Hacker News Reader

Comment by haalcion3 | Hacker News Reader

And, as probably everyone knows, Google runs everything in containers and has been using containers for a decade:

http://www.nextplatform.com/2016/03/22/decade-container-cont...

Docker may be flawed, but containers aren't. If you need some enterprise leader to tell you this instead, here are some Gartner posts showing this is the way:

VMs may be well established and "magic quadrant", but they are also on decline, and containers make better use of hardware/resources:

http://www.informationweek.com/cloud/infrastructure-as-a-ser...

It's important what you use to roll out containers and requires forethought:

https://www.gartner.com/doc/3267118/containers-change-data-c...

View on HN

What really surprises me about Google is why they don't open source some of these great core technologies (MapReduce, Containers etc.) instead of publishing theory as academic papers. On the one hand, it may be a great way of promoting the creating of these tools from the ground up, inspired by the theory alone. On the other hand, Google's invaluable experience with using these technologies probably means their versions are more stable, faster, resilient etc.

Kubernetes:

http://kubernetes.io/

Google Container Engine:

https://cloud.google.com/container-engine/

Open-source MapReduce implementation:

https://gigaom.com/2015/02/18/google-open-sources-a-mapreduc...

https://github.com/google/mr4c

Dataflow [aka Flume]:

https://github.com/GoogleCloudPlatform/DataflowJavaSDK

http://research.google.com/pubs/pub35650.html

Kubernetes just got a pretty sweet update as well. The improvements to the dashboard alone were worth revisiting.

I think they've said in talks/presentations that the challenge they have with some stuff (like Borg) is that they can't really extract individual components. It's all too tightly coupled. It wouldn't be fair to ask them to open source their whole stack. The fact that they took the lessons learned and created Kubernetes or published papers on their technologies is more than enough.

Why enable their competitors to such an extent? What benefit is in it for them, when they seem to be at the absolutely forefront of this work? They have to let their researchers publish academic papers or they won't get the best researchers, but when they also have the most experience with ops and such a lead in implementation, the theory only gets competitors so far.

It's entirely possible that they consider those core technologies part of their "magic sauce". If their development and production environments are so far ahead of everyone, their development cycle will be much faster.

Reply on news.ycombinator.com