Re:What we've learned from building Ghost after 5 years and $3M
by hawkice on Wednesday May 16, 2018 @11:28PM
(#17088483)
Woah! That information about Docker is bananas. I've worked on projects with more than 200 VMs. Talk about things that don't show up on HN: scaling issues with popular technology where you never find a solution.
[ Parent
| on HN ]
Re:What we've learned from building Ghost after 5 years and $3M
by Groxx on Thursday May 17, 2018 @02:16AM
(#17089177)
Until fairly recently with docker enterprise, from what I've been able to gather Docker's official stance on Enterprise™-use has been "don't use docker, use runc, and build / orchestrate / etc it with kubernetes or something" which basically means "rewrite whatever you intend to use". (afaik this all changed with https://blog.docker.com/2018/04/announcing-docker-enterprise... )
Which is fine. "can execute binaries in isolation" and "can schedule rollouts and efficient disk/networking structures across many thousands of instances" are vastly different problems. Even "can deploy to 100s" and "can deploy to 1000s" are quite different, without something targeting 1k+ I wouldn't expect it to handle beyond a couple hundred without compounding issues.