What is our primary use case?
To look at the full transaction flow, from the customer all the way through back-end systems, for problem analysis, triage, performance optimization.
It does it really well. The more visibility you have through the agent implementations, the deeper you can get down through the stack and identify what you're looking for and how to fix it, find it, tune it.
How has it helped my organization?
The Managed SaaS is going to simplify things a lot and allow us to consolidate a lot of tool sets. We are going to be able to get rid of some of the old Legacy monitoring tools and replace them with something much better that puts everything in one place.
What is most valuable?
The stuff that's coming with the new pieces around the Dynatrace Managed SaaS implementation. The ease of implementation there is significant. We've spent a lot of time with AppMon and DC RUM - that's a lot of time to set up, configure. With Managed solution, you just drop it in and everything pretty much auto-instruments, which makes life a lot easier.
What needs improvement?
One of the things that I was talking about with Simon earlier was getting mobile native replay. They don't have a timeline on it yet, but that's one of the key things we're looking at, to get rid of one of our incumbent products that does replay.
Also, extended support for some of the agents, the one-agent technologies under Managed. We've got some old legacy platforms that don't have one-agent support yet.
For how long have I used the solution?
One to three years.
What do I think about the stability of the solution?
The Managed solution is much more stable. We've been running that environment for about six months now and it hasn't had hiccup yet. We had problems with AppMon before. We've taken AppMon beyond it's stated capabilities. We are running 2,400 agents off of a single server in a single system profile, which is well beyond the stated capacity. I think the stated agent cutoff limits are 1,200. We passed that like a year ago. We measure the internal workings of AppMon to make sure it's not going to fall over, but we hit those thresholds almost daily now. So we are throwing data out just because it's over capacity.
What do I think about the scalability of the solution?
Managed, we can scale it out massively. AppMon doesn't scale well. That's the other piece, why we're looking at Managed, because we're already over capacity there. We've got to move stuff to the other platform, to Managed, so that we can get the horizontal and the vertical scale ability, and get away from our "problem child" on the other side.
How is customer service and technical support?
We use tech support. How I would rate them depends. So some of our problems are quite complex, because of how we've stressed the system to the extent that we have. So, we've had some stuff on the AppMon side that's because we're well beyond the usable limits. They help where they can, but in some instances it just is what it is.
On the Managed side it's been pretty good. Stuff gets turned around pretty quickly. With them being able to do the remote management from the back-end, they are able to fix stuff up if they need to.
In terms of response time, relative to the complexity of the problem, I think it's reasonable. Some of our problems have been not normal. But the normal stuff, they turn it around fairly quickly. We don't pay for Premium support either, so you get what you pay for. They are usually quite responsive, and we've got really good connections into most of the folks back in the labs, so it works out well.
Which solutions did we use previously?
With the company I'm consulting with now, they had Dynatrace before there was a desire internally to look at some AppDynamics stuff. They did a bit of a bake-off and decided that it wasn't the right way to go just because of the capabilities and features. So it's always been on that side of the Dynatrace stuff.
But I've used New Relic and AppD and the IBM Application Performance Monitoring solutions, so I've seen lots of them, and Dynatrace is definitely the better of them all, by a long shot.
How was the initial setup?
Set it up myself. It took 20 minutes. Put a piece of hardware in, run two scripts, done.
We ran into a bit of a technical issue where we had to engage the support guys. They identified the issue, fixed the scripts, and then the people after us didn't have to deal with it. I think we were one of the first Managed on-prem implementations; maybe, not the first, but one of the few. So we weren't doing a normal implantation I guess, so little hiccups, but they were quickly resolved.
What other advice do I have?
I think is role of AI when it comes to IT's ability to scale in the cloud and manage performance problems is a big game changer. That's one of the reasons that we're looking at moving a lot of our stuff from the legacy AppMon over to the Managed platforms, so we can take advantage of that and get rid of some of our more archaic event management, event lifecycle, alerting-type platforms. A lot of that stuff doesn't add the value that it should. So we are looking at the AI engine and the anomaly detection to basically replace a lot of that manual effort.
These archaic solutions are all siloed monitoring tools. That's one of the things I presented on here at the Perform 2018 conference yesterday, about all the data silos across all the old platforms, and being able to pull them together into Dynatrace so you can get that single pane of glass. The siloed solutions all had their own purposes, but the data was not inter-relatable. You'd have OS monitoring tools, and even AppMon, DC RUM, Synthetics; they all have great data, but they're not tied together.
The immediate benefit of just one solution that could provide real answers, and not just data, would be you could look at it in one spot. Even one of our groups that came to us a couple weeks ago said, "Oh yeah, we are going to do a new native mobile app and we're going to use this piece of freeware from Google and this piece of freeware from there." And I said, "Okay, so how are you going to pull that all together?" And they said, "Well, you can look here, and then look it up there..." And when I said, "Why? We already have a solution that does that," they said, "But, they're free tools", and I said, "Yeah, free necessarily isn't always the best option."
In terms of vendor selection, I think one of the key things with Dynatrace is that they are very open to influence on product development side. So, we've influenced them fairly heavily on development and capabilities for Citrix and DC RUM. They've given us integration and support components around some odd technologies that we've got, and they have always been very open and accommodating to going after and developing capabilities around the stuff that we are looking for, which has been good.
I rate it an eight out of 10. I don't know if it could ever get to a 10 because there are always going be anomalies and idiosyncrasies that, commercially, it doesn't make sense for them to cater to everything. There is stuff that I'd like it to be able to do but commercially it just doesn't make sense. But, at the same time, they are evolving into things that need to happen as technology advances.
In terms of advice to someone who is research this type of solution, I get pulled on from the Dynatrace Accounts team regularly to do those sorts of conversations. I'm a pretty firm believer in the products and what they can do. I highly recommend them to anyone who is looking at them. I've used the competitive - or non-competitive - products that are out there, so it's pretty clear for me as far as why it's the right choice. I'm happy to have those conversations with them to take them down that path and let them understand the why's and what decisions they should make.
Disclosure: My company has a business relationship with this vendor other than being a customer: Partner.
Feb 08 2018