Autopilot: Workload Autoscaling at Google

dl.acm.org

29 points·aratno··8 comments

Joyent (SmartOS/Triton) has promoted something like this they called the "Autopilot pattern". However it's more comprehensive than scaling because it also includes other operations like startup and recovery (similar to Kubernetes operators but not for Kubernetes).

http://autopilotpattern.io/ https://www.joyent.com/blog/dbaas-simplicity-no-lock-in

terrible name. Microsoft has had something called Autopilot for a decade now. see: https://www.microsoft.com/en-us/research/publication/autopil...

This leads me to believe that, given it’s roughly in the same space, either this team is not aware of things that a major competitors are doing in the space or they know and just don’t care.

Autopilot is still used today by first party Microsoft teams. Even some azure teams are using AP now.

yes. last I heard it's alive and kicking.

Have you considered the possibility that internally Google's Autopilot was available even before Microsoft's paper?

It's possible but I highly doubt it. We're talking 10+ years (2007) since Microsoft made it public via the research paper.

Autopilot is the name of the vertical scaling system inside Google used for borg. It's been there as long as I have, which is 2012.

I imagine the team that named it didn't need to read a research paper from MSFT to understand the benefits of dynamically scaling machine resources to match load.

I just skimmed the abstract and had an impression that this looks more similar to Borg, rather than Google's autopilot which solves a very particular problem of machine resource utilization within Borg. And yes, Borg has been there more than 15 years.

Reply on news.ycombinator.com