Our main use case for Dynatrace is that we deployed Dynatrace OneAgent across our AKS nodes to monitor pod level metrics and service dependencies. Smartscape topology helped us visualize the entire environment in real-time and identify issues such as memory leaks and high response times during load. For our backend services, we were facing intermittent latency issues, but using Dynatrace we could pinpoint the exact methods and queries causing the application slowness and understand what exactly was happening.

Dynatrace offers several best features. When comparing before integrating and using Dynatrace, we also used other tools including Prometheus and Grafana, but Dynatrace provides that additional feature. Nowadays, everything incorporates AI, and Davis AI that Dynatrace built-in dashboards are clean and powerful. The AI integration allows us to outperform the issues, reduce the downtime, and timely resolve the issues. Their AI integration is exceptional and gives Dynatrace a better capability compared to competitors.

Davis AI and the built-in dashboards have made a difference for our team because whenever we were stuck on any level issue or metrics and wanted to dig into some logs and metrics to identify where exactly it was going wrong on our application side, the AI helped us to narrow down where exactly it was going wrong. It identifies the particular specific pod, application, or container and provides a whole overview of the application and container level. We had to fine-tune it to align with our environment. We fine-tuned it in such a way that whenever something goes down on our containerization or VM level, we could drill down and by using the AI analysis, it looks into the entire system and provides an exact pinpoint solution when we give a detailed outline. This reduced the resolution problem to a lesser time compared to using other tools.

Dynatrace has positively impacted our organization. We have had major outages across the organization, whether application side or performance level related, with users experiencing slowness or blank pages. However, Dynatrace setup on an enterprise level brought down the outage time by half. All cross-functional teams can integrate and see exactly what is going on, allowing us to work on the actual resolution based on the metrics provided by Dynatrace. Using this tool, it pinpointed the problem instead of us scratching our heads trying to identify where it could go wrong, and we could directly proceed with the solution. This reduced our man-hours and outage time while increasing productivity based on performance metrics and observability.

We encountered some challenges while using Dynatrace. Although the initial setup was smooth, fine-tuning alert thresholds and custom metrics took some time. Another challenge was that Dynatrace charges based on host units, so we had to carefully plan our agent deployments. The licensing model is expensive. Additionally, the complexity of setup is an issue. While OneAgent and auto-discover services are powerful, the setup is more complex compared to other tools such as Prometheus and Grafana. These integrations are simple and basic, but Dynatrace setup requires more complexity based on the environment. For new users wanting to use Dynatrace, it is difficult. However, the AI-related solutions and metrics took us to the next level for identifying and fixing things.

Dynatrace requires an agent for operation. OneAgent is powerful, but it is also resource-heavy. On lightweight nodes or older systems, the agent can slightly impact performance. If Dynatrace could implement a lightweight agent behavior, we could make things faster. Additionally, if Dynatrace could add a long-term retention policy so that we could store more data and find fine-grained details, that would help us. While Dynatrace managed edition supports on-premises deployment, the SaaS version depends on cloud connectivity. For highly regulated or air-gapped environments, setup and updates can be challenging. Although the initial setup is smooth, if someone wants to fine-tune it and fully understand the tool end-to-end, it could be tricky.

Dynatrace is stable.

Dynatrace scalability is great. It is a powerful tool and helped us to reduce customer downtime and increase work efficiency. We could identify any issues causing problems in our application or environment.

Customer support is very prompt. Whenever we faced any issues, we could get timely resolution from their support, so we did not face any issues.

Positive

Before Dynatrace, we used open source Prometheus. It is a very standard solution but not exceptional. It is a standard option because it is open source and free.

My experience with pricing, setup cost, and licensing is that Dynatrace licensing model is based on host units and can be complex initially to understand. If setting up in a large scale environment, it is overwhelming because it is expensive. We had to plan carefully when deploying OneAgents across our nodes or clusters, ensuring we did not exceed our licensing capacity.

For purchasing Dynatrace, we explored the AWS Marketplace, but on an enterprise level, we reached out to their agents and they set up a call for integration. We deployed on an enterprise level because while the marketplace is fine, if implementing a solution at a huge enterprise level, we had to work with the actual Dynatrace team to set it up. For some of the bare metal hosts or engines, we went with the marketplace only.

When it comes to metrics and performance, we have clear return on investment. Before Dynatrace, we spent considerable time manually correlating logs, metrics, and traces from multiple tools to find root causes. Using Dynatrace directly improved application uptime and reduced customer impacting incidents. It helped our engineers focus on optimizing rather than firefighting.

My experience with pricing, setup cost, and licensing is that Dynatrace licensing model is based on host units and can be complex initially to understand. If setting up in a large scale environment, it is overwhelming because it is expensive. We had to plan carefully when deploying OneAgents across our nodes or clusters, ensuring we did not exceed our licensing capacity.

Before choosing Dynatrace, we considered DataDog and other tools available in the market, but compared to all other tools and their values that could bring to our organization, we chose Dynatrace.

My advice for others looking into using Dynatrace depends on the organization. My main advice is that if you want an enterprise-level observability and monitoring solution for a hassle-free experience with troubleshooting, identifying and reducing customer downtime, and increasing work efficiency, or if you want a stable and standard solution, Dynatrace is the best fit. This review received a rating of nine out of ten.

My main use cases with Dynatrace include a plethora of functionalities. Real user monitoring (RUM) is being used extensively. Synthetic monitoring is used extensively. We use Kafka monitoring. I have awareness of how AWS metrics are being sent, whether it's a direct integration or only account-level integration. We use it significantly.

Some of the best features I appreciate about Dynatrace include synthetic monitoring, where you can get into HTTP level monitoring and click browser monitoring, which is available in both Splunk and Dynatrace. This is very useful in our environment. Apart from that, mobile-based monitoring, which we have embedded in some cases with the apps that are connected, is also beneficial for monitoring APIs.

Synthetic monitoring has had a significant impact on my ability to track performance proactively. It has been very useful. It serves two aspects: synthetic monitoring is primarily for the front-end side where availability and tracking whether the website is running, and we can verify if users are able to log in and see things running.

From the infrastructure point of view, the availability of the infrastructure at AWS level with Kafka, EC2 instances, and Lambda functions is covered for the monitoring system and infrastructure team. We are catering to both audiences.

Dynatrace's AI-driven Davis engine absolutely helps identify performance issues by showing root cause analysis for us up to 200%; whatever is integrated, if it is visible, it can stitch and show.

The comprehensive application topology visualization in Dynatrace, called Smartscape, has benefited my understanding of system dependencies. Even though Splunk has something similar, since we are using Dynatrace primarily for observability, we can get top-to-bottom visibility, from infrastructure to network to application front end. When integrated based on application requirements, we get a good grasp of what's happening during issues.

Dynatrace could be improved primarily with regards to pricing. These tools are expensive. They have been the pioneers from inception, and they remain at the top of the Gartner chart. The quicker we learn, the better we can serve is what the team believes.

Learning is another aspect that needs improvement, specifically how the tools can educate users. Education needs to be tailored differently for front-end user monitoring, infrastructure-based monitoring, and centralized monitoring teams. The segregation of educating how to use the tool is something we have recommended to all tools teams and product owners.

They could help with more learning capabilities. A monitoring tool will be used by different types of users. One is from the infrastructure point of view. Another is an application developer who wants to see if strings are getting attached and code-based applications are being stitched together. A front-end monitoring person or business wants to know about website and infrastructure availability, and what kinds of dashboards they can create for their comfort. That's how a tool gains visibility and inclusiveness. It depends on the owner of the tool to address these different aspects and let users choose their preferred way to use the tool.

Regarding the initial setup, while I can't speak for how my company implemented it overall, I can say that the Dynatrace setup is good enough, not an issue. Integration with the cloud is straightforward. In cases of new aspects of cloud, some drilling is needed to determine installation possibilities. That's a challenge in the cloud - whether to integrate with the cloud directly or opt for agent installation. These issues arise when new features or services are enabled in AWS; parallelly, Dynatrace and Splunk and these tools need to adapt to see if these services can be monitored. Users start asking if we can enable new AWS features in our tools. This synchronization should happen at the back end; users should not be involved in that process.

Positive

The monitoring team is able to work without any user intervention, which is appreciable. Comparing to other tools such as Splunk, CloudWatch, or other tools which I have researched for RFPs, I feel DataDog is good enough. However, based on the experience from my users, Dynatrace is more flexible for them.

We deal with APM solutions and monitoring or logging solutions by having Splunk already in place in my environment, along with Dynatrace. We have other tools such as CloudWatch.

I have dealt with Splunk on-call and have all kinds of experience. I have used the PeerSpot platform extensively, and it does help significantly. I will be happy to provide individual product reviews.

On a scale of 1-10, I rate Dynatrace an 8 out of 10.

Neutral

Positive