What is our primary use case?
It is one of the core enterprise tools that we use for detecting incidents and paging out to teams when a problem happens. Ideally, it is proactive. This is what we are looking for and what you would expect with Dynatrace: To be looking for things that are going south and getting people on the phone to deal with them beforehand. With a large enterprise, where we are, there are a lot of different teams on the phones, so Dynatrace is hopefully giving us where in the large ecosystem of the application the problem actually is. Therefore, basic use cases, but this is what Dynatrace is good for.
How has it helped my organization?
We do have the pockets today where teams have started to get better at giving us a cohesive view. We use AppMon a lot to fuel the metrics that we have. Therefore, we are starting to create dashboards on top of AppMon that pull together information from a few sources and actually provide that cohesive view. We are just now starting to proof of concept it with some of the business teams to see if we can get traction and start to fill in some of the gaps that we have.
What is most valuable?
PurePath is a very staple thing for it, because within one transaction all the way through you can see the bits and pieces from when the user first came in to the database. We have mainframe components and a lot of middleware layers as well. To have one place where you can see the entire flow all the way up and back is invaluable and it saves lots of time.
What needs improvement?
It looks like they are actually fixing the issues in version 7. Therefore, I am real excited to get it in, because the core problems that we are having, the newest version seems to be fixing. If we can get out of actually having to handle every problem, it can let teams start to get steam on their own.
For how long have I used the solution?
One to three years.
What do I think about the stability of the solution?
The stability has been wonderful for the couple of years that I have been involved with it. It has really been the key tool of choice that anytime anything goes down anywhere, it is typically the tool that is shown on the big screen in front of everybody or shared through a Skype session. It is getting to the point that the CTO of the organization knows the tool by first name and will look to have it brought up, because it is so consistent, reliable, and a cornerstone for solving problems.
What do I think about the scalability of the solution?
This one I would defer to my team a little bit, but I do hear a few gripes scalability-wise. It is a very good tool, and we have got it to do what we need to do. However, I do know that for the size of our organization, we're talking thousands of agents and hundreds of applications, it does get to the point where the servers themselves that house Dynatrace are at a point where, in some cases, they are just too big for one machine, since you have to have an entire application ecosystem all funnel into a single system. One of the things we run into is, when we deploy agents, it is a bit finicky sometimes about how that happens. We have had to put in measures to make sure that applications do not get an upgrade for Dynatrace until we specifically quarter them off, making sure that we have a very careful process to troubleshoot them because we have had several instances where applications have had issues after an upgrade.
How is customer service and technical support?
I have a very good team who are very good at this, so from my perspective, I would say is Dynatrace is one of the few companies where it seems their people are still ahead of my team in terms of troubleshooting things. Some of the other applications that I have, I feel like my guys should be paid by the companies that troubleshoot. The Dynatrace support team that we have is great and the staff that we have had onsite has been consistently good.
Which solutions did we use previously?
My company has used siloed monitoring tools. We do not actually try to get rid of those best of breed tools, but there are some obvious problems with them. Today, a network tool does not use the same terminology as a database tool, and the database guys do not talk the same language as the app teams. Thus, very frequently there are very silent organization or we will have huge gaps between these things. Not only do you have a conversational barrier between teams, you will frequently have whole sections of the network that are not monitored or people who think, "Well that's not my side, the network is good," and the database guy will say, "My side is good and the database is fine." However, there is obviously something in the middle that is not there. That siloing is very damaging to working on a big team trying to fix things quickly.
Before Dynatrace, it was a smaller list of niche tools. Dynatrace was the first tool that started to slice horizontally through all the different silos and provide feedback.
How was the initial setup?
I was not at the company three years ago for the initial setup.
What other advice do I have?
I would wholeheartedly recommend the platform as it is, but looking for someone who is just getting into it and does not have a lot of experience, Dynatrace 7 seems to be easier to get into than AppMon was before. So, it is a great starting point today.
If I had just one solution which could provide real answers, not just data, the immediate benefit to my team would be two things, and they are both on the business end of what we do. Today, our business customers are very frustrated with the ecosystem that we have in place because there are so many complex components. They really want a solution where they can see what is the actual impact for the people who are trying to use the application. If I am trying to go in and check a balance or trying to buy something, they really want to know how many people are being impacted by that today. Then, what actually is the technical problem behind the scenes, but so often, we have a lot of technical problems, but we have a really hard time prioritizing what those are as a cohesive solution. If our business customers could say, "80,000 people are impacted by incident A, but only 200 people are affected by incident B." This would provide an entire view that would be so much better for trying to prioritize development teams to fix problems, and we don't have this typically today.
We are in AppMon 6.5 today. We have people on my team who are sort of a tiger team that have to get involved whenever there is a performance problem because there is almost an art form to using AppMon today. What I have seen so far of Dynatrace and the OneAgent today, it removes a lot of the AppMon art form. I see a lot of value in moving to 7.1 later this year. I am very excited to see when some of our teams, who are not as familiar with Dynatrace but know the application, can start using the application more. Hopefully, it will reduce and back off the need to constantly bring in my Tier 3 team as super experts and help and to maybe focus more on key problems, letting teams deal with things themselves.
Most important criteria when selecting a vendor: Having a company that has been around for a while and has multiple products that we can leverage cost of scale.
Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
Feb 19 2018