Incremental improvements to an infrastructure or rather the software ecosystem of a company isn't something many people talk about, indeed. Shiny new tools are interesting and flashy to look at.
For example, I'm very happy with the development of the company I work for over the last 2 or 3 years. I've pushed them from old unmanaged systems to mutable systems managed through chef. There's still some really ugly things, it's mutable, it's not shiny, it's VMs running tomcat running java 8. I like to call it really 2000s and/or vintage. It took time to migrate all systems, to gain trust of stake holders, to educate and convince the ops team. And I guess in some cases, we're running a really ugly mess of a system, but it's a reproducible mess.
But this was a massive improvement and value gain for the company. Suddenly we have an ops team with a lot of leverage and competence to manage an ever growing SaaS setup. And overall, the mindset of most people involved has changed over time towards standards, automation and the value of this. Automation has improved the cost efficiency of some default projects dramatically. We're still old school overall, but we're generating value.
And now the wheel of time has turned some. We've been bought, now there's 8 more development teams, now there's new products being brought in. At this point we're picking up containers at a larger scale because we have to move faster than the config management can handle with the current manpower. So now we're handling some stuff we can using the config management, some stuff with containers.
Overall, moving slow and deliberately in an infrastructure is a very valuable skill. Solve the important problems. Sometimes a trusty, ugly, old java application server isn't your important problem.
For example, I'm very happy with the development of the company I work for over the last 2 or 3 years. I've pushed them from old unmanaged systems to mutable systems managed through chef. There's still some really ugly things, it's mutable, it's not shiny, it's VMs running tomcat running java 8. I like to call it really 2000s and/or vintage. It took time to migrate all systems, to gain trust of stake holders, to educate and convince the ops team. And I guess in some cases, we're running a really ugly mess of a system, but it's a reproducible mess.
But this was a massive improvement and value gain for the company. Suddenly we have an ops team with a lot of leverage and competence to manage an ever growing SaaS setup. And overall, the mindset of most people involved has changed over time towards standards, automation and the value of this. Automation has improved the cost efficiency of some default projects dramatically. We're still old school overall, but we're generating value.
And now the wheel of time has turned some. We've been bought, now there's 8 more development teams, now there's new products being brought in. At this point we're picking up containers at a larger scale because we have to move faster than the config management can handle with the current manpower. So now we're handling some stuff we can using the config management, some stuff with containers.
Overall, moving slow and deliberately in an infrastructure is a very valuable skill. Solve the important problems. Sometimes a trusty, ugly, old java application server isn't your important problem.