much to anyone in a 70 year old home with a full panel, all electric appliances, and 2 AWG cable buried in the yard.
To try to save some money, make sure you really need the 40 amp charger circuit and make sure your panel really needs the upgrade. I don’t want to imply that you are getting into $6k of work without thinking very hard, but here’s some thoughts and observations, just in case your electrician didn’t suggest all this.
Tandem breakers can be used to move two single-pole circuits into one breaker slot. Doing this with four circuits can free up space for a double. Also check on how much peak load you are actually pulling, then figure on how many reactive starting loads are realistically going to happen at once (like AC coming on). I have the 40 amp charger circuit, a 50 amp hot tub circuit, a dozen servers pulling ~7 amps all day, and two air conditioners (main 40 amp and garage mini-split 15 amp). My peak 1 minute load is like 59 amps while charging the car. All that said, if you have electric baseboard heat in your home that can be a lot tighter of a squeeze.
I have had the EV for two years and I also find that a 240v 15 amp charger circuit, 12 amp max charge would have been perfectly fine for me. That would allow recovery of about 60 miles overnight, or ~90 miles in a typical commute-to-commute time-span of 14 hours. That’s with a big chonky Nissan Ariya, the Bolt will do better.
Do you need a “cloud”, as in a programmatic API to access and provision a pile of compute and storage? Or do you mean a “cloud” as in a bunch of locally hosted productivity tools?
The first case, creating a pool of resources with API access to deploy VMs, can be accomplished with something like CloudStack or OpenStack. This is what you would want if you are just handing off compute and storage resources for other teams to provision and deploy software on.
If you are managing internal IT resources and software, especially critical infrastructure like DNS, NTP, virtual routers/firewalls, identity management platforms, then you probably want a a hypervisor solution like Proxmox, or Nutanix or VMware if you have need of the extra features and have the budget for it.
My base infrastructure and productivity suite looks like this:
I have a dual-boot Windows partition just for games and I keep using it less and less. Valve has done incredible work with Proton. It has been months since I have booted into Windows, and that’s with a weekly games night where we are generally trying something new every week.
Alternative experience here: I have managed 7 GitLab installs for the past 12 years, two of these having ~300 users and ~1000 repositories, and I update monthly on average. I have only had one update break in all that time, and it was because of starting another upgrade step without waiting for all of the background migrations to complete. GitLab support got us sorted out within a couple of hours. A great experience overall.
There is a complexity floor, however. I use GitLab personally because of its CI/CD implementation. Upgrading it requires some care because it’s a large and complex project. You have to read the errata and use the upgrade planner.
For small installations with a few users, and if there is no CI/CD preference, Forejo is probably the better choice for maintenance simplicity.