We recently explored how operations teams need to evolve from monitoring to observability to avoid being overwhelmed by alerts. But observability is just as important for developers. It gives them better insight into how applications behave in the production environment. The question is: how do you give your developers this insight in the best possible way?
Think of development and operations as the makers and drivers of cars. Development assembles the car — the IT application, in our case — while operations is the driver. The driver receives instant feedback: if they press the accelerator harder, they see the needle on the speedometer rise. But the manufacturer can’t always see how the car responds to inputs in the real world.
Developers need feedback too
Operations immediately sees the effect of a functionality change or a setting adjustment. But development needs this insight too in order to make the best adjustments and improvements over time. And it’s not enough for development to simply see the raw metrics — observability is just as important. Developers must be able to investigate what’s going wrong at every step of the development process.
The problem is that developers often don’t get enough of this insight. When they receive an issue report, they want to resolve it quickly. But without direct visibility into the production environment, the only thing they can do is request logs from operations or ask operations for custom dashboards. The problem is that operations often doesn’t have the time to do that or doesn’t even always know exactly what those dashboards should display. After all, operations didn’t write the applications and views them very differently than development.
In other words, developers still too often encounter a technical barrier to getting their job done, preventing them from quickly arriving at the correct solution. In practice, development needs to see every metric and every log in real time so that they can immediately understand the impact of any proposed actions.
So what can be done?
At BRYXX, we’re tackling this by building a monitoring and logging plane into our internal development platform. This layer provides a home for all conceivable data about application performance, standardized or otherwise, using open-source observability frameworks such as OpenTelemetry.
Additionally, the monitoring and logging plane integrates modern observability tools such as Grafana, Splunk, and Dynatrace, which developers can use to easily develop dashboards that display exactly what they need to know about application performance. The platform provides all the necessary building blocks to do this. In addition to security by default, the platform offers monitoring and observability by default, managed by a separate platform team, ideally made up of both operations and development people.
More efficient collaboration
A monitoring and logging plane eliminates potential friction between development and operations. Instead of operations drowning in alerts while development receives too few (or the wrong) notifications, both teams have quick and easy access to the information they need, thanks to the observability tools of the monitoring and logging plane. This allows everyone to do their jobs better and provide the customer with the best possible user experience.