This topic applies to Dynatrace Managed installations only.
Supported software platforms
For details on which 64-bit Linux distributions you can install Dynatrace Managed on, please see Which environments and versions does Dynatrace support? Note that installation on both physical and virtualized hosts is supported, but installation in containers is not supported.
Security Gateway requirements
Please visit How do I install Dynatrace Security Gateway?
|Node type||Max hosts monitored (per node)||Max user actions/s (per node)||Minimum node spec1||Disk IOPS (per node)||Transaction Storage2 (for 10 days code visibility)||Long-term Metrics Store3 (per node)|
|Trial||20||20||4 vCPUs, 16GB RAM||30||250GB||100GB|
|Small||250||250||8 vCPUs, 32GB RAM||150||500GB||1.2TB|
|Medium||600||500||16 vCPUs, 64GB RAM||300||1TB||2.5TB|
|Large||1500||1000||32 vCPUs, 128GB RAM||750||2TB||5TB|
|XLarge||3000||3000||64 vCPUs, 256GB RAM||1500||4TB||10TB|
1 CPU and RAM must be exclusively available for Dynatrace. Power saving mode for CPUs must be disabled. CPUs must run with a clock speed of at least 2GHz.
2 Transaction data is distributed across all nodes and is not stored redundantly. So for multi-node clusters, the storage per node is divided by the number of nodes.
3 For multi-node installations we keep three copies of the metrics store. For 4 or more nodes the storage requirement per node is reduced.
All nodes of a cluster should have the same hardware. Recommended cluster sizes are 1, 3, or more (up to 10). The hardware requirements are estimates based on typical environments and load patterns. Requirements for individual environments may vary.
For high availability installations with nodes distributed across different data centers or installations with more than 3 nodes please contact firstname.lastname@example.org for detailed recommendations.
To monitor 8k hosts with a peak load of 3k user actions per second, you need 3 extra large nodes with a combined storage of 4TB for direct storage and 30TB for long-term metrics.
To monitor 200 hosts with a peak load of 500 user actions per second, you need 1 medium node with storage of 1TB for transactions and 2.5TB for long-term metrics, or to have failover you can also use 3 small nodes.
Disk space requirements of Dynatrace Managed directories
|Directory symbol||Directory path||Description||Maximum expected disk space|
|PRODUCT_PATH||/opt/dynatrace-managed||Main directory for Dynatrace Managed binaries||5GB|
|DATASTORE_PATH||/var/opt/dynatrace-managed||Main directory for Dynatrace Managed data||depends on data size|
|LOG_PATH||/var/opt/dynatrace-managed/log||Logs of all Dynatrace Managed components, services and tools||1 GB|
|CASSANDRA_DATASTORE_PATH||DATASTORE_PATH/cassandra||Metrics repository||depends on data size|
|ELASTICSEARCH_DATASTORE_PATH||DATASTORE_PATH/elasticsearch||Elasticsearch store||depends on data size|
|SERVER_DATASTORE_PATH||DATASTORE_PATH/server||Transactions store||depends on data size|
|AGENT_BUILD_UNITS_PATH||DATASTORE_PATH/agents||Installation packages of agents (if downloaded by Server or installed from standalone agents package)||5GB|
|SERVER_BUILD_UNITS_PATH||DATASTORE_PATH/installer||Dynatrace Managed installer for adding nodes to cluster, prepared during installation/upgrade||2 GB|
The directory paths included in the above table are the default ones. They will be different in case you have chosen custom locations during installation.
For Dynatrace Managed installations with more than one node, all nodes must:
- Use the same Linux distribution
- Have the same hardware configuration
- Be synchronized with NTP
- Be in the same timezone