I understand for a production environment with x amount of devices, you may need something robust to capture a lot of netflow and syslog data. However, I'm trying to spec out a home lab for the CCIE Enterprise and I'm probably going to have to spend $1000 on a used server from ebay to run this monstrosity. It's not that I couldn't afford that, and I understand I will have to spend some money to get the CCIE, but it seems like such a waste.

My understanding of DNAC is that it runs everything in containers. I'm thinking these containers are being manually allocated a certain amount of CPU / memory that makes the requirements for DNAC so insane. Has anyone looked into modifying these settings at all? I'll admit I don't know anything about docker, but this is something I think should be investigated to see if we can get this running with more reasonable requirements.