Dynatrace Managed has to be considered as a packaged solution with several components.
Cassandra stores configuration data and metrics. Elasticsearch stores RUM data and log files for Log Monitoring v2 (logs classic). NGINX servers as load balancer and routes communication to particular endpoints and between cluster nodes.
@natanael_mendes in my previous post - it's a boxed solution, so you should not care about what data is stored in which technology as it may change between versions. Also you should not make any changes in the Cassandra or Elastic setting on the Managed node unless instructed to do so from the Dynatrace support.
About the data - I don't think this is documented in the public docs, but you can find relevant data in this post.
Please be aware on the following fact :
There are no sync transaction store so this means ElasticSearch and Cassandra sync between nodes which can also have an overall impact on storage as well; Thus the suggestion would be to spread this out and give each node its own storage type (a mount for cassandra, a mount for ElasticSearch and a mount for transaction storage for instance) as a minimum recommendation.
For example, where Elastic storage and Cassandra all storages are on the same disk then reserved disk space is calculated as the sum of minimum required free disk space required by the respective data contributors (plus a 10% safety margin).
The individual calculation is done as:
For Cassandra reserved disk space: Current Cassandra usage * 0.6 (enough space for cassandra to do reorganization). Consequently take care during the backup operation due to disk compaction.
interesting post : Solved: Re: Backup Size Ratio Cassandra vs. Elasticsearch - Dynatrace Community
More info hardware requirements :
Managed hardware requirements | Dynatrace Docs
I hope that answered