So, before I get started on blog posts with proper tech, just a quick introduction and an overview of what I’ll be doing here.
Yeah, I know, insert standard blog welcome here. But that’s not really what this blog is all about.
I love tech. I also love creating stuff, notably coding although with the lockdown and overall lack of any urge to do anything other than breathe and obsess about the latest piece of bad news, I’ve found it hard to do anything.
Hence, in some part, this blog.
As I said, possibly badly, in the introduction, my goal of this blog is to try and share some of the cool new tech in a way that makes sense to us developers; we are at an interesting point in the industry where the momentum of creation of new tech is far outstripping actually using it – this norm was coming into ascendency when I moved out of the career of producing software for a living and into vendor space. The advent and succession of Open Source and proliferated a massive amount of very specialised technologies, a lot of which is done purely to solve pragmatic problems.
This deluge of tech is a nightmare for the day-to-day dev. We all like our comfort zones but nowadays the stuff seems to change on a daily basis. I talk about some of the tech I was proud to have mastered and most people glaze over and look at me like I’ve just dropped a bus pass – STRUTS, JSP, JSF, ECMAScript, OpenVMS, the list is endless.
But the problems the new tech solves are basically the same, and that’s where it is fun to just dig down and find them.
I’ll give a nice example of this, pre-empting an blog post or two – we now have OpenShift Virtualisation, a technology that runs VM images directly on the Kernel of a Worker Node using a wrapper/launcher container; it effectively is running a VM as (or rather in) a container. Now that sounds nuts until you look at some of the topologies people now have for their apps; I dealt with one customer who had a transaction system comprising a backend database, which for licensing reasons couldn’t be containerised, and a frontend system written using .NET Core, which could.
They had effectively two orchestration systems for their end-to-end product, a Container orchestration system for their frontend/Ux and a Virt orchestration system for their old school database backend.
And, due to a change in the law, they had to change the schema of the database and the text/entry points on their system; it was a popular system and they couldn’t afford a huge amount of downtime, but because of the physical separation of the components they couldn’t do a single point rollout of the changed frontend and the backend; they literally had to take their 24 hour available system offline for a good amount of time to respin the VMs and release the frontend simultaneously. And it involved two separate and completely disparate control systems.
So, moving the VMs into OpenShift they could orchestrate a release of both the DB and the frontend simultaneously with minimal downtime (unlike the standard Containers you can’t really do a rolling deployment of the VM but the action of rollout was much simpler and single-point controlled).
Anyway, this blog will have a number of fun tech articles that hopefully explain the new stuff in a way that I can understand it; if you have any questions please add a comment…..
Devepiphany 