cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

autoscaling automation of dynatrace managed

krunal_shimpi
Newcomer

We are running dynatrace managed solution.

According to the link "https://www.dynatrace.com/platform/dynatrace-architecture/" , the SaaS offering of dynatrace is hosted on AWS cloud and uses autoscaling with automation to provide the following benefits.
1. AWS Auto Scaling and Elastic Load Balancing capabilities ensure that new Amazon EC2 instances spin up automatically to meet spikes in demand and/or traffic.
2. Real-time failover redundancy built into all system components.
3. Automated management layer detects faulty components and replaces them immediately.
4. One-third of excess capacity in our clusters always reserved for real-time failover and ability to cope with peak loads in real time.
5. Guaranteed ≥99.5% uptime.

According to the link "https://www.dynatrace.com/platform/deployment-models/" with Dynatrace managed we "Gain all the conveniences of Dynatrace SaaS".

Can you tell us the steps to achieve automation of the dynatrace managed cluster with autoscaling on AWS. Setting up a cluster node is a manual process it seems, where the commands to set up the master node are sent on activation mail with an expiring token. The commands have to be copied and manually executed following by adding more cluster nodes.

How do we derive the benefits mentioned above if the steps require manual intervention. Can you please provide technical steps to achieve the automation that SaaS solution provides.


3 REPLIES 3

You will not have autoscalling ability like on AWS. Cloud solutions can extend resources like memory, cpu and storage on the fly. This is something needed in such approach because Dynatrace has to optimize consumption and costs.

In manage you are responsible for infrastructure, you can handle resources extensions. Isetting up extra cluster on deman when traffic is bigger is not good idea because some of data are kept only on cluster where they was recorded. When you are using Dynatrace in browser you don’t see this because you are queried both clusters. So if such cluster would go offline you can loos access to some part of data.

Sebastian


Hi Sebastian,

Thanks. Our application hosted on AWS- almost all of them are automated using some form of Infrastructure-as-code fashion. But with dynatrace managed we can't achieve that. We are not always looking for setting up a new cluster but we want our infra to be codified and manageable. but current managed solution doesn't allow that.


Managed is different production than saas on such case. You can monitor Dynatrace host via OneAgent or cloudwatch and decide how to manage resources from such aproach.

Sebastian