Container Application Monitoring is a DPS capability that provides complete application performance monitoring for containerized workloads, including distributed tracing.
Built for teams running cloud-native container workloads, Container Application Monitoring is licensed with a flat container-hour model that makes costs easier to estimate and scales with modern workloads.
It is specifically built to support:
This page explains how Container Application Monitoring consumption is calculated, and how you can manage your Container Application Monitoring spend in Dynatrace.
The unit of measurement for Container Application Monitoring consumption is container-hours, using the Container Application Monitoring rate card item.
In the context of Container Application Monitoring, a container is any actively monitored container running in a supported environment.
The unit of measure for Container Application Monitoring. One container-hour represents one hour of monitoring for a single container.
With Container Application Monitoring, you get automated root-cause analysis,
Services monitoring, Distributed Tracing, and metrics. It includes:
6,000 KiB volume of traces ingested per 15-minute interval, with a default 10-day retention.
300 metric data points per 15-minute interval, with a default 15-month retention. This covers billable metrics that can be attributed to a single opted-in container, including span-derived service metrics (keys starting with dt.service.*).
Each container generates 0.25 container-hours for each 15-minute interval in which Dynatrace receives spans or metrics from it.
Container-hour consumption is calculated in 15-minute intervals. One 15-minute interval is equivalent to 0.25 container-hours. If a container is monitored for fewer than 15 minutes in a given interval, its consumption is rounded up to 15 minutes. This design accurately captures short-lived services, which are common in cloud-native environments.

As an example, a team runs a data processing workload consisting of three application containers, each with an Istio sidecar.
The workload lifecycle is:
api-service starts when the workload begins and exits after 23 minutes.checkout-service starts when api-service finishes and runs for 15 minutes.web-frontend starts 2 minutes after checkout-service and runs until the workload terminates at minute 54.istio-proxy sidecar.Because container-hours are billed per container per 15-minute interval, 3.50 total container-hours are consumed.
The included volume is calculated per 15-minute interval based on the number of monitored containers:
| 15-minute interval | Containers monitored | Included trace volume | Included metric data points |
|---|---|---|---|
| First | 2 | 2 × 6,000 KiB = 12,000 KiB | 2 × 300 = 600 |
| Second | 6 | 6 × 6,000 KiB = 36,000 KiB | 6 × 300 = 1,800 |
| Third | 4 | 4 × 6,000 KiB = 24,000 KiB | 4 × 300 = 1,200 |
| Fourth | 2 | 2 × 6,000 KiB = 12,000 KiB | 2 × 300 = 600 |
The following examples demonstrate how to calculate your monthly cost for Container Application Monitoring.
Calculations are based on the following list prices (which may differ from your rate card prices):
One month is equivalent to 30 days, 24 hours per day.
200 containers are running continuously.
200 containers × 24 hours × 30 days = 144,000 container-hours per month
144,000 container-hours × $0.007 = $1,008 per month
This example assumes the following included trace/metric volumes:
Included trace volume: 6,000 KiB × 4 × 200 containers × 24 hours × 30 days = 3,295.9 GiB per month
Included metric data points: 300 × 4 × 200 containers × 24 hours × 30 days = 172,800,000 data points per month
This extends the example above with additional trace and metric volume: 1,500 GiB of trace volume and 10 million total metric data points are ingested over the last 30 days beyond the included volume.
200 containers × 24 hours × 30 days = 144,000 container-hours per month × $0.007 = $1,008 per month
1,500 GiB × $0.2 = $300 per month
(10 million excess data points) × ($0.15 per 100,000 data points) = $15 per month
Total: $1,008 + $300 + $15 = Total consumption: $1,323 per month
Dynatrace provides several ways to monitor and analyze your Container Application Monitoring consumption.
License managers can view usage and costs in Account Management.
Go to Account Management > Subscription > Overview > Cost and usage details > Usage summary.
Select Container Application Monitoring > View details.
For more information about the subscription overview UI, see Overview (Dynatrace Platform Subscription).
The Usage – Container Application Monitoring ready-made dashboard shows real-time estimates of your Container Application Monitoring consumption.
Open
Dashboards > Ready-made and search for Usage – Container Application Monitoring.
The dashboard covers:
For more information about ready-made dashboards, see Ready-made dashboards.
Billing usage events (billing_usage_event) are system events emitted by Dynatrace that represent the authoritative record of billable usage.
Use them to build reliable cost dashboards and chargeback reports by host, Kubernetes namespace, cloud account, or cost-center tag.
Each billing usage event for Container Application Monitoring contains:
Query Container Application Monitoring consumption programmatically via the Account Management API for integration with external reporting systems. You can also query the underlying metrics through the Environment API - Metrics API v2.
Yes, with one exception: host monitoring of the same host. Container Application Monitoring can be combined with any other rate-card item, except host-based monitoring of the host its containers run on, such as Full-Stack Monitoring, Infrastructure Monitoring, or Foundation & Discovery.
Don't enable Container Application Monitoring on containers that run on a host monitored by OneAgent in any host monitoring mode. The host is already licensed through its host monitoring capability, and its containers would additionally be licensed in container-hours under Container Application Monitoring, which results in double licensing. For details, see Get started with Kubernetes platform monitoring + Application observability.
Kubernetes Platform Monitoring, Log Analytics, Code Monitoring, Real User and Synthetic Monitoring, and Application Security are also commonly used alongside it.
Container Application Monitoring requires:
No. When your environment migrates from a Dynatrace classic license to DPS, Adaptive Traffic Management and the Full-Stack trace volume calculations are migrated as well. This is an automatic process that takes one week, and it must finish before you can use Container Application Monitoring.
You can start using Container Application Monitoring one week after the initial migration.
Container Application Monitoring treats OneAgent and OpenTelemetry as equals: trace data from both sources counts against the same included trace volume per container-hour. Billable metrics emitted from both sources count against the same included metric data point volume per container-hour.
There is one important difference in how sampling is handled:
We recommend using Cost Allocation with enrichment rules to attribute Container Application Monitoring usage to specific teams, or products.
Set up Cost Allocation and enrich your data with the appropriate Cost Allocation attributes, such as dt.cost.costcenter and dt.cost.product.
Usage details are also available at the application level. For more information about the available Container Application Monitoring attributes, see the Semantic Dictionary.
Each monitored container adds 6,000 KiB of trace data per 15-minute interval to an included pool. Unused volume doesn't roll over to the next interval.
The pool covers trace data from both OneAgent and OpenTelemetry that can be attributed to a single opted-in container. How the pool is protected from overage differs by source:
Container Application Monitoring includes a defined volume of custom metric data points for metrics from your monitored containers. This volume is based on total monitored container-hours and is calculated per 15-minute interval—300 metric data points are included per 15-minute interval. The included volume covers only applications that run in containers, and unused data points don't roll over to the next interval.
dt.service.*) also count against it when the span data comes from a Container Application Monitoring-monitored container.To attribute metrics to containers, enrich your metrics with the context that identifies the container they come from. Metrics without this context aren't correlated with the container, and are billed as Metrics - Ingest & Process from the first data point. Additionally, metric data points that exceed the included volume are charged as Metrics - Ingest & Process.
It is not. A container-hour is independent of the container's memory size, CPU requests, or resource limits. Each monitored container contributes one container-hour per hour, regardless of how large or small it is.
Usage is metered per 15-minute billing window. If a container is licensed under Full-Stack Monitoring and under Container Application Monitoring within the same window, it's charged for both in that window.
To count distinct billable units, Dynatrace generates a stable container identifier (dt.container.uid) during ingest.
You don't set dt.container.uid yourself, but Dynatrace generates it if the telemetry carries the container identity attributes for your platform, such as k8s.pod.uid and k8s.container.name on Kubernetes or aws.ecs.container.arn on Amazon ECS.
OneAgent and Dynatrace Operator can add these attributes for you. For OpenTelemetry, you may need to set them yourself or enable a resource detector.
For the attributes and setup per platform, see Where it's supported.
Telemetry without these attributes can't be attributed to a container and is charged as Traces - Ingest & Process and Metrics - Ingest & Process instead.
For the supported environments and technologies, see Where it's supported.