News for hackers. Stuff that matters.
Re:The end-to-end refresh of our server hardware fleet by raz32dust on Friday March 10, 2017 @08:17PM (#13843552)
I feel you. But I think the lock-in problem can be solved if we can have some standardization of cloud services such that you can always move to another provider. That has to start with some big company developing an abstraction layer and open sourcing it, and then we can go from there. I think Netflix has a switch from Amazon to GCP; I hope be they'll standardize it and open source it.
Re:The end-to-end refresh of our server hardware fleet by cowardlydragon on Friday March 10, 2017 @09:10PM (#13843756)
Chef, puppet, etc, and other Apache projects all offer this already.

It's like leveraging Oracle specific database features. Your a fool to do so.

Re:The end-to-end refresh of our server hardware fleet by jacques_chester on Friday March 10, 2017 @10:02PM (#13843956)
> That has to start with some big company developing an abstraction layer and open sourcing it, and then we can go from there.

I work for a company (Pivotal) that's had such a product -- Pivotal Cloud Foundry -- for several years. It creates an abstraction layer for apps or container images, your choice.

Deploy with BOSH to raw metal, OpenStack, vSphere, AWS, Azure or GCP. BOSH creates an abstraction layer over the IaaS.

We're also the main driving force behind Spring, Spring Boot and Spring Cloud Services; the latter is in part a generalisation and integration of Netflix OSS.

We cooperate a lot with Google and Microsoft. For example: https://cloud.google.com/solutions/cloud-foundry-on-gcp