How are you approaching Observability in the Cloud across both non-prod and prod environments? Are you paying for a tool (e.g. Dynatrace) across all environments or only production? If only production, are your devops teams able to use the tool effectively in production after not having used it in development? If across all environments, how is the cost not prohibitive? Or, did you manage to accept it as a cost of doing business?

Cloud observability is indeed a crucial topic, especially when it comes to balancing costs and ensuring effective monitoring across all environments. At Ternium, we prioritize using Dynatrace in production environments and critical applications. This approach helps us gain a clear view of the application's architecture and associated services, ensuring we have comprehensive visibility into our infrastructure, user experience monitoring, resource utilization trends, and, most importantly, faster analysis of critical incidents. This is achieved through troubleshooting problems at the code level and identifying root causes via the integrated artificial intelligence.

We understand that cost can be a significant factor, which is why we decided to begin using the tool primarily in the most critical applications. Since the tool is mainly utilized in production environments, our development team and functional leads were trained through workshops conducted by Dynatrace consultants in collaboration with our technology team, allowing them to fully explore the tool’s key functionalities.

I’m available to discuss this topic further or answer any other questions that may arise.

I’ve found that a 'tool-first' approach can sometimes lead to unnecessary costs. Instead, consider a 'problem-first' mindset: pinpoint specific observability needs in both non-prod and prod environments before selecting tools. If using tools like Dynatrace across all environments, ensure they’re utilized effectively by your DevOps teams. For those managing costs, evaluate if a tool’s expense is justified by its value and how it contributes to overall efficiency. Balancing cost with effectiveness is key to a sustainable observability strategy.

We are currently working on that and I suggest you the a balanced approach across environments by implementing a balanced observability strategy that covers both non-production and production environments, but potentially with different levels of depth or tools.

We are working with MS DevOps, more exacly a DevSecOps with FinOps stack.