What is our primary use case?
Dynatrace is an extremely helpful APM solution in large, complex environments, whenever we need to have an integrated vision of our users, infrastructure, and applications. It excels when we need to do all this with the lowest resources possible, with the best data quality, across all layers, and with the least management overhead.
How has it helped my organization?
The first time I came across Dynatrace, I simply remember the WOW factor. Being a performance enthusiast since the mid 80's, I cannot remember more than a handful of such technical experiences!
Dynatrace excels in troubleshooting scenarios. You do not get samples or gaps. You get the real thing. That is absolutely important when troubleshooting the strangest situations.
Focus on root cause analysis has dropped significantly for the importance of alerts. I was used to be drowning in alerts, but now have refocused. I still have them, of course, but now they are about minor things. I see myself getting rid of them very soon.
What is most valuable?
The most valuable feature is Dynatrace's commitment to bringing in new features. It is very difficult to keep pace with the new developments that are always occurring.
Given the full-stack approach, you are able to view from the end user perspective, almost down to the bare metal. It gives you all the way from the user perspective, network experience, web and application server view, application detail, process detail, operating system detail, and you even get to manage virtualization and datacenter support.
Dynatrace is capable of following all user actions across all monitored infrastructure, even to systems not monitored, but interconnected. This technology is known by the name PurePath, and being able to do this sometimes feels like magic. I have been told, time and time again, by very knowledgeable IT personnel, that this is not possible. Sometimes, you have to see it, to believe it.
One of the most promising features is the AI capability. I have seen some of its capability in production environments, and it is pretty impressive! It is not a marketing buzzword being abused, but it actually works. Being able to offload problem handling and root-cause analysis, and get to the point analysis of complex systems is something that I have not seen with my eyes, and in this depth, in any other IT solution to date.
Dynatrace's online presence in APM "education" is second to none. Viewing the YouTube online videos, hearing the PurePerformance Podcast, getting to learn in Dynatrace University, and following the documentation is like being in performance heaven! You just can't grasp it all!
What needs improvement?
In the new Dynatrace solution, support for legacy applications is still not there. Given there is excellent support of legacy applications and protocols in the Appmon & DC RUM offers, they have the knowledge to put it there. Knowledge that their competition simply doesn't have.
For how long have I used the solution?
One to three years.
What do I think about the stability of the solution?
Not seen one stability issue since I been dealing with it.
What do I think about the scalability of the solution?
No issues with scalability. The solution is multi-tenant and based in Big data technology.
What impresses me more is the ability to be able to analyze 100% of all activity going on, and do that with so low overhead. I have not been able to observe more than 1% overhead, despite Dynatrace saying that it can be slightly higher in some situations. I sometimes get the impression that overhead might be negative (underhead?), because you can get rid of inefficiencies like log dumping, that typically have high (disk and CPU) overhead.
How are customer service and technical support?
Support is just great. Part of it is public, so everyone can check it out. Dynatrace direct support is even better, and you sometimes get direct answers from the teams that are implementing the functionality.
Which solution did I use previously and why did I switch?
I have worked with other APM, IT, and UX monitoring solutions before. None of them is even near what Dynatrace has to offer.
How was the initial setup?
Never seen anything so complex be so simple to install.
What about the implementation team?
What's my experience with pricing, setup cost, and licensing?
Pricing can be high, especially for Portuguese standards. But as one says, you get what you pay for.
Setup cost is very low considering that it is an almost totally, automatic process. Installing SaaS or managed is only some minutes away. Given that there is no configuration involved in the agents, you can develop how many you want per hour. It only depends on your IT deployment strategy. TCO is thus much lower than expected. Licensing is very interesting, as you pay only for what is being monitored. A lot of things are given away for no additional cost. If you have a great IT consolidation, it will be pretty cheap. If you have a lot of servers, it will be heavier.
What other advice do I have?
Since you cannot manage what you cannot measure, I do give the most importance to data quality. This is priceless. If you manage based on bad or incomplete data, you are leaping into the bad decisions direction.
The return might just be immediate. You install it and after minutes you are getting the full data in. The other day I compared Dynatrace to another APM solution, and the other person had been struggling nine months to get the data out. When he saw what Dynatrace did out-of-the-box, he simply could not believe it.
Finally, be prepared to be surprised! It is very fast pace. The Session Replay and the Augmented Reality are just two recent examples. Almost every day I get some new perspective in this field, like AIOps, and it just keeps getting faster!
Disclosure: My company has a business relationship with this vendor other than being a customer: Dynatrace Partner.