Have been looking for minio alternative for long already. Found versitygw lately and would like to share the joy. It feels very promising. Fits to many small or lab use cases.
It does not actually solve the trickiness of managing large storage but relies on the backend (that is usually fs like zfs in small setups).
However, seems to be quite new project plus the risk, that the owning company takes it to bad direction, is there too.
Using none currently, the following seems to be enough: containers for apps and KVM+vagrant on a laptop for experimenting.
However, for VM orchestration. OpenStack has been my choice for long time, but about to check KubeVirt and (old and underappreciated?) Apache CloudStack.
Is it so that the fragmentation (the long list of distributions), or variety of configuration options, is a bad quality? This seems to be used as a counter argument to Kubernetes.
Probably, it could be end result of unconsistent design or bad technical choises. However, most likely it just means that there are multiple organizations and interest groups pushing changes and ideas to the project. This should be seen as a good thing. The downside is that there is no single source of best practices and this is confusing to newcomers. You just need to pick one distribution and trust the choises, or develop the competence and understanding.
And we could imagine that the userbase or the number of developers in single distribution (take OpenShift or Rancher) could be bigger that in Nomad itself.
Having said that, I still would like to see more stable Kubernetes landscape, and that has to happen eventually. The light distributions k3s and k0s are pushing things to nice direction.
OpenStack had similar, or maybe even worse, fragmentation and complexity issue when the hype was high. There were probably technically better alternatives (Eucalyptus?), but people (and companies) gathered around OpenStack and it won the round. However, comparing OpenStack to Kubernetes feels bad, as Kubernetes is technically far superior.
i3 is great and it's almost a must for my productivity nowadays.
However, I would highlight, that it's ability to work in multi-monitor is huge selling point.
You would like to show something from an another screen, plug the cable, press suitable hotkey, and the workspace, that you were just working on, jumps to the new display. No need to move windows.
You go back to your desk and plug in the laptop. The workspaces will automatically jump to the monitors, per your config.
Next, you run to a meeting, the very same workspaces collapse back to laptop's monitor automatically.
There was no need to touch windows in any of these scenarios, And no more "I forgot the browser windows to the attached display" that has happened with some other systems.
Very welcome suprise! I was excited about pNFS four years ago, but couldn't find any open source system to play with. Finally it is time for some mirrored pNFS testing.
One important thing, that is not pointed out yet: FreeNAS has easy upgrade path for major OS versions. That is something that I miss when running plain FreeBSD.
Plus, you can always switch between plain and FreeNAS installation.
I was interested in the thin client concept few years ago and tried virtualgl. It is really nice and impressive. There are few different configurations available. I aimed for low latency and 30 fps gaming was easy to achieve over 1gb link (it needs wide bandwith as expected).
I have been looking for a Document Management System, and I would like to understand how does this compare to other open alternatives like dSpace.
I would favor older and more established project, but maybe I'm focusing too much in Document Management and haven't thought about the collaboration aspect. PaperNIC might be better there.
- packer, that is a really nice and simple image building tool. It saved lot's of headache compared to tools like oz, because packer fails fast and saves debugging time.
- terraform, that is comparable to OpenStack Heat or AWS CloudFormations. Each have their own issues. But terraform being platform agnostic is an advantage for me.
- vagrant is not modern anymore, but it's ubiquitous.
Yes, because it is a great platform to learn system administration. First, it's so simple that eg. it takes only few lines to automate installation or create a customised internal repository. Second, the documentation.
edit. And oh yes, the memory usage of the default installation is so low, that you can virtualize dozens of them on your laptop. This helps when you want to locally reproduce the stack you run in cloud.
It does not actually solve the trickiness of managing large storage but relies on the backend (that is usually fs like zfs in small setups).
However, seems to be quite new project plus the risk, that the owning company takes it to bad direction, is there too.
https://github.com/versity/versitygw/