Grafana Labs Adds Kubernetes Support to Observability Platform
Grafana Labs this week at the Kubecon + CloudNativeCon conference demonstrated for the first time an ability to monitor and observe Kubernetes clusters via its cloud platform.
In addition, the company revealed Grafana Agent software has been optimized to support OpenTelemetry Collector, an open source tool for receiving, processing and exporting telemetry data. OpenTelemetry is open source agent software being advanced under the auspices of the Cloud Native Computing Foundation (CNCF).
Grafana Labs CTO Tom Wilkie said adding Grafana agent to Kubernetes clusters using a Helm installation tool it provides to pull Kubernetes metrics, logs and events into pre-built Granfana dashboards for visually tracking alerts. The goal is to make it easier to use the Grafana Labs observability platform to surface issues before they impact applications, he added.
In general, most organizations are pursuing a programmatic approach to observability, noted Wilkie. While there is a clear need to query data whenever an issue is discovered, IT teams are still monitoring pre-defined metrics to understand how applications are performing. The aim is to strike a balance between traditional monitoring and more advanced observability capabilities, he noted. The key is unifying the data as it is collected from multiple computing environments, said Wilkie.
At the same time, DevOps teams as part of that are also being tasked with implementing FinOps best practices to programmatically consume IT infrastructure more cost-effectively. As such, there is a pressing need for tools to monitor costs in a way that is integrated into the flow within an existing IT management workflow.
The challenge is providing access to a set of tools to achieve those goals at a time when most organizations are not increasing the overall size of their IT organizations, noted Wilkie.
As the number of Kubernetes clusters deployed in production environments continues to increase steadily, monitoring and observability of cloud-native applications environments is becoming more challenging. IT teams are under considerable pressure to ensure the resiliency of cloud-native application environments while reducing costs.
The issue at hand is that Kubernetes is simultaneously the most powerful and complex IT platform to find its way into enterprise IT environments in recent memory. The level of expertise required to optimize Kubernetes environments is still hard to find and retain. The only way to narrow that gap is to provide IT teams with tools that are more accessible, noted Wilkie.
Arguably, there would be greater adoption of Kubernetes if the platform was simpler to manage. Kubernetes itself was designed by software engineers for other software engineers. Despite the rise of DevOps to programmatically manage IT environments, most IT teams are still made up of a mix of engineers and administrators who tend not to have as much programming expertise.
IT teams will inevitably find themselves managing Kubernetes environments along legacy IT platforms. In fact, for the foreseeable future, IT teams will be managing a mix of cloud-native and monolithic applications. The challenge is finding frameworks that make it possible to manage those environments without necessarily having to acquire and maintain a separate monitoring and observability platform for each one.
