Google Kubernetes Engine: GKE Observability and Application Performance Management
The Google Kubernetes Engine (GKE) is a hosted Kubernetes cluster (IAAS – Infrastructure As A Service) provided by Google. Similar to other Kubernetes (K8s) engines (where Instana has OpenShift Observability, Azure AKS Observability and IBM BlueMix Observability), Instana’s comprehensive GKE Observability includes the ability to discover and monitor the entire Kubernetes system, all orchestrated containers, infrastructure and any services are running inside GKE.
To help optimize GKE application performance and automate performance management workflows, Instana’s automatic GKE Observability &p; Monitoring goes beyond simple metrics to provide comprehensive Cloud, Kubernetes, Infrastructure, Application and Service monitoring capabilities:
- Discovery of Kubernetes nodes and deployed services
- Automatic code instrumentation and tracing for Java, Node, .NET and 7 other languages
- Automatic mapping of applications, containers and Kubernetes clusters
- Tracing of all end-to-end requests across all systems and services
- Application, Service, Infrastructure, Kubernetes and Cloud health monitoring
Comprehensive observability of Google’s Kubernetes Engine infrastructure requires performance visibility into the virtual hosts, running pods, containers and orchestration, and any applications and services deployed on the cluster.
Instana is the quickest and easiest way to monitor the Google Kubernetes Engine across the stack and provide comprehensive application insights. The Instana agent automatically discovers all Kubernetes instances, deploys the necessary monitoring sensors and begins tracing applications and requests. Instana also automatically determines the health of the GKE cluster.
Monitoring GKE Performance
The Instana Agent automatically identifies all running Kubernetes nodes and automatically deploys a fully configured Kubernetes Monitoring sensor. Instana’s curated knowledge base already knows what performance metrics are relevant to collect for health monitoring of any GKE Service cluster any and all containers and infrastructure running within the cluster, as well as the applications / services operating within GKE. Since Instana automatically collects all relevant information, monitoring a hosted Kubernetes cluster couldn’t be easier.
Observe GKE within context of Applications and Services
Instana automatically maps, monitors and traces all services and applications that pass through the GKE cluster, detecting all performance / availability events and issues, while also identifying any service incidents, tying all monitoring, tracing, profiling and data analysis together for correlated analysis.
Google Kubernetes Engine Configuration Visibility
In addition to performance and health data, Instana’s Kubernetes Monitoring sensor also collects configuration data for Kubernetes, pods and containers, allowing Instana to analyze and correlate configuration data and changes with application and service performance information.
All Kubernetes performance and configuration information is summarized in a single Monitoring Dashboard, showing all relevant information in a single place for easy problem-solving and performance optimization.
Google Kubernetes Engine performance monitoring centers around service metrics and their interactions with other services or data stores. Instana automatically identifies and collects the relevant service metrics.
Google Kubernetes Engine Observability and Monitoring Data
Instana’s GKE Observability includes four types of data; Cluster Data, Deployment Information, Pod Measurements, Node Measurements:
| Cluster Data | Node Measurements |
| KPIs | KPIs |
| Node Count | Pod Allocation |
| Pod Allocation | Pod Capacity |
| CPU Request Allocation | CPU Request Allocation |
| CPU Limit Allocation | CPU Limit Allocation |
| Memory Request Allocation | Memory Request Allocation |
| Memory Limit Allocation | Memory Limit Allocation |
| CPU Resources | CPU Resources |
| CPU Requests | CPU Requests |
| CPU Limits | CPU Limits |
| CPU Capacity | CPU Capacity |
| Memory Resources | Memory Resources |
| Memory Requests | Memory Requests |
| Memory Limits | Memory Limits |
| Memory Capacity | Memory Capacity |
| Pods (running / allocated / pending) | Conditions |
| Pods Capacity | Labels |
| Replicas (available / desired) | Pod list |
| Node list with KPIs | |
| Deployment list with KPIs | |
| Component Status | |
| Pod Measurements | Deployment Information |
| KPIs | Conditions |
| Phase | Labels |
| Restarts | CPU Resources |
| CPU Requests | CPU Requests |
| CPU Limits | CPU Limits |
| Memory Requests | Memory Resources |
| Memory Limits | Memory Requests |
| Conditions | Memory Limits |
| Labels | Pods (available / desired) |
| Container List (state / restarts) | Pods (pending / ready / unscheduled / unready) |
| Pending Phase Duration |
Further information on the different sensor information is available in the
Instana Google Kubernetes Engine Management Documentation.
Instana Agent Installation: Getting Started
Ready to start? You’ll need an Instana Trial or Account first. Already got one? The best place to begin is Instana’s Getting Started Guide.
