As a boy I remember my mother and her brother, both in their forties at the time, torturing my grandmother with inquiries about which one of the two she loved more. Her refusal to answer the question was as predictable as their attempts. She insisted, correctly, that each one was unique, and that she loved both of them equally. I was my grandmother’s favorite grandchild, so I will use her wise approach to compare Dynatrace APM and Dynatrace SAAS.
Recently we, at GQP, LLC (http://www.gqperatners.com) guided a good client of ours, a major university, in their taking a first look at Dynatrace SAAS. We had already introduced Dynatrace APM to the organization about a year and a half prior to that, and they were happy and enthusiastic users of Dynatrace APM.
If that was the case, and they indeed were so happy, why look any further?
Good question. I’ll try to answer it below.
Dynatrace APM is the “Cadillac” of the application performance monitoring solutions. It collects, stores and analyzes every data point relevant to performance of an application. The granularity and the size of the collected data is limited only by hardware resources allocated for the deployed APM solution. Dynatrace APM allows end users to drill down to a line of code, to slice and dice memory dumps, to integrate seamlessly with load testing tools and much, much more.
But there is a price to pay for all this power. While there is a web browser gui capability, for real, deep analysis one would need to use the tool’s thick client. This User interface could be at times both less than intuitive, and quite overwhelming.
In the case of this customer its users congregated at two poles of expertise while working with the tool. A minority of heavy users, who just loved the tool and what they could do with it, and a majority of users who learned to productively use it for a few quick and vital checks. The first group, consisting of performance experts and architects, would use the tool to diagnose performance problems, to conduct root cause analysis, and detailed review and validation of the application architecture. These users accumulated serious deep expertise in how to use the tool.
The second group would also use the tool regularly, but much more lightly. If a question would arise, they would need to talk to experts. Thus the organization limited its use of the tool to development support of a major application and its deployment and operation pre-production environments.
When the question arose about a broader use of an APM solution across multiple environments and applications, together with the client we, decided to take a look at Dynatrace SAAS, which since the winter of 2017, had become the flagship product of Dynatrace.
We are happy that we took this step.
Dynatrace SAAS is a revolution. Immediately the client fell in love with the product’s ease of use — everything is accessible and reachable via web, with Smartscape — a visual model that automatically creates an interactive map of the environment, and last, but not least, with an AI powered way of processing thousands of data points to detect and analyze problems. In addition, One Agent installation proved to be much easier and faster than agent installation in Dynatrace APM.
Also, there are clear advantages of using a SAAS model, as it saves a lot of time on procurement and maintenance of hardware on premise.
Old Dynatrace observers would quickly recognize Ruxit at the core of the new Dynatrace SAAS, but the new incarnation has significant improvements and additions.
While Dynatrace SAAS is a cloud native application it can be used to monitor both cloud and on premise applications. PurePath support (ability to explore and trace transactions in a detailed manner) was recently added to the tool.
In cases where organizations are not yet comfortable with running a SAAS based application monitoring solution, there is Dynatrace Managed version, that is an ability to deploy the same solution with identical functionality on premise.
While the product makes excellent first impressions, there are some features that we have become used to in Dynatrace APM, that are missing in the current version of Dynatrace SAAS. They are – the ability to conduct and analyze memory dumps, and a way to integrate Dynatrace SAAS with third party load testing tools like NeoLoad or JMeter. Our friends at Dynatrace assure us that these features are coming.
All in all, Dynatrace SAAS is an amazing tour de force that provides a concrete example of the power of AI used to solve immediate, practical problems. Dynatrace’s AI driven ability to analyze problems democratizes performance analysis and makes it accessible to a much broader group of users than before. Experts are free now to focus on more difficult problems. As we all know, there is no shortage of those. Among which, is the need for organizations to learn how all this newly acquired power, can enable them to fix problems faster, and to deploy better software.
So which tool is better? It really depends on your tasks. As my grandmother used to say about her children — both are unique and both are very worthy of love.
Incidentally, I never asked her whether I was her favorite grandchild. I just knew it.