Dynatrace monitors your hosts, containers, and the applications that run on them. The capabilities you choose determine how deeply each workload is monitored and how its consumption is calculated under your Dynatrace Platform Subscription (DPS), which is what your billed costs are based on. Use this page to compare the options and pick the right one for each workload.
The following capabilities are available for host-based and container-based workloads:
For more information about OneAgent installation and setup, see OneAgent setup.
Host and container monitoring don't cover:
Dynatrace offers in-depth monitoring for host environments and the applications that run on them, including cloud instances, on-premises servers, VMs, and IBM Z mainframes.
Each host is monitored in exactly one mode, so you choose one mode per host. Start from what the people who rely on that host need to see: application performance and traces, detailed host and process metrics, or just availability and inventory.
| Feature | Full-Stack Monitoring | Infrastructure Monitoring | Foundation & Discovery |
|---|---|---|---|
Availability and topology | |||
Basic host health indicators |
|
|
|
Host and process topology |
|
|
|
Filesystem and OS services |
|
|
|
Performance and metrics | |||
Process, disk, memory, and network analysis |
|
| |
Custom metric data points included | 900 data points per contributing GiB in each 15-minute interval (about 21 million per month for an 8 GiB host) | 1,500 data points per host in each 15-minute interval (about 4.3 million per month) | |
Extensions |
|
| |
Application performance monitoring | |||
Distributed tracing |
| ||
Automated root-cause analysis |
| ||
Code-level profiling |
| ||
Trace volume included | 3,000 KiB per contributing GiB in each 15-minute interval (about 66 GiB per month for an 8 GiB host) | ||
Kubernetes | |||
Kubernetes Platform Monitoring |
|
Use Full-Stack Monitoring when you need application performance visibility (distributed tracing, root cause analysis, and code profiling) alongside infrastructure observability on the same host. It includes distributed tracing, automated root cause analysis, CPU and memory profiling, Infrastructure Monitoring, and Kubernetes Platform Monitoring.
For more information, see Understand and manage your consumption of Full-Stack Monitoring (DPS).
Use Infrastructure Monitoring when you need deep host and process metrics without application-level tracing or profiling. It includes detailed host and process performance metrics (CPU, memory, disk I/O, network throughput, and OS-level analysis), with support for Extensions.
For more information, see Understand and manage Infrastructure Monitoring consumption (DPS).
Use Foundation & Discovery for lightweight availability and topology on non-critical hosts, when uptime and infrastructure inventory are enough and you don't need deep performance metrics. It includes basic host availability, disk status, OS service health, and topology detection.
For more information, see Understand and manage Foundation & Discovery consumption (DPS).
Use Mainframe Monitoring when you need to monitor IBM z/OS workloads together with the rest of your infrastructure. It includes host-level health and performance metrics and distributed traces for z/OS transactions, correlated with the rest of your monitored infrastructure.
For more information, see Understand and manage Mainframe Monitoring consumption (DPS).
Dynatrace offers in-depth monitoring for containerized environments and the applications that run inside them, including serverless container platforms such as AWS Fargate.
Container capabilities build on each other rather than being alternatives. On Kubernetes, start with Kubernetes Platform Monitoring for visibility into clusters, nodes, and workloads, and add Container Application Monitoring when you also need application performance monitoring. On serverless container platforms, where there's no cluster to monitor, you can use Container Application Monitoring on its own.
| Feature | Kubernetes Platform Monitoring | Container Application Monitoring adds |
|---|---|---|
Platform visibility | ||
Kubernetes metrics and events (CPU, memory, network) |
| |
Health views and explorative analysis |
| |
Resources and topology |
| |
Application performance monitoring | ||
Distributed tracing |
| |
Service detection and dependency maps |
| |
Automated root-cause analysis |
| |
Trace volume included | 6,000 KiB per container in each 15-minute interval (about 16 GiB per month) | |
Custom metric data points included | 300 data points per container in each 15-minute interval (about 864,000 per month) |
Use Kubernetes Platform Monitoring when platform and DevOps teams need visibility into clusters, nodes, and workloads. It includes resource metrics, events, topology, and alerts across clusters, nodes, namespaces, workloads, pods, and containers.
For more information, see Understand and manage Kubernetes Platform Monitoring consumption (DPS).
Add Container Application Monitoring when you need application performance monitoring for cloud-native workloads, charged at a flat rate per container-hour regardless of container size. It includes distributed tracing, service detection, and automated root cause analysis.
Container Application Monitoring can't be combined with host monitoring of the host its containers run on. For example, if a host is monitored with Full-Stack Monitoring, don't enable Container Application Monitoring for its containers: the host is already licensed through Full-Stack Monitoring, so its containers would be licensed twice.
For more information, see Understand and manage Container Application Monitoring consumption (DPS).
You can extend any host-based or container-based monitoring mode with these capabilities, which are licensed separately.