Operations | Monitoring | ITSM | DevOps | Cloud

Do more with less. That’s the mandate we’re all hearing. AI has fundamentally changed how we work. Modern AI workloads generate 10-100x more queries than humans ever could, pushing legacy architectures past performance limits. And the audacity of it all? Legacy logging vendors continue to raise costs without delivering meaningful innovation. IT and security teams are still forced to choose between speed and retention. Investigations are still slow. Data onboarding is still painful.

International travel comes with amazing sights, cultural experiences, and local delicacies. However, most travelers know that it comes with differing economies that impact a money’s value and various currencies. When people need cash, they have to translate the money in their wallets to the local currency, which means different coins and bills. Depending on the exchange rate, the currency’s value can change as the person moves from one country to another.

Your on-call engineer gets paged. A critical service is down. Error rates are spiking. They open Sentry for errors. Flip to Grafana for metrics. Pivot to Kibana to search logs. Then jump to Lumigo, but that only covers the Lambda functions, not the Node.js backend throwing the actual errors. Three tabs become five. Five become eight. Half the incident is gone and your team is still piecing together what happened instead of fixing it. Sound familiar?

As tools such as AI agents become more integrated with the instrumentation, governance, and centralization of product analytics data, product managers (PMs) still own the meaning of those events and the connected outcomes. Knowing when to trust the data, forming strong hypotheses, and being able to act on the insights requires an expert in the loop.

In this series, we’ve explored Karpenter’s architecture, the key metrics that reflect its health and performance, and the vendor-agnostic tools for collecting and analyzing its telemetry data. In this final post, we’ll show you how Datadog helps you monitor and alert on Karpenter alongside your Kubernetes cluster and the infrastructure that runs it.

In the first two parts of this series, we explored how Karpenter’s architecture enables just-in-time provisioning and active node consolidation, and we identified the key Karpenter metrics you should track to keep your cluster performant and cost-efficient. In this post, we’ll look at vendor-agnostic tools you can use to capture these signals.

In Part 1 of this series, we explored how Karpenter’s architecture enables just-in-time provisioning and active node consolidation. Because Karpenter is constantly making infrastructure decisions based on real-time scheduling pressure, its metrics can give you early warning of provisioning slowdowns, cloud API throttling, and misconfigurations that prevent it from scaling the way you expect.

Karpenter is a fast, flexible Kubernetes autoscaler designed to improve cluster performance and cost efficiency. When the cluster doesn’t have capacity to schedule a pod, Karpenter requests additional compute from the cloud provider, specifying a right-sized instance that matches the preferences you’ve set (for example, instance family).

There is a quiet moment that sometimes happens right before a meeting begins. The slides are ready. Dashboards are open. The numbers look neat on the screen. But the revenue doesn’t match last week’s number. A trend line suddenly looks different. Someone says, "That’s strange." And the conversation shifts. Instead of talking about strategy or growth, the room starts trying to figure out what happened to the data. Moments like this rarely happen because someone made a mistake.