The primary use case is for application performance management. So, we are using it to identify outages of different parts of the application as well as how we can make the application more efficient and rightsize it.

How has it helped my organization?

We can see down into the layers, such as with databases. We can see database queries which are causing problems.

We can see CPU usage for different containers. I can do a run and see what errors exist in containers which are causing problems. We can rightsize containers on the fly and understand what is happening with our Docker, microservices, etc.

What is most valuable?

The most valuable feature is it has AI in it. The artificial intelligence (AI) engine in it is able to do alerts and some good analytics. During outages, it is able to identify and correlate where the actual root cause of a problem is. This connectivity allows us to be able to see a bit further into the application down through the layers. If it is a problem within AWS, a problem within a container or something that a user did. We are able to see and coordinate that, then we are able to tell the developers how to fix it.

What needs improvement?

The GUI has the most room for improvement. Sometimes, it can be a little cumbersome to find things and be able to create your own views, or be able to dig in and understand where things are.

Some additional features would be the ability to break out some of the permissions and allow some additional or different ways to tag services, events, and different things which run. We want to push down the ability to do that, so developers and other folks can get in there. Currently, more permissions are needed to be able to do certain things, and we want more people to be able to use it, own it, and understand it.

For how long have I used the solution?

One to three years.

What do I think about the stability of the solution?

We don't put very much stress on it. We could probably stress it some more, but we don't have enough systems right now on it to stress it. For the most part, the ships don't cause as much stress.

We are going to have it on about 40 ships around the world which will run it independently of our AWS platform. Those are don't stress it too much. We will probably stress it at a certain point, along with AWS, but we still very much growing the platform.

What do I think about the scalability of the solution?

It can scale very well and very high. We don't need it to scale as much right now. It is able to absorb a lot of the systems that we have with the agents and and the API Gateways. It seems like it can scale very well when we need it to, so scalability is good for us right now.

How is customer service and technical support?

If we needed technical support, we usually call our account team to help us figure out where the errors are, whether it is something with an agent or management servers.

How was the initial setup?

It is pretty easy to integrate it into the AWS environment. You give it a username and password and it asks some basic permission. It can pull a lot of information very quickly. We are able to correlate more and provide more data for it. So, it was easy to integrate it into that environment.

We have it running on AWS. It integrates pretty well there. We have it on Red Hat Linux servers, as well as Windows servers. We have it running on VMware where it integrates very well. It understands these productions and understands our platform. It is able to read into Docker containers and all the databases that we run. However, it is limited as far as how many of a certain type of database that we can have, but for the most part, it runs pretty well and integrates very well.

What was our ROI?

It has been doing a good job of alerting us to issues. It has been very helpful and effective at identifying how we can do things to make our infrastructure and application a little better.

Which other solutions did I evaluate?

We considered AppDynamics, Datadog, and Crashlytics. We even considered things like Splunk for different pieces of it.

We chose Dynatrace because we needed something which could run both on AWS and VMware on our ships that might lose their Internet connectivity. This product gave us the flexibility of being able to do both. Dynatrace had the ability to run independently, so we could access it while it retains information.

What other advice do I have?

A PoC is the best way to go. Put it against an application and go through the paces of tagging, analyzing, and alerting on it. You can understand what it does and how it does it. Give it a very complex application, so you can see how well it works.