The modules that we are finding most valuable are as follows:
Applications: This provides us insight into how our applications are performing within our environments and affords us the ability to identify opportunities and make changes to code / environment to effect positive performance lift.
Databases: This module is amazing; allows our developers the ability to view into the SQL nodes themselves to quickly rule in/out any hardware issues. It also provides them with access to download execution plans directly via AppDynamics, which in turn helps improve turnaround time for fixes and such.
End User Monitoring: Whereas the Application module gives us insight as to how our applications are performing within our environments, End User Monitoring provides us similar data from a browser perspective. In other words, it provides us the complete experience from the clients’ perspective; tremendously valuable indeed.
Improvements to My Organization:
The totality of the product has allowed us to decrease our mean time to resolution when issues are detected. Also, because we’ve implemented this in our test dev environments, we now use it to measure and quantify performance improvements prior to ever releasing to production.
Room for Improvement:
The product has a pretty diverse metrics browser, but I’d like to see better out-of-the-box visual reporting so that we can roll this up to management. I have no doubt that will be improved in the next release.
Use of Solution:
I have used it for one year.
The deployment is as easy as could be… simple and straightforward.
The product has been rock solid from a stability standpoint.
Scalability is easily controlled via scripting, a REST API and included utilities for copying configurations between applications.
Technical support is 10/10. They are very responsive to support requests.
The initial deployment of .NET and database agents was done in minimal time. We had our first application set up in a matter of minutes.
We chose to deploy in-house and that turned out to be the correct decision for us, as it allowed us to become more familiar with the behind-the-scenes workings of the software. As far as advice, while auto-discovery is nice, we’ve found that prior planning in regards to configuration of applications, tiers and nodes pays off by ensuring the returned data is in a format that is most digestible to your organization.
We are still in the process of completing our ROI calculations, but we already have examples of a 89% reduction in MTTR for one application... and a multitude of examples of money saved in others, including savings of around $150,000 for one particular issue that we were almost set to throw more hardware / software at.
Other Solutions Considered:
We evaluated several other products in addition to AppDynamics (Ruxit, New Relic & Dynatrace). While each possesses some great tools, none provided the totality or completeness of toolset that we required for our complex environments.
Take the time to plan what exactly it is you are looking to get out of the software; it will pay dividends after implementation.
Disclosure: I am a real user, and this review is based on my own experience and opinions.