 In this video, I'll demonstrate a few ways you can use Pixie to monitor your infrastructure health. Pixie is an open source Kubernetes observability tool for developers. Pixie uses Linux EVPF technology to automatically capture telemetry data such as network and resource metrics, full body requests, and application profiles without the need for any manual instrumentation. Installing Pixie takes less than five minutes using one of the install guides or videos linked below. Let's use Pixie to see high-level resource usage by node. I have the live UI open on my screen and my cluster is selected in the cluster drop-down menu at the top. I'm going to select the script drop-down menu and start typing nodes. Then I'll click on the script. The PX node script lists all of the nodes in the cluster and graphs each node's CPU usage, network traffic stats, and memory consumption. It also lists all of the pods running on these nodes during the time window. You can increase the time window or rerun the script up in the top right corner here. Pixie's UI makes it easy to quickly navigate between Kubernetes resources. Clicking on any pod, node, service, or namespace name in the UI will open a script showing a high-level overview for that entity. Let's click on the name of a node up in the top left nodes table here. Clicking on that node name will follow a deep link to the PX node script with the node argument filled out. Pixel scripts take arguments. These arguments are listed in the top next to the script name. This script has two arguments. The UI denotes required arguments with an asterisk after the argument name. So here you can see that the node argument is required for the script. If you had navigated to the script directly using the script drop-down menu, you would have needed to provide a node name before seeing results. So this node script shows a similar set of information as the node script. It shows CPU usage, memory usage, and network traffic, but for just the selected node. Now, right now, resource usage is shown for the entire node, but to see information broken down by the pod, click the drop-down arrow next to the group by argument and select pod. The graphs will update to show the resource usage by pod instead of aggregated for the entire node. Let's use Pixi to see high-level resource usage by pod. The PX node script contains the list of the pods running on the node. Let's click on the name of a pod in the top left pods table. We'll do the cube DNS pod. Clicking on that pod name will follow a deep link to the PX pod script with the pod argument already filled out for us. Now, this script shows an overview of the specified pod, including high-level HTTP application metrics and resource usage. It also lists containers on the pod, live processes, and more. If you scroll all the way to the bottom, you can see a CPU flame graph for this pod. This flame graph can be used to identify performance issues in the application code. The wider the bar, the more time spent in a specific function. To learn more about Pixi's continuous application profiling, see the tutorial linked below.