Do you document your code? Imagine that you need to get back to your code in 6 month after you wrote it, there is always a big possibility that you will have to spend some time to find out how this code works. Or if someone else wrote some code, which is already in production and your task is to fix a bug in it and there is no documentation and no one actually knows what this code does.
There are more benefits of implementing continuous documentation for the code:
easy to onboard new team members,
easy to share knowledge,
if this code is open source - easy to start contributing,
easy to see purpose and motivation of each piece of code,
easy to keep versioning for each new release of the code.It this talk I will show the difference between documentation types and will show a demo in the end of the talk.
If you nowadays work with and on the cloud or simply *-as-Code, chances are you came into proximity of ‘Yet Another Markup Language’ (YAML). Deceptively simple in appearance, YAML provides you with a plethora of ways to shoot yourself in the foot. This talk discusses a number of common pitfalls. In a nutshell: Why YAML is not as easy as you may think.
The general consensus is that is too hard or too expensive to run a small business on entirely free software. After running my company for 5 years on entirely free software I think this is not the case. We have instant communication, shared calendars, groupware, single sign on, and a single administration interface! Things have worked out really well and in many ways are better integrated than trying to combine multiple different cloud based services.
In this talk I will walk through how our infrastructure works, what the pitfalls were, and what the benefits are. At the end I'll give a short demo on a couple of virtual machines to show off the user and admin experience.
OpenStack Swift is the massively scalable object storage of OpenStack. It can store billions of objects in a single cluster without any disturbances. But what happens if all of your objects are "small objects"?
Because OpenStack Swift stores objects as files on XFS, disks will end-up storing millions of small files, creating large performance and stability issues. This is all related to the constraints of POSIX filesystems.
During this presentation, we will give you an overview of these issues, and explain their root causes. You will then discover what we tried to overcome them, as well as the chosen solution. Eventually, we will present you results of our production deployment.