cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Looking to upgrade from Dynatrace Managed to SaaS? See how

What’s new with Dynatrace Managed?

miriamderenbach
Dynatrace Enthusiast
Dynatrace Enthusiast

We have been busy developing and delivering updates to Dynatrace Managed! In this blog post, we're taking a close look at a few selected features that will be available to our Dynatrace Managed customers in the next few weeks.  

 

Now that we have a growing team dedicated to working on Dynatrace Managed exclusively, we can deliver both bigger and smaller feature updates simultaneously. Today, we'd love to share a sneak peek at a few smaller updates ahead of their release.  

 

Updating the JDK for Dynatrace Server to JDK 21 from JDK 17 


The Dynatrace Cluster Server nodes will be upgraded to run with JDK 21 instead of JDK 17. Large-scale internal load-testing has indicated that switching the JDK can show a considerable reduction of time spent for Java garbage collection and thus can lead to improved performance. For now, only Dynatrace Server switches and other components remain untouched. ActiveGates will continue to run with JDK 17, and the current JDK will continue to be used for building packages, just like before. There is no need for manual updates to benefit from this update, as it should happen via rolling updates as part of the normal version updates.  

 

Watch your disk storage go down 


We have changed the default compression algorithm for Cassandra to zstd from the current LZ4. As a result, the size of timeseries column families could be reduced by up to 30%! Only timeseries data is affected, so the impact on your environment can vary depending on feature usage. Because existing data can't be rewritten and the changes will only affect newly written data, the change won't be instant, so it can take some time for users to see the difference. There will be no impact on cluster update time. There is also no need for manual updates, as we have changed the default with the production rollout for version 314, and the new compression will be rolled out as part of the normal upgrades.  

 

Faster assignment of tags, management zones, and naming rules 


A new periodic worker in Dynatrace will apply tags, management zones, and naming rules faster to recently created monitored entities (hosts, processes, services, etc.). This worker runs continuously and considers all monitored entities created in the last 30 minutes and seen in the last 2 hours. 

It is automatically enabled by default for all Managed environments. No additional configuration is required. 

 

Updates of third-party components 


We have updated several third-party components to ensure critical security updates and bug fixes are addressed. These updates enhance security features, improve performance, and increase stability, helping us stay ahead of potential vulnerabilities and maintain a robust infrastructure. 

  • Netty 4.1.94 will be upgraded to Netty 4.1.118 for Dynatrace Server 
  • Cassandra 4.1.7 will be upgraded to Cassandra 4.1.18 

 

What’s next? 


Next, we're working on getting ActiveGates to run with JDK 21 as well. Some new features in JDK 21 will allow us to tune garbage collection settings and further optimize CPU usage on Dynatrace Server.  

We also continue to update components and continue to look for ways to optimize Dynatrace Managed to provide the best performance. 

 

Watch out for the new release notes to hit your inbox and find the features listed above alongside other updates for Dynatrace Managed.  

4 REPLIES 4

Islam_Zedan
Champion

Thank you for the useful updates :dynatrace:

Dynatrace Certified Professional - Dynatrace Partner - Yourcompass.ca

Theodore_x86
Helper

Will the new CASSANDRA compression algorithm reduce also the size of reserved storage for Compaction process? This storage causing most of disk exhaustion issues.

BR

Houston, we have a problem.

Hi,

The compaction process should also use less disk-space, but we did not adjust reserved storage settings/guidance at this point as the amount of actually reduced disk-space depends a lot on usage-patterns.

We plan to tune the guidance for reserved disk-space for compaction in a follow-up step after we assessed the actual reduction seen in production systems.

 

Regards... Dominik.

Thanks for the reply Dominik!

Houston, we have a problem.

Featured Posts