Unveiling the Magic Behind On-premises Infrastructure – DevOps Series, Part 4
Now that you are familiar with how Developers work in their local environments (see previous blog), it is time to explore their peer Operations teams. Once developers feel comfortable with code running in their laptops, Ops needs to start thinking about how it will go into production. That production environment might be on-premises (private) or in the Cloud (public), depending on many different factors – like cost, data confidentiality, or even local country regulations.
Today we will discuss On-premises deployments, and leave public Cloud ones for my next blog entry.
On-prem might be, for example, a QA (Quality Assurance) environment where code is tested (unit testing). Only on successful completion would it be moved to production (whether on-prem as well, or in the Cloud). Or it might be the other way around. There are multiple possible scenarios and it would be great if they were all kind of similar, right? That way we could leverage the same skills to work on one or the other.
On-prem environments are very common and they provide total flexibility as they are run by in-company personnel. Of course, they come at a cost in terms of purchasing and operating the solution, but the knowledge obtained in the process will allow you to understand and optimize many aspects that will be of ultimate benefit for you, your software, and your company.
Build and Configure Your Own Mini DataCenter
Please follow my new Learning Lab on ‘Operations on-premises’, where I cover how you can build and configure your own Mini DataCenter from scratch, and provide it with external connectivity in a home environment. You will learn how to deploy a modern microservices-based example application into it, including publicly reachable user interfaces. And finally, as we know it will be a very successful application, I will show you how to implement a queuing system to alleviate the pressure it will suffer from multiple concurrent requests.
Don’t worry if you have no hardware at all to build your own datacenter. There are several alternatives that might be of interest to you as you start learning. My personal favorite is Cisco DevNet Kubernetes Sandbox. It provides you with a comprehensive environment, including 1 master and 3 worker nodes. It is based on Virtual Machines and comes pre-configured, so you do not need to fight to get it working. It is accessible via SSH from your own terminal application, and you can make a reservation for it according to your needs (hours or even days). And it is reliable. You will not be evicted from your environment without notice, and have to start all over again.
Now you have everything you need. Please go through my new Learning Lab and have fun! Any questions or comments please let me know in the comments section below, or contact me via Twitter or LinkedIn.