The Davis data units model counts all incoming data points from your metrics. Each data point deducts 0.001 DDU from your available quota. If you send a metric via the API at 1-minute frequency, this translates into
1 data point x 60 min x 24 hours x 365 days x 0.001 DDU weight = 525.6 DDUs per year, per metric.
- All built-in metrics sent via the default OneAgent (for example, host metrics) are provided free of charge.
- With the exception of metrics booked by extensions on custom devices, any metric that is booked on a host (or process), regardless of source (extension or API), will first consume included metrics per host unit (i.e., 1,000 included metrics for Full-Stack Monitoring) before any DDUs are consumed. See below for full details. There are no included metrics for metrics booked by extensions on custom devices.
- DDUs are deducted for metrics that aren't included in your quota of included metrics per host unit (See table below). For more details about host units, see Full-Stack vs. Infrastructure Monitoring.
- Custom metrics typically consume DDUs. However custom metrics from OneAgent-monitored hosts are first deducted from your quota of included metrics per host unit, so they don't necessarily deduct DDUs from your quota. The same is true for custom log metrics.
- All metrics derived from Dynatrace monitoring extensions consume DDUs.
- All API-ingested metrics consume DDUs.
Included metrics per host unit
1,000 metrics per host unit are included on each monitored host (in Full-Stack Monitoring mode). These metrics are included with your Full-Stack host units and therefore don't consume DDUs. If a host has 16 GB RAM (i.e., 1 host unit), that host provides 1,000 included metrics that don't consume DDUs. For example, if an extension produces 1,500 metrics each minute, 1,000 of those metrics are included while the balance of 500 metrics will consume DDUs. Note that included metrics are bound to specific hosts that are monitored via OneAgent and can't be used for other product capabilities.
All metrics coming from OneAgent that are not extensions (host metrics, etc) are included as part of your host units and therefore don't consume DDU's.
For Infrastructure monitoring mode, consumption of host units is capped at a maximum of
1.0 host unit. Infrastructure monitoring mode provides 200 included metrics per host unit (see table below for details).
|Host size||Full-Stack mode||Full-Stack mode||Infrastructure mode||Infrastructure mode|
|Based on RAM GB||Host units||Included metrics||Host units||Included metrics|
|N x 16||N||N1 * 1,000||2001|
1 The minimum of included metrics is always
200, even when N is less than
Included metric consumption vs billed metric consumption
All values below are based on metrics sent at 1 min intervals.
|Mode||Host units||Included custom metrics per host unit1||Metrics reported from extensions||Included metrics available||Paid custom metrics||Deducted DDUs/min|
1 Included metrics are bound to individual hosts.
How dimensions per metric affect DDU consumption
These examples show how DDU consumption for a simple file-count metric can vary based on the number of hosts and dimensions that the metric is collected for:
2 hosts(i.e., 2 dimensions) equates to
2 x 0.001 = 0.002 DDUs
5 folders(i.e., 5 dimensions) equates to
5 x 0.001 = 0.005 DDUs
10 hostsequates to
5 x 10 x 0.001 = 0.05 DDUs
DDU consumption for built-in extensions
The Dynatrace monitoring extensions listed below initially consume each monitored host's quota of included metrics. Once the included metrics quota is exhausted, DDU consumption begins. The figures in the table below represent the expected consumption per process for each technology type that an extension monitors. The Custom metrics count reflects the default set of metrics that are generated by each extension.
|Elasticsearch Cloud||ActiveGate Extension||17||0.017||744.6||8,935.2|
|MS SQL||OneAgent Extension||18||0.018||788.4||9,460.8|
|MySQL||OneAgent or ActiveGate Extension||31||0.031||1,357.8||16,293.6|
|ActiveMQ JMX||JMX Monitoring||15||0.015||657||7,884.0|
|HornetQ JMX||JMX Monitoring||5||0.005||219||2,628.0|
|Apache Hadoop HDFS||JMX Monitoring||43||0.043||1,883.4||22,600.8|
|Apache Hadoop YARN||JMX Monitoring||48||0.048||2,102.4||25,228.8|
|Apache Spark||JMX Monitoring||17||0.017||744.6||8,935.2|
|Cassandra JMX||JMX Monitoring||28||0.028||1,226.4||14,716.8|
|Jetty JMX||JMX Monitoring||7||0.007||306.6||3,679.2|
|Kafka JMX||JMX Monitoring||42||0.042||1,839.6||22,075.2|
|Solr JMX1||JMX Monitoring||24||0.024||1,051.2||12,614.4|
|WSO2 API Manager||JMX Monitoring||27||0.027||1,182.6||14,191.2|
1 For Solr version 7+, each Solr Core will consume additional 24 custom metrics or 0.024 DDUs/min.
1 For Solr version 7+, each Solr Core will consume additional 24 custom metrics or 0.024 DDUs/min.
DDU consumption for remote extensions
Metrics sent via remote extensions always consume DDUs (OneAgent, ActiveGate, built-in, and remote extensions).
|Remote extensions||Consumes DDUs||Included per host unit|
|Citrix Virtual Apps and Desktops||Yes||Yes|
|F5 BIG-IP LTM||Yes||No|
|IBM Message Queue||Yes||No|
DDU consumption overview for metrics
Included metrics (no DDU consumption)
- Built-in OneAgent metrics (CPU, Memory, etc.)
- Kubernetes cluster metrics
- Kubernetes node/pod metrics on OneAgent-monitored hosts
Metrics that consume DDUs
- Calculated service metrics, custom DEM metrics, and log metrics
- Metrics ingested via the API
- All metrics from OneAgent and ActiveGate extensions (built-in and custom)
- Cloud services monitored by AWS, Azure, and GCP integration 1, 2
- VMware integration 1, 2
1 For the GA release only of Davis data units, cloud integration metrics are provided free of charge. These metrics will consume DDUs and be billed for in a release planned for early 2021. For more information, see DDU consumption estimates for cloud services.
2 Cloud integrations also provide metrics for virtual machine instances (AWS EC2 instances, Azure VM instances, GCP VM instances, VMware VM instances). Virtual machine metrics count against your allotment of host-unit included metrics when the underlying VM is monitored by OneAgent. For example, if you have 10 VMware VMs monitored via the VMware integration and 5 of those VMs are monitored via OneAgent, then the metrics of those 5 VMs will be included with the host units and no DDUs will be consumed. The metrics of the other 5 VMs will consume DDUs.
DDU consumption examples for metrics
- Metrics reported on a host monitored in Full-Stack Monitoring mode with OneAgent
(1 host unit = 1,000 host included metrics)
- Reported metric scenarios
- 450 reported built-in metrics from OneAgent (such as CPU idle time)
- 1,500 additional reported metrics from extensions (for example, JMX and RabbitMQ) booked on a OneAgent-monitored host
- 300 reported metrics via public API with no relation to a OneAgent-monitored host
The evaluation of whether or not a host exceeds the host-included metrics limit is made based on the total number of metric data points that are reported at a 1-minute frequency (i.e., there's no sliding time window for metrics). When the limit is exceeded for a certain time period, DDUs are only billed for those data points that exceed the limit during the time period.
If a host is licensed based on host hours rather than host-units, the mechanism works basically the same way; you can consume up to the number of host-included metrics per host unit hour. Each data point above the limit consumes DDUs.
Reported DDUs vs. consumed DDUs
You can compare the reported DDUs for a specific host with the number of host-included metrics by creating a chart with the Data explorer using the following metrics. This allows you to gain insights into the number of available vs. consumed DDUs for a specific host.
This metric is split by host ID. It shows you the host-included DDUs. For example, if a host has 1,000 included metrics, this metric will have a value of
1,000 x 0.001)
byEntityRawmetric is split by monitored entity and reflects the total reported DDUs before any host-included considerations are performed.
Similar to the
byEntitymetric is split by monitored entity. It reflects the total consumed DDUs (i.e., all consumption above the number of included metrics per host unit). This is equivalent to the information shown on the DDU consumption overview page (see example below).
Recommendations for working with this chart
- As DDU billing is calculated on a per-minute basis (take a look at the DDU consumption for metrics example for more details) it's recommended that you work with a 1-minute resolution in the chart as well.
- Using Split by and Filter by allows you to select a specific host that you're interested in.
The example Data explorer page above shows the following information:
The purple line reflects the host-included DDUs. In the above example this is
0.25DDUs (i.e., 250 custom metrics) on a per-minute basis.
The yellow line reflects the
byEntityRawmetric. It shows the total reported DDUs for a monitored entity, which is filtered by the host
Dynatrace. Any DDU consumption exceeding the host-included budget exceeds the purple line threshold.
The turquoise line reflects the information from the
byEntitymetric. It shows you the total consumed metrics. In the example above the host-included limit is exceeded about every 15 minutes between 09:30 and 10:45, and continuously beginning at 10:45. This results in the consumed DDUs, which are represented by the turquoise line chart.
Please note that data points from the
byEntity metric can be delayed up to a few minutes.
Automated conversion of custom metrics to DDUs
For existing customers with Dynatrace versions earlier than 1.198, your quota of existing custom metrics will be automatically converted into DDUs (there's nothing you need to do to make this happen). New accounts (Dynatrace version 1.198+) will have DDUs per default and no conversion will be triggered.
A customer bought 500 host units and one volume of custom metrics (500 metrics). Each Dynatrace environment (SaaS) comes with 100 free custom metrics. Additionally, 10 free custom metrics are provided for each of your allotted host units. There is a maximum free tier limit of 10,000 custom metrics.
100 + 10 x 500 = 5,100 custom metrics + 500 paid custom metrics = 5,600 custom metrics
The DDU conversion only converts paid custom metrics, as we also include 1,000 metrics per host unit (Full-Stack Monitoring mode). For Infrastructure mode, we include 200 metrics. This means that 500 paid custom metrics equates to 262,800 DDUs per year. On top of this we provide an additional free tier of 200,000 DDUs. In total, you get 462,800 DDUs plus the 1,000 included metrics per host unit (i.e., bound to the host). Note that only the portion of existing custom metrics in this example is converted to DDUs. More conversion examples can be found in the table below.
Davis data units licensing model
|Host units||Host unit hours||Free tier of DDUs||Paid custom metrics in DDUs||Included custom metrics per host unit1||Included custom metrics per host unit hours1||Total available DDUs|
1 Included metrics are bound to individual hosts.
Original custom metric licensing model
|Host units||Host unit hours||Free tier of custom metrics||Paid custom metrics||Included metrics from host units||Included metrics from host unit hours||Total included custom metrics (max. 10k)|
All metrics sent via the default OneAgent (host metrics, etc.) are included free of charge and do not consume DDUs. Metrics that are bound to specific hosts (or processes), regardless of source (extension or API), first consume all available included metrics before consuming any additional DDUs.
The following metric types consume DDUs:
|Metric type||Consumes DDUs||Included per host unit|
|Calculated service metrics||Yes||No|
|Calculated log metrics||Yes||Yes|
|Calculated DEM metrics||Yes||No|
|Kubernetes Cluster metrics||No||No|
1 Applies to the GA release of Davis data units only and is subject to change. These metrics will be billable in a future release.
2 If you use the REST API to send metrics to a OneAgent monitored host, those metrics will count against your allotment of host-unit included metrics. If metrics are sent to other entities, then those metrics will consume DDUs.
3 Cloud metrics that are bound to OneAgent-monitored hosts are deducted from your allotment of host-unit included metrics. For example, an EC2 instance metric counts against the host-unit included metrics of that EC2 instance. All other cloud metrics are subject to DDU consumption.
- Any metric that is booked on a monitored host or process group instance as these can be mapped to a specific host.
- Since Dynatrace version 1.203, custom metric ingestion can be booked directly on monitored hosts so that you can take better advantage of included host metrics.
- In a future release, ActiveGate extensions will also be able to book metrics on monitored hosts and thereby take advantage of included host metrics, just as OneAgent extensions do.
No, any metric that is booked on a host (or process), regardless of source (extension or API), will first consume included metrics per host unit (i.e., 1,000 included metrics for Full-Stack Monitoring) before any DDUs are consumed. For details, see metric cost calculation.
No, calculation of host units and host unit hours hasn't changed.
Each data point consumes
1 data point x 60 min x 24 hours x 365 days x 0.001 DDU weight = 525.6 DDUs
No, included metrics are bound to the host units you've already purchased. Those metrics are bound to the specific hosts where OneAgent is installed. There is no longer a global pool that host units contribute to.
Calculated service, web application, mobile app, and Synthetic Monitoring metrics can consume fewer DDUs per data point than expected if some of the servers in the cluster don't make contributions to a given data point. As metric data points are independently generated on each server in the cluster, the recorded consumption per server equals the total consumed DDUs divided by the number of servers. Therefore if, for example, only two of four servers in the cluster write data for a given metric during a given minute, only half of the normal number of DDUs will be consumed during that minute.
The DDU consumption breakdown shown on the
Monitored entities and
Metrics tabs on the Davis data units overview page (Settings > Accounting > Davis data unit overview) (see image below), as well as the
byMetric metrics in the API, can slightly differ because:
- Breakdown by monitored entity takes into account the included DDUs per host unit, which the breakdown by metric does not.
- Consumption by monitored entities can be delayed up to 10 minutes in comparison to the
byMetricmetrics in the API.
Therefore, for licensing and billing purposes, only the breakdown by monitored entity is considered (breakdown by metric is provided purely for informational purposes).
For metrics not related to a monitored host, consumption per monitored entity and consumption per metric are identical.
Yes, custom metrics via OneAgent extension are covered by included metrics. For Full-Stack Monitoring mode, this means 1,000 included metrics. For Infrastructure mode, this means 200 included metrics.
No, custom devices don't have included metrics. Included metrics are only available for hosts and processes that are monitored by OneAgent.
The following extensions utilize custom devices to ingest metrics: F5, Windows Server, Apigee Edge, SAP, Juniper Network, IBM I, IBM DataPower, Citrix Netscaler, and others.
ActiveGate extensions don't consume host-unit included metrics. Included metrics are only available for hosts and process metrics that are monitored by OneAgent.
The configuration change should only take a few seconds, with no negative impact on your metric consumption.
We recommend that you ingest custom metrics at a 1-minute frequency. When more metric data points per minute are ingested, more DDUs are consumed.
For example, when a custom metric is ingested at a frequency 10 seconds:
6 data points x 60 min x 24 hours x 365 days x 0.001 DDU weight = 3 153.6 DDUs
As stated above, for the GA release of Davis data units, cloud integration metrics (including AWS and Azure supporting services and VMware) are provided free of charge. These metrics will consume DDUs and be billed for in a future release. The logic here is that all metrics that belong to a OneAgent monitored host are either part of the basic value package (these are not custom metrics) or they count against the included custom metrics that come with each host unit. All other extension or integration metrics consume DDUs.
Yes, you can select which individual AWS and Azure services are monitored. All "supporting services" are available for monitoring on an opt-in basis. Also, by leveraging tag filtering, you can opt-out of monitoring individual "Core services" (those services that are automatically monitored).
Kubernetes nodes are automatically monitored when OneAgent is installed; these metrics are part of our core OneAgent offering and are included per host unit.