Host monitoring (DPS)

Dynatrace offers Full-Stack Monitoring and Infrastructure Monitoring for the hosts and containers in your digital infrastructure. To switch between Full-Stack Monitoring and Infrastructure Monitoring modes, see Enable Infrastructure Monitoring Mode.

Full-Stack Monitoring

Full-Stack Monitoring offers comprehensive application performance monitoring, including distributed tracing, code-level visibility, CPU profiling, memory profiling, and deep process monitoring. That is true for hosts and containers. For hosts, all Infrastructure Monitoring functionality is included with Dynatrace Full-Stack Monitoring. A host is defined as a single physical or virtual server or cloud instance running an operating system, whereas a container is defined as an isolated space on a host or any cloud infrastructure for running individual application processes.

A Full-Stack monitored host has OneAgent installed in Full-Stack Monitoring mode, actively sending data to Dynatrace. Billing of Full-Stack monitored hosts (whether run in the cloud or on-premises) depends on a host's memory size and the duration a host was monitored by OneAgent.

GiB-hours

Dynatrace uses GiB-hours (referred to as "memory-gibibyte-hours" in your rate card) as the unit of measure for calculating your organization's consumption of host monitoring in Full-Stack Monitoring mode. The more memory that a host has, and the longer that the host is monitored, the higher the number of GiB-hours that the host consumes.

The advantage of the GiB-hour approach to monitoring consumption is its simplicity and transparency. Technology-specific factors (for example, the number of JVMs or the number of microservices hosted on a server) don't affect consumption. It doesn't matter if a host runs .NET-based applications, Java-based applications, or something else. You can have 10 or 1,000 JVMs; such factors don't affect an environment's monitoring consumption.

Billing granularity for GiB-hour consumption

Dynatrace is built for dynamic cloud-native environments where hosts and services are rapidly spun up and destroyed. Therefore, billing granularity for GiB-hour consumption is calculated in four 15-minute intervals per hour. When a host or container is monitored for fewer than 15 minutes in an interval, GiB-hour consumption is rounded up to 15 minutes before consumption is calculated.

GiB-hour calculation for physical hosts and virtual machines (VMs)

Each installed instance of Dynatrace OneAgent running on an operating system instance (deployed on either K8s, a container platform, or a physical or virtual machine) in Full-Stack Monitoring mode consumes GiB-hours based on the monitored host's physical or virtual RAM, calculated in 15-minute intervals (see the diagram example below).

The RAM of each VM or host is rounded to the next multiple of 0.25 GiB (which equates to 256 MiB) before monitoring consumption is calculated. A 4 GiB minimum is applied to GiB-hour consumption for physical and virtual hosts. For example, a host with 8.3 GiB memory is counted as an 8.5 GiB host, being the next multiple of 0.25 GiB, while a host with 2 GiB memory is counted as a 4 GiB host (no rounding needed, but application of the 4GiB minimum).

GiB-hour calculation for containers and application-only monitoring

In cloud-native environments, services and hosts are often short-lived. Therefore calculating monitoring consumption in 15-minute time intervals, rather than full hours, better reflects your actual usage. Containers, which are an essential mechanism in cloud-native environments, are typically smaller in memory size than hosts. Therefore, the minimum memory threshold for containers is 256 MiB, rather than 4 GiB, the minimum memory threshold for hosts. The same rounding as for hosts, to the next multiple of 0.25 GiB, also applies to containers. For example, a container with 780 MiB memory is counted as a 1 GiB container (780 MiB, which equals 0.76 GiB, being rounded up to the next multiple of 0.25 GiB).

This diagram illustrates how memory is counted for GiB-hour consumption calculations at 15-minute intervals.

Figure 1. GiB-hour consumption for hosts and containers is based on memory size, calculated in 15-minute intervals each hour. Each interval is divided by 4 in order to reach the GiB-hour consumption unit of measure.

Metrics

Full-Stack Monitoring includes all Infrastructure Monitoring metrics, application performance monitoring metrics, and other built-in metrics. In addition, Full-Stack Monitoring includes a fixed number of custom metric data points for each GiB that contributes to your environment's GiB-hour consumption. As of April 26, 2023, Dynatrace offers 900 included custom metric data points for each GiB of host memory, calculated at 15-minute intervals.

Your environment's included metric data points are applied automatically to metrics that originate at hosts and containers that are monitored by OneAgent in Full-Stack Monitoring mode. You will only be billed for consumed metric data points if you exceed your included volume of metric data points.

Included metric data point calculation example

Considering the example shown in Figure 1, here are the calculations for the included metric data point volumes for each of the four 15-minute intervals, assuming a volume of 900 included metric data points for each 15-minute interval.

First 15-minute interval
900 (included metric data points) x 13.5 (GiB memory) = 12,150 included metric data points

Second 15-minute interval
900 (included metric data points) x 9.5 (GiB memory) = 8,550 included metric data points

Third 15-minute interval
900 (included metric data points) x 8.75 (GiB memory) = 7,875 included metric data points

Fourth 15-minute interval
900 (included metric data points) x 0.25 (GiB memory) = 225 included metric-data-points

How metric data points are consumed in Full-Stack Monitoring mode

Metric data-point consumption takes many forms. An equal number of data points can be consumed:

• By a few high-resolution metrics or numerous low-resolution metrics.
• Equally across multiple 15-minute intervals or all at once in a single minute.
• By all monitored hosts, a subset of all Full-Stack monitored hosts, or a single host.

Additional metric data points that your environment consumes (beyond those data points that are included with each GiB of memory that contributes to your consumed GiB-hours) are billed as Custom Metrics Classic.

Distributed traces

Full-Stack Monitoring includes Dynatrace PurePath® distributed tracing. OneAgent automatically manages the volume of captured trace data. For details, see Adaptive Traffic Management documentation. The peak trace volume available in an environment at any time depends on how many GiB of memory contribute to your GiB-hour consumption. Every contributing GiB of host memory adds peak trace volume of 45 KiB/min. Each environment has a minimum trace peak volume of 14 MiB/min.. This trace volume is available for all traces sent by OneAgent code modules or via OneAgent Trace API.

Taking the hosts shown in Figure 1 as an example, the peak trace volume for each minute equals the environment minimum (14 MiB/min). For the calculation of trace data volume in this next example, we'll apply a multiplier (100) to host memory size to generate a more realistic scenario. This assumes there are 100 times more GiB contributing within each 15-minute interval.

Peak trace volume/minute calculation example

First 15-minute interval
45 KiB (peak trace volume) x 13.5 (GiB memory) x 100 (example multiplier) = 59.32 MiB/min

Second 15-minute interval
45 KiB (peak trace volume) x 9.5 (GiB memory) x 100 (example multiplier) = 41.75 MiB/min

Third 15-minute interval
45 KiB (peak trace volume) x 8.75 (GiB memory) x 100 (example multiplier) = 38.45 MiB/min

Fourth 15-minute interval
45 KiB (peak trace volume) x 0.25 (GiB memory) x 100 (example multiplier) = 14.00 MiB/min (the minimum trace volume)

Note that traces sent via the Custom Trace API are billed as Custom Traces Classic.

Dynatrace retains the total amount of ingested trace volume from your environment for ten days, with index-level access for 35 days. For an average host with 16 GiB memory, this translates to 10 GiB of retained trace volume (45 Kib per GiB/min x 16 GiB memory x 60 min x 24h x 10 days / 1024 / 1024).

CPU, memory, and thread profiling

Full-Stack Monitoring includes CPU, memory, and thread profiling for technologies like Java, .NET, Go, Node.js, and PHP. OneAgent uses an intelligent patented mechanism to manage the volume of profiling data. Dynatrace retains the total amount of ingested profiling data from your environment for ten days.

Infrastructure Monitoring

Dynatrace OneAgent can be configured for Infrastructure Monitoring mode, which provides physical and virtual infrastructure-centric monitoring, along with log monitoring and AIOps.

Host hours

The unit of measure for calculating consumption of host monitoring in Infrastructure Monitoring mode is a host-hour. Each instance of Dynatrace OneAgent installed and running on an operating system instance (deployed on either a physical or virtual machine) with Infrastructure Monitoring mode enabled consumes host hours. The longer that a host is monitored, the more host-hours you consume.

Billing granularity for host-hour consumption

Dynatrace is built for elastic cloud-native environments where hosts and services are rapidly spun up and destroyed. Therefore, billing granularity for host-hour consumption is based on 15-minute intervals. When a host is monitored for fewer than 15 minutes in an interval, host-hour consumption is rounded up to 15 minutes before consumption is calculated.

Figure 2 below illustrates how host-hour consumption per host is calculated.

Figure 2. Host hour consumption per host, calculated at 15-minute intervals.

Metrics

Dynatrace Infrastructure Monitoring includes important built-in metrics. In addition to built-in metrics, Infrastructure monitoring includes a fixed number of custom metrics data points (we currently offer 1,500) for each 15-minute interval of monitoring per host. This equates to 100 custom metrics data points per minute at no additional cost. This "included" data point volume is available for all Infrastructure-monitored hosts in your environment. It applies only to metrics that originate at hosts that are monitored by OneAgent in Infrastructure Monitoring mode.

Looking at Figure 2 above, the included custom metric data point volume for the four 15-minute intervals is shown below.

Included custom metric data points calculation example

First 15-minute interval
1 (hosts monitored) x 1,500 (metric data points) = 1,500 Included custom metric data points

Second 15-minute interval
2 (hosts monitored) x 1,500 (metric data points) = 3,000 Included custom metric data points

Third 15-minute interval
1 (hosts monitored) x 1,500 (metric data points) = 1,500 Included custom metric data points

Fourth 15-minute interval
1 (hosts monitored) x 1,500 (metric data points) = 1,500 Included custom metric data points

How custom metric data points are consumed in Infrastructure Monitoring mode

Custom metric data-point consumption takes many forms. An equal number of custom metric data points can be consumed:

• By a few high-resolution metrics or numerous low-resolution metrics.
• Equally across multiple 15-minute intervals or all at once in a single minute.
• By all Infrastructure-monitored hosts, a subset of all Infrastructure-monitored hosts, or a single infrastructure-monitored host.

Additional custom metric data points that your environment consumes (beyond those data points that are included with each Infrastructure-monitored host) are billed as Custom Metrics Classic.