Dynatrace Reviews

The primary use case is application monitoring. We are using APM to test for performance, bugs, and hoping to resolve the issues faster, and hoping to catch them before we go to production.

It's been great, it has helped us a lot. It can do more, but it's definitely helped us a lot and I'm a big believer in Dynatrace products.

First of all it's mitigating issues before they are in production, and if they do go into production, it's reducing the time to find the issue and actually resolve it. I believe, in the organization that we're in right now, that is challenged for resources and time, a product like Dynatrace helps immensely.

The PurePaths, because that's where somebody who is a non-developer can figure out where the problem is and send appropriate PurePaths, clean charts, or even the link to the developer. The developer can then look at it and figure out exactly where the problem is, this is the piece of code that took the longest time, and then resolve it.

Right now, since I'm primarily an AppMon user, so maybe the Dynatrace product addresses this: The challenge with AppMon is, what if you don't have an AppMon agent on a host, but it talks to the database. It talks to it, but I don't have either a host agent or an AppMon agent on it. That has been a challenge, but I believe the Dynatrace agent, the OneAgent, will solve that, potentially. You ask me three months from now, after we take a crack at the Dynatrace product, maybe my answer will be different, but I'm hoping that addresses some of the issues.

The configuration of the alerts, that's been a challenge in AppMon for me, right now. Some of the alerts are too noisy, but that might be my lack of some configuration. Again, it's just me primarily handling it, so that could be an issue. Somebody asked a question in one of the sessions, here at the Perform 2018 conference, about noise and how many alerts to your problem count, and the person doing the session answered right away saying, "I checked my dashboard before I came to this session and I had one alert on it." So I'm guessing that will resolve itself.

I think I like the direction it's going in, the only challenging part for me is to keep up with the name changes. But other than that, as far as stability, I think I'm happy with it.

I think for the deployments we have right now, it has not been a challenge. My goal is to increase the usage throughout the organization, maybe that's where I'll face some challenges, but at this point there are no challenges.

I have used technical support in the past and they are pretty quick to respond. The other thing is, the APM community is available, Andy answers pretty much any question I post pretty quickly, so I think that group community help is really good.

I have use siloed monitoring tools in the past. When I started at Nemours 17 years ago, I had custom scripts that I would use to apply to various servers. They were on the host level, but the deployment was challenging. How to tie in a CPU alert to application slowness is challenging, because you had to go to the timestamp, look at the log and say, "Okay, this might be the issue." Dynatrace tells you how it is, and I think that's the most important feature.

I'm the primary Dynatrace admin, if you want to call me that, and it was pretty easy. But keep in mind, my skill sets are probably unique in the sense that I understand applications well, so I know how to insert the agent - because we use AppMon - how we insert agent into JVM.

But I believe the new Dynatrace product is probably the way to go, because you don't need to actually talk to the application folks, you just deploy it on the host and you're done. I believe it's definitely going in the right direction. It is complex, it wasn't for me, but I can imagine it being complex for some people.

I don't know if there were any other vendors on my list because we've been users for about 15 years. We started out with a Vantage product that moved to server monitoring, and then we had the Gomez platform, and then you also had Dynatrace, but then they all came under the same umbrella. So we never really evaluated any other vendor. We had some of the free tools we used to use, like  Profiler, but from what I've heard from developers, nothing ever came close to this so I'm a fan.

If we had just one solution that could provide real answers, as opposed to just data, we could spend less time on troubleshooting and trying to figure out what the problem is, and actually do our jobs, which is to design, build, and develop software.

The criteria we look for when adopting an APM solution are ease of use, and does it truly get you down to the problem area - and I believe that Dynatrace does - and the third one, it's true for everyone, is the cost.

I would rate it a nine out of 10. I'm not giving it a 10 yet because I would like to see the Dynatrace product in action and truly want to understand it. If we move to Dynatrace, away from AppMon, are we missing out on something?

My advice would be go with Dynatrace.

I mostly just onboard different applications in the company under the Dynatrace platform. Occasionally, they will have issues, then I use AppMon in order to tell them what the issue is. It usually is something simple: The URL, this particular service is slow, or your database is not responding correctly. 

It is performing well.

It is different for me than other users. I like the PurePaths dashlet the most. This is mostly because (and I can count a handful at times where this has not been this scenario) as soon I open the PurePaths dashlet and sort by response time, there is the problem. Every time.

If the PurePaths dashlet pulls up 750,000 PurePaths, I really only needed to know about seven or eight of them. Then, being able to look into the code-level dive about it, that is just a sanity check. Just to make sure that it is the same issue multiple times, not a random anomaly where everything else was crap.

Also Errors dashlet, I use that a ton.

I would love to see a better data export, because AppMon's charting capabilities leaves a lot to be desired. The dashboarding capabilities leaves a lot to be desired. There are a lot of times, for example, at my last company, they wanted Dynatrace data in addition to a bunch of other stuff dumped into one place. It was not just performance metrics. The CEO wanted his business metrics in the same place as the performance metrics along with a lot of other stuff. However, Dynatrace could not export this type of stuff. 

You have about a 5,000 line limit or you have to set up your CSV file just exactly. DC RUM can do it. It might take like an hour sometimes, but DC RUM can do it. I would really like to see the ability to export, in Dynatrace and AppMon, in essentially in a nice format of whatever you want to whatever else. That would be fantastic.

For AppMon, there is always room for improvement: charting, dashboarding, and user management. However, that is pretty much our fault with LDAP. The onboard process itself is a pain, even though we have scripted so much it, it is just very repetitive. There is a lot of alerts and things like that out-of-the-box that do not need to be there or that just do not do the right things.

For Dynatrace, I feel like it just needs a lot more technology support. I know they are trying to essentially get rid of AppMon and move toward the Dynatrace way of doing things. However, we are a multibillion dollar bank. We are not up-to-date. We are not going to be microservices for a long time. We are not going to be container for a long time, and we are probably one the most expensive clients that they have.

We are the ones who are going to drive a lot of the money factor so they need to have that. They need to have integration between the current set of tools so we have the ability to onboard five or six apps, then we'll also put the AppMon agent on it and show people the difference between it. It needs to be better integrated.

All of our team will go to a five minute sales meeting, if they were like, "Look, you can do this with a script." We are sold.

We do not want to do any of the regular AppMon stuff. However, when you have to convince the CTO that we are going to completely rip out the entire monitoring solution which we just spent the last 15 years trying to get a process set up for, and now we are going to redo it. That is not going to go over well. That is not a good conversation. 

You need to have that ability to do MQ. I don't care who uses MQ, but apparently we do. If you can't look into those messages, then you just lost a half of our organization which can't be monitored with it.

With stability, I have never run into issues with Dynatrace causing an issue yet. 

I have run into AppMon causing issues. There have been a lot times when I have waited for a release. AppMon does the release, then it ends up taking down our application. Now, the fix is immediate, but I have already loss face. 

The last time, I took down the main application that lets you call tow trucks. It was just a monitoring loop, a simple thing. They fixed it in a patch. They knew about the issue and they told me immediately what the issue was. I got it fixed in 15 minutes. It took me six months to convince the team to install it in the first place and took me another seven for them to give me another shot at it. It was not a problem that showed up in QA, for whatever reason. So, I could not convince them that it does not exist anymore, because I could not show them any evidence that it existed in the first place. Let alone that I fixed it.

I have never been more confident in a tool because I've seen how easy it is for it to deal with the .NET Applets. That is a big problem in AppMon. Unless every single person is naming their Applets, the exact same way and following the exact same pattern, it becomes an issue. The new tool does not run into that at all. Similarly, you can script it so it just automatically blasts across the organization. As long as it has the PurePath capabilities, somebody who is running, for example, the actual web application that tells people their accounts, that might be a different, more in-depth use case for AppMon versus Dynatrace. So far, the scalability of the solution is phenomenal. 

Anytime I can't find the answer immediately in docs or answers, I just open a ticket. They are very good about giving you a response. 

Initially, I am talking about two or three years ago, I think they did not have enough personnel staffed there. Therefore, it would take them maybe two or three days to get to your ticket. Now, it is maybe the next day you will have a pretty reasonable answer and that is provided you did exactly what it says in the support ticket. For example, make sure you upload your support archive. Otherwise, you will burn a day and they will send you an email requesting you just upload this. That is shooting yourself in the foot, and that is not their fault. 

I have previously used siloed monitoring tools. I have used SiteScope. My current company uses ITCAM. They are okay. They get the job done.

At my previous employment, we used SiteScope and that was quite literally the way that I thought about it in day-to-day life, if you do not really give a crap about it, just put SiteScope on it. However, if you actually need to know if it is working, it needs to have Dynatrace.

I was always pushing for that sort of thing. There is stuff like Wiley where you are not getting 100% monitoring. There is another tool, one is a very new company, and it seemed to get the job done but that was only because we were using Citrix Xenapp. It was specifically able to decode the traffic for Xenapp and XenDesktop, which was what we were looking at. Apart from that, I have never had a situation where I was like, maybe we should not put Dynatrace on this. I have never run into that. 

I was not involved in the initial setup at my current company, but I was at a previous company.

For upgrading at my current company, that is in process. We are trying to figure out if it is better to blast it across the organization. We have five Dynatrace servers. They are all completely at capacity. They are all set at large. It is a really big deal for us to try to switch anything over. Right now, we are trying to figure out, do we just upgrade our collectors and hope for the best, or do we do it in QA and then in production? 

A lot of people do not run the right thing in QA. It is never the case that their QA is identical to production. So, is that a good indication? I have run into issues before when upgrading from 5.6 to 6.1 expecting that all the bugs were ironed out. That is when I took down that application. Now, I do not have confidence in this upgrade process.

For the Dynatrace Managed version, that setup process was incredibly easy. It took 15 minutes. 

Just go with Dynatrace. Just start with Dynatrace. Do not go into AppMon. Start with Dynatrace, because AppMon is going to give you so much extra stuff that 99% of your user base will not need it, including yourself.

You don't need AppMon. I am a hardcore AppMon guy and I am still saying this. It is a lot nicer to be able to start with the Dynatrace solution, be able to script everything, and start integrating the new thing than it is to try to do the old tool set. 

I started in the PDP program at Dynatrace. That was when they were still Compuware. Then they became Dynatrace, and I went to a different company, now I am at PNC. I have done the exact same thing for several years in different places.

In my current and previous positions, AI is not important when it comes to IT's ability to scale in the cloud and manage performance problems.

The previous company bought it, and they did not even set it up. I onboarded a bunch of apps. So, they were way too fledgling to try to start looking into it. 

What I would use AI for is if it could assist me in saying, "These are your common PurePath patterns." A lot of times in the ending part of an URL, they will have /apps, /data/, then they will put something lie the date or some big custom code. For example, we had one application for tow trucks, which would tell them the URL contained in PurePath, for the actual seven decimal place of the geographic coordinates of that tow truck.

This is not a good way to look at data. If the AI could tell me something about it, just mask it out, or just know this is the same type of data as these other ones and not worry about the extra text part of the piece. That would be the foremost use case for me. After that, I am not sure. 

I would need to use it a lot more to maintain my own trust factor in it before I would want to try to tell somebody that is asking me what the problem is. Just immediately saying AI says this. I do not have a high confidence enough factor in it, because I have never really used it.

If my organization had just one solution that could provide real answers, not just data, it would probably put me out of a job. Most of the time, when I get a ticket, they will ask me what the problem is. I will point out the problem, and it is something which is you need to code this better or you messed up these settings. Therefore, as far as helping me not have those mundane sort of tickets where I don't really want to waste my time with people. It is fine for the first few, but after the thirtieth or fortieth person, you tell them that you wrote this very poorly. It is better to just have some tool tell them that this is probably not the best way to do this.

That would be the initial benefit of the one solution. A part from that though, all I am doing is onboarding. The new Dynatrace already takes care of this. So, I am not really sure what my role would be afterwards. Right now, the APM is siloed off from the development teams. If you are going the full Dynatrace route with AI and getting the opportunity of the AIs already going to tell them what the issue is. Then, the APM team does not really need to exist anymore, apart from doing migrations.

Most important criteria when working with a vendor: That initial pairing of sales versus FTS. If I could reach out to them and get answers within a day, or better yet, within an hour. That is one of the best things because a lot of times that initial conversation can get derailed so quickly. You are not going to get more than five or ten minutes to pitch it to your boss. They are always at meetings. For example, my boss, at my previous place, I would be able to sit with him and talk to him about this thing. Then he would get, maybe, five minutes a week of his bosses level. That is the person who is going to sign the paycheck. 

Therefore, when he goes to a meeting, and it is a week later, he gives the spiel and has it all ironed out. Then, his boss asks him, "What about this?" Now, he does not know the answer, and I can't get the answer, then I need to get somebody on the phone stat to give him an answer. Otherwise, we have to wait another week. That is a big deal for us to have that communication open. 

My primary use case is troubleshooting production issues to see which endpoints are mostly calling and get some logs about them.

It's performing well. We use AppMon, and whenever any production issues occurs, we get a time window from the customer explaining, "This time it is a software failure."

We actually did some training on the AppMon UI, then it almost never failed on us. We could trace back exactly where the issue occurred. We came to the Dynatrace event to understand more about it, and see how we could learn more to perform our troubleshooting better.

It used to take a lot of time to troubleshoot. Now, with Dynatrace, we can actually see the logs anytime we want. I can just find the problem. It has improved performance from a time perspective.

The most valuable feature the solution offers right now is the PurePath. When we see a web request, and something failing, we can drill down using PurePath. Then, on PurePath, we can get to the database SQL statement. So, it is a very cool thing. We have our logs and all the stuff, but this is an area that stands out.

The AI thing is a boost, but it is not in AppMon. It is in Dynatrace. If AppMon could also incorporate it, that would be great.

The main thing is more about the dashboard. Currently, it does not keep the snapshot data. It only keeps it for a very short duration. Because of that reason, we cannot get more reports. If we can somehow option a way to preserve the data and keep it for a longer time, or have that feature, that would be good.

It is always requiring us to update the Dynatrace client. There are some issues with Dynatrace. Many times I thought there was a production night deployment and we would do a smoke test. Therefore, I started Dynatrace, and suddenly it says your client needs to be updated. Then I had to go do that two or three times. So, there are some glitches in it.

In the future, Dynatrace should actually try to resolve those issues with AppMon. The new Dynatrace version, I think will not have these issues, because it will be only browser based. So, there will be no need to install anything on our machines.

I sometimes see issues with the loading of the data. If we give you criteria, like you want to see the last 15 days of data, it has a lot of requests in it. To refresh, it will take a lot of time. Therefore, we actually narrow it down to a particular instance or event.

My main purpose is to troubleshooting issues, so this way I know exactly what time it happened, then I can just narrow it down to that. However, if you want to see a month's data, it keeps on spinning. Here is an improvement which needs to happen, which is the case with all applications or tools. There is a lot of data, and either we have to change the way we are logging or the application needs to be enhanced.

For some reason, we wanted to create a report about what is the maximum used end point or the maximum called end point. Also, we wanted to create a certain dashboard at that time. Myself, with a DevOps engineer, we called the technical support team, and they helped us create that graph. We got it, actually.

We have used production logging. We have our own custom logging system, where we read the logs and transfer to the database. We have a nice UI for it too, but it does not scale well. It can read the log, but it can read only those logs which we added.

Most of the time it works, but it is very slow. If we want to trace back, for example, to find out what happened in the previous week, we cannot get that data.

The DevOps engineer worked extensively on it. I do the troubleshooting.

It saves a lot of time.

It is the best solution that I have seen so far.

AI is like a new feature or benefits, and it is a cool thing. We have not tried it, but I really like to see it working. It is a great program, and it automatically makes a trendline baseline. Whenever something goes wrong, it can send an alert. There is also a web check feature.

Currently, we do not have a baseline. If AI is there, it can see the trend. Based on the trend, it can send notifications. It is also integrated with various platforms, like social platforms, so we can also get alerts from there.

Most important criteria when selecting an APM solution: 

We look mainly for how it will scale and what are the features currently available. 

I provide this information to the decision-makers.

We use Dynatrace to detect where we can improve performance. We use AppMon to do that. We're just getting started, in the past six months is when we started using Dynatrace, so we don't have definitive results yet, but we have definitely identified problem points that we can work on.

It's given us a lot better insight into how our application performs.

The fact that it can really analyze what parts of the application are slower than others.

I like that the new version has aggregated Waterfall, but I'm told that it's not gonna come to AppMon. I would like that to happen to AppMon.

I would give AppMon an eight. And the reason I don't rate it higher is that the learning curve on the tool is... You have to have some experience with it to be able to figure out where to find things, and the best way to get to things. But once you know that, it's a really good tool.

I would recommend Dynatrace. I don't have any problem recommending it.

I haven't had any problems with it working.

We have a pretty small environment, so we don't have scalability problems.

We had training from them and they're all been knowledgeable. I'm happy with the training that they provided. Their support people are responsive.

We really didn't have anything. We went to an APM because we had a concerted effort to improve application performance. Dynatrace seemed like it was the best solution for the problem. It covered what we wanted and more.

It was complex, but well-documented. Getting AppMon set up has a lot of moving parts, but they're all documented and it went relatively smoothly.

There was SOASTA and one other.

I don't have any personal experience with AI in IT, but I can see that, as environments get more complex, it will definitely help with finding problems.

If we had just one solution that could provide real answers, and not just data, on our ops side it would help because they use that to dig out the root cause of problems.

When we look at a vendor we're mainly looking to see that their product does what we want it to do, that we can analyze performance problems and that it gives us good feedback on where we can improve things.

We are old school, we don't have proper documentation on how our application interacts with the downstream components. We tried CA Wily, that didn't provide a solution. We had CA Wily for three years, but when we deployed AppMon it gave us the complete picture, how the application interacts with downstream components, the application flow.

Based on that, my development team is developing the application automation diagram using the app map. It's like reverse engineering I guess.

It's really helping out. The term we call the mean time to repair, that has improved a lot. Before we had a "command center" type of setup. We had all the dashboards on the screen and five to six people sitting in front of the computer. When there was an issue they used to call a "bridge" meeting and everybody would login to the bridge.

The AppMon solution helped the OCC, the operations guys, to pinpoint one problem area and to call the specific group, instead of opening the bridge for everyone and asking them to join there.

The mean time to repair and the resource utilization time, they are totally reduced deploying APM.

The Transaction Flow diagram and the class and meta level information, those are really key selling points in the automation for AppMon. Also, the meta level instrumentation and the dashboards that most of the people use in our organization. 

Most importantly the back-end components. Most of the back-end components that the application connects to; nobody knows how our application interacts with, for example, the DataPower Gateway. But AppMon really provides that information for us. So finding the gaps is the key here.

One thing I'm missing and that is the JMX from the MBeans, it's missing completely. 

Also the reporting. We have a load testing team, they completely rely on the reporting for analyzing the data. They should have a template to create a report and they should have something to auto-deliver the report into your email box. 

They also need to develop how to capture the JDBC and MBeans metrics. That is something they're lacking. Also integration for the extension to DataPower and MuleSoft Gateway.

The other feature that Dynatrace should have is - from what I see in Dynatrace in our PoC - when you auto-upgrade the agents, the JVM or the application has to be restarted. But if you have something like an "auto-attach" feature, to attach the agent for the running process, it would not require a JVM restart. That would be nicer. That is a killer point.

We haven't seen it that much but it's promising. We haven't had any downtime.

We have a small portion deployed right now, but they're planning to go enterprise-wide, we're thinking about 8,000 to 10,000 agents, so probably at that time we'll know the scalability. But the small one we have right now, it's good. It's doing well. The UI is faster, people like it.

We used tech support one time. We did in Fabric. We have internal cloud, called C3. So C3 and Fabric, we had a little bit of an issue, we called the support guys to fix it so and they were knowledgeable.

I have used HPE Diagnostics, CA Wily. I'm now doing a PoC with Dynatrace and AppDynamics.

With CA Wily, the problem was the complexity of our environment. It's not the tool, we have too many security layers. CA Wily, there's a breaking point. When we have the agents on multiple servers, it's not creating that link because of the security. But Dynatrace, I don't know how they did it, it's really awesome.

We switched because of the Transaction Flow diagram. There is a complexity in CA Wily that it is not able to integrate into our security layer, whereas AppMon can. That's the main thing that's missing. We were looking for that flow diagram. Nobody knows it. The developers call it but they really don't know what the main functionality is that's behind it: How the Apache layer connects to application server, and the application server connects to the different services, and the services to the back. Nobody knows it unless you have proper documentation. Proper documentation is very difficult to get in any organization. 

Upper management really liked the Transaction Flow diagram. No matter what, that's the key.

Comparing AppMon to Dynatrace, I like Dynatrace very much because of the ease of use. AppMon is complexity. You need to know more to use the tool. The adaptability comes when you have a nice GUI that is easy to navigate. I saw that in the Dynatrace SaaS model. 

When I'm looking at AI, the problem identification and anomaly prediction is important. It's good to know, beforehand, when the problem is going to happen. The anomaly detection is the key area, and part of the AI I think.

If we had a solution that gives you an alert that said, "This is your problem, this is how you're going to solve it," that would be really awesome. Pointing out a problem to a specific group is a key point. That would really help, instead of globally alerting everybody, alerting upper management. If before they know it, you can solve the problem that would be nice.

When looking at vendors, we have a key set of requirements. Among them are container health monitoring, flow diagram. Also extension monitoring the non-Java applications, or non-supported applications, because Dynatrace works mostly on the .NET or Java applications. There are applications out there which are non-Java based like PeopleSoft. At least we can see the interaction with those components, but it would be nice to see what's going on inside those external components. That's what I'm looking in future releases, more support for things like PeopleSoft.

I would rate Dynatrace an eight out of 10 compared to other tools. The amount of data it provides is awesome. Other tools work on a sampling methodology but Dynatrace captures all the RAM sessions that are running. It's more data, but they have the filtering options so I can pick the data I want. Capturing all the data gives me more insight into what's going on, I can compare a bad response to a fast response, that type of thing. Capturing all the data is awesome.

I've worked with HPE Diagnostics, I've worked with CA Wily, now I'm doing a PoC with Dynatrace and AppDynamics. Compared to all these products Dynatrace stood out.

Currently, at my organization, we have DC RUM, Gomez synthetics, and AppMon. We are getting ready to PoC the OneAgent technology. Our use case for OneAgent will be containers monitoring, AWS monitoring, and microservice monitoring. Our organization currently has a digital services team that will be doing a lot of this cutting edge stuff and our architecture is not being fully monitored by the current stack that we have. We also have a tool called ScienceLogic that we use with our Ops bridge and we have Splunk. Therefore, we are really looking for the OneAgent to fill in the gaps that we are not covering. 

Out of the gate, I can tell you that just changing the agent model from having to go, "This is a Java, this is a .NET, this is a web, or this is an app," And getting rid of that to having the OneAgent is an improvement. Then, being able to monitor anything, literally anything on the infrastructure, that will be huge for us. 

Another thing is the log analytics. We have to work with the Splunk team. We have a very fractured organization, so anytime an application team wants log monitoring, they have to work with the Splunk team. Anytime we want log monitoring we have to work with Splunk team, it is a paperwork process. It is not that they are not nice people, it is just that it takes a really long time, and by then, the problem may be gone.

We really just need the immediate log analytics, because we might not need long-term analytics all the time. Like Splunk, they do awesome data analytics, but sometimes when we are troubleshooting an issue, we just need to look at where the problem is.

I would like better plugin support, because they are constantly asking us to do plugins, saying "Yeah, we can do that. Use this plugin."

Then, the moment something goes wrong with that plugin, I have no way of getting help. They recommend, "Contact the plugin author." You are kidding me? Those guys do not have any obligation to respond. They wrote it, but they do not have to support it. 

I do have concerns about stability. I have heard there are gaps being mentioned and I think that we will lose some features. I am not entirely trusting that this is fully-baked yet. I am probably going to go back to my organization and say we really need to do a PoC and we need to get on the train. However, I am not stepping down AppMon or our current DC RUM for another year or two, because this really needs to mature.

I am not worried about scalability. OneAgent is going to more scalable than the clunky old solutions that we just put to bed.

I evaluate technical support on a criteria of:

• Did they help solve my problem?
• How long did it take them to help solve my problem?

I have not had very positive experiences in the last year. I found them to be arrogant, rude, not solving my problems, and not interested in solving my problems. 

They are very insulted if you ask them questions. They refer you to read the documentation. I am like, "Well, I already read the documentation and that is why I am asking you a question." 

"Google that," is not the answer that I want hear out of tech support, because I already googled it. So, I have been rating them very low. It is actually the one pain point I did bring up to my sales engineers and sales support. They are aware.

Our organization just dumped HPOM and the plug was pulled on it in 2017. We are getting ready to sunset SiteScope and HPE BSM. The really siloed monitoring tools are archaic and high maintenance. The infrastructure is too big and moving too fast to be constantly updating old-fashioned tools.

What I really am excited about is that they just announced that it will be doing session replay for the DC RUM part of it. We currently have IBM Tealeaf and we hate it. We hate its guts. Our management hates it because we rolled it out to one application and it cost us over a million dollars to do that. We wanted to roll it out to a second application, and IBM wanted another million dollars. We already have the on-prem and a trained administrator. That was the licensing that they wanted that million for.

Dynatrace just said the session replay that DC RUM will have, it will put IBM Tealeaf out of business. Thank God, because that solution literally is from the 90s. I am not kidding you, it was coded in the 90s, and it is extremely brittle. It is hard to maintain, and it is clunky. If they can get that replay to be smoking, then they will make so much money. All the Tealeaf customers will stampede over here.

We stood up AppMon five years ago. It was not straightforward; it was complex. 

We paid for a Dynatrace guardian to be onsite for two years. The guardian rocked. We really loved the guardian. If we had not had the guardian there, we would not have had an adoption rate like we got.

Thursday and Friday last week, I got pulled into what they call a triage team. They had a problem in production where all these people were in a war room with all these fingers being pointed. Nobody had the full picture. I am using AppMon, which performed beautifully and management was super impressed. Just using AppMon 6.5, the visits, being able to drill down, and find the answers there worked. However, if I had had the OneAgent, DC RUM, and SaaS solution altogether, I would not have been in there for two days. It cost our company so much money and they still do not know what the answer is.

This is an expensive solution, but it is also worth the money.

My organization evaluated Nagios and a start-up company. Then we evaluated Dynatrace as well, and it just outperformed the others.

I am really excited about the AI that I am seeing out of the OneAgent because I was just at the IBM conference last year. The IBM AI is still pretty much a toy and I have not really seen the rubber hit the road with their stuff, but the rubber hitting the road is here with the Dynatrace AI. From what I have seen, it will be a key tool set for us just to pin down problems and get answers immediately. 

Most important criteria when selecting a vendor: Everything that they are doing is right, except tech support and plugins.

I'm an admin so my primary use case is to manage the product and the software. And whenever there is a user, they have a question, they don't know how to browse or look at the PurePath, that is where I come in, to show them how to.

It has helped us reduce time spent on root cause analysis. Whenever there's an issue, there are multiple teams involved, and we are using multiple tools: Splunk, Nagios, etc. If we have one product that gives one solution to the root cause, the issue, it reduces some of the operation time for our team, so we can spend most of our time on automation or something else.

The PurePaths. The ability to see the transaction flow of the web request. It's very useful.

It's valuable because our developers, when we have an issue, they can drill down to see exactly where in the application, the call in the application; where the high response time is, or where something is wrong. At least they know where to look for the solution.

I want to see an increase in the load, at least to 7,000 or 8,000 transactions per second, just for it to stable for this upcoming holiday, if we do not transition to the new Dynatrace product yet.

What is missing right now is, it doesn't tell me where the problem is. So whenever we find a problem from the customer, or from our incident management team, then we go into the tools and look for it. So we want it to have more problem identification, somehow, just like the new Dynatrace product. They have the problem tab, you can go in there an see it. But AppMon lacks that.

Right now, for AppMon, the maximum handling load, the transactions per minute, is around 6,500. We had an issue on Black Friday or Cyber Monday, some kind of stability issue, that users who could not log in.

It's not very scalable, we cannot add another server onto it. We have a little trouble there, so that's why we're trying to upgrade to this new Dynatrace product to see how it will go.

We just submit a ticket, and it will probably take them one day. That's sufficient for us, for now.

The resolution time is also about one day. Right now, one day is not good enough. We would like it to be in the hours. So that's why, another thing with this new Dynatrace product is they have the messaging system, something like that. And they have the server that would monitor our Dynatrace server, where they can see everything, what the issue is already, so they can resolve it quicker, instead of our submitting a ticket and waiting for one day.

I was not involved in the set up of the app. I remember they were doing it manually. They installed the agent, set up the server manually, on each server. So when I came in, we used Jenkins to write a script to automatically deploy to our server. There is still a lot of manual configuration. It's not straightforward.

When it comes to the nature of digital complexity, and IT's ability to scale in the cloud and manage performance problems, I think moving to the cloud is very agile and provides scalability. Right now, we're using AppMon which doesn't have the AI feature. That's why we're interested in upgrading to the new Dynatrace product, which has the AI feature and scalability.

If we had just one solution that could provide real answers, as opposed to just data, there would be a lot of benefit. If somehow the new Dynatrace product - the AI engine that tells you where the problem is - would tell us where the problem is and bring it down to the pinpoint, to the root cause, so we don't have to spend time looking into it, or the developer team doesn't have to manually look into it, that would be good because those actions require time.

When we were selecting a new APM solution, the criterion of our company was performance. The performance team, they mainly use this tool to test how load capacity runs. So they are mainly testing for the performance stability of the application.

I would say that if you haven't used AppMon before, then just go and order Dynatrace, instead of going to AppMon and then transitioning.

The one that we recently scripted was just to see if an application was up, it was a very simple script. We had an issue with a vended solution at the university in which the application would just go down and the vendor had said that, "Oh, it never goes down, we have 100% uptime." We didn't have a good way to monitor that before. We said, "Our students our reporting that they cannot log in." With Synthetics, we wrote a very simple script that went to the landing page and tried to type and hit "enter," and that's it.

We have two-factor and some other things that prevent us from going too far into it, and we haven't figured out a technical solution for it, but it's a very simple use case of just making sure that the application is up.

Dynatrace itself is performing well.

We now know when the application is down, as opposed to students opening tickets and letting us know. So it's more proactive.

That it's always running.

From what I've learned today, here at the Perform 2018 conference, there are two things that I really want to see.

Number one, the thing that is preventing MSU from moving forward with Dynatrace right now is that we can't tag our customer traffic with a customizable tag. All of our students have a unique identifier and in AppMon we tag that and we can search by it very easily and it's very useful. But in Dynatrace, you can't yet customize and find people like that, so that's really preventing us. I heard that it's being worked on but I'm not sure when it's coming out.

The second thing that I'd really love to see is - I'm very impressed by all the new features that I've learned about at the conference - and one of the features is "impacted users." I would like to see a rate of impacted users. For example, how long has the problem been going on: 100 users in five minutes. Does that mean that in 3 hours if we don't get this solved, we're impacting x number of people? Understanding the rate at which the problem is impacting people would be a cool feature.

It seems very stable.

With the AI, it seems very scalable. 

We're going through a proof of concept right now, so we work very closely. They're knowledgeable and we get the right person whenever we call.

We didn't have a previous APM solution. We didn't know we needed this solution until we saw it. It was literally students calling in with problem tickets.

I'm just a supervisor, so I sat with our technician during setup. I didn't do any of the actual work, but it seems seamless. It installed in about two minutes. I really wanted to see it. I have to go to the assistant director and, eventually, the director and try to explain why I think we need the new technology.

They went through a whole eight month process. I wasn't there, I'm a new manager, but I understand they had over a dozen companies. They had spreadsheets of all the pros and cons. They went with Dynatrace because it just has more features.

Regarding AI and the ability of IT to scale into the cloud and manage performance problems, we don't have the new technology yet, we don't have the new AI, we have the old AppMon and Synthetics. But it seems like it's very important.

We have used siloed monitoring tools in the past. The university is very old and very segmented and different departments use different tools and we don't all talk to each other. We still have this problem today and we're trying to get more user adoption for Dynatrace. But it's difficult.

If we had just one solution that could provide real answers, and not just data, I think that we would spend less time working on things that probably don't matter, like mundane routine operations.

To me, the most important criteria when working with a vendor is responsiveness to issues.

My advice to colleagues would be, do your homework. But, I would be surprised if anybody would beat Dynatrace.

When we first got Dynatrace, it started just as a developer's tool. As we used it more and more, we started using it not only for applications but for hardware. Now we're using it for servers. it's evolved over time.

The use cases are monitoring, alerting, triaging, and for statistics and dashboards.

It's basically used to make sure that everything is running okay. You can check dashboards in each group. The DevOps group can use the dashboard and say, "This is how my code is working." So it's more peace of mind, and to keep things available, which is really important to us: availability and reliability. That's what it's being used for.

At product start-up, it was not very visible at first. But now, since people have actually been using the product, and they've actually seen the visibility and what it can provide, it's doing well. So that's why we're deciding to upgrade.

We've had the product for about eight years now. They started using it more and more, so the company decided to invest in it more and more.

Right now in an AppMon, when you have DevOps teams looking to see where transactions are going through from business transactions, and they want to set up measurements to see how things are. And dashboards - dashboards are extremely important now. Thresholds are important.

After today, at the Perform 2018 conference, I can say more because I can see, even though we're AppMon 6.5 now, and we're going to Dynatrace, I was really impressed with everything I saw. The fact that I was impressed, and I was seeing everything, what the future looks like, made us feel like we made the right choice as to what's going on.

Dashboards is one, troubleshooting is another. I come from the monitoring perspective, so the ability to triage quickly is important, and the ability to alert and tell people where the problem is, that's what I really like about the product.

I haven't had a chance to go through all of it, but I would like to see the ability, from an administrative standpoint, for it to collect statistics. I want to be able to see the servers that the agents are installed on. I want it to be able to start doing collections for me by platform: How many Linux servers do I have? How many Windows servers do I have? Statistically give me the information of how things are performing, but I want that in a dashboard, where I can look at a dashboard and I can look at a section. So the ability for me to drill down will make it easier for me.

Ours is pretty stable, being that we have AppMon. We have collectors and DTs in there, they're load-balanced all over the place. And every now and then we have issues, but it's been pretty stable. It has to be, because it's being relied upon more and more. If it was not stable then management wouldn't say, "Okay, we're going to go this direction." 

Stability is extremely important because when you're in my business, you need to have data available 24/7. If I have systems that take a hit, and I can't depend on my dashboard to pull up data, then that's a problem. And if they can't triage, then that's a problem.

We have encountered downtime. When we have downtime we look at it and say, "How quickly can you get it up?" The good thing is that when we had downtime, we also had load-balancing in place. If you have that in place that really helps out, because then you go from one to another. It doesn't mean that it wasn't down, but it wasn't down long.

We can't afford to be down, because being down costs money. Everything is equated to money, to some degree. If this tool is not available during this time, how much money am I losing because I can't triage it? How many resources am I using? We've had some downtime, but not much. Even when you do maintenance, it's done a certain way.

Nothing is 100%, but it's pretty reliable. It had to be impressive enough for our management to allow us to invest into it going forward.

It scales pretty well. I'm from a pretty big company. The solution started out small and scaled out, because we have a lot of collectors.

We scaled up. The good thing is, that's where Dynatrace SaaS is going to come in, when you look at what it's costing us to maintain the hardware on-prem, and do the analysis. 

Scalability: We're always building more applications. And the fact that we can get rid of a lot of hardware and run that tool is really important.

Scalability is really important with growth. You need something that can scale quickly and not have to go through this whole process. It's one thing to look at it and do an analysis and say, "Okay, I need to scale." But, it would be easier if you could look at something and put in a formula and say, "Alright, we can scale based on this." Or something that could tell us. "You're at a point where you need to scale," so that would help out.

It's easier when you can get to a level-2. We've had guardians on site for a while. Because we've had guardians on site they made it easier for us to get to support. So we know how to navigate support. 

From my perspective, support has been quite satisfactory, because it's not like we're a novice going at this. We understand, these are the people you need to talk to. And they recognize the urgency when you're calling. So, I feel comfortable with saying support has been pretty good for us.

I wasn't involved in the initial setup, but since then we've upgraded so many times. The upgrade is pretty easy. Complexity comes in when you have to schedule things. If I have a DevOps team that's in prod, once we were able to actually cookie cut the upgrade process, it was easier.

It also depends on the different platforms you're dealing with. You're dealing with Windows versus Linux, and you have to inject code into certain places, that's an issue. 

I'm glad when Dynatrace says they have a different agent, because I wasn't crazy about the agent, the way we were deploying it. Not only me, but you have people who are quite protective of their code, and if you've got to inject an agent, they're not really comfortable with that. But with the new agent, that makes it easier for me too. That's another reason I was happy, that's one of the things I asked about: How are we doing the agent? Even though we were automated, there were still some configurations. But with this new product, it seems like deploying agents is going to be awesome.

I'm excited because I like the AI piece. I want to get rid of the thresholds piece. That's part of the excitement as to where we are going to go, because a lot of the conversation occurs when you have to say, "What's the real threshold based on the applications that we have?" We have over 500 applications, easily. Each one has its own behavior. Then everyone wants to discuss a threshold. Now, I'm thinking, with Dynatrace SaaS, we don't have to do that anymore. Bit it will be awhile before we get there.

AI is really important. We're on-prem right now for hybrid to go to the cloud. So when you have the numbers and trends based on what AI can do, it helps you along the way. I can run reports and see the things that I need that will satisfy my customer, that will make it easier for me. We are looking towards the cloud, we do have applications up there. It becomes a sizing issue, as far as what do you need, because you have to pay to be in the cloud. So AI is really important because it will not only help us troubleshoot, it will actually predict some of the things we need to help us get there. I can't put a number on the importance of AI right now.

It's going to change a lot of analysis and things that you have to do. The one thing I would like to see is what's involved in setting up the AI. Because I'm excited about it, I want to see what's involved in setting it up, and see that component.

Regarding siloed monitoring tools, portability is one of the challenges. Siloed monitoring tools make it really hard to port out to other places. You get data, but you can't necessarily use it as far as fitting it into other areas. That's the problem with siloed monitoring tools. If you're good for that scope then you're okay, but when it's time to go beyond that scope... Going enterprise-wide is important. So, where siloed monitoring tools were okay at the time, they're just not keeping up what's going on now. The monitoring field is evolving. It's more than about just monitoring. Monitoring is one thing, alerting is another thing. But the AI part puts it together, makes it easier, not only for you to do your job, but the self-healing piece that goes with it. That's really exciting, when things can self-heal and then just report on it. That makes life a lot easier.

If we had just one solution that could provide real answers, and not just data, the benefit would depend on how you look at a benefit. First of all, we would be more efficient. However, I'm not sure if it would help from a resource standpoint, because then I'm not going to need all those resources, if I have a tool that's going to do everything I need. But, it would help in the sense that, if I can resolve something like that, and things fall into place. That would greatly help. And not only that, then I could just depend one tool. Right now, people look at a solution as, "I need a tool for this, and a tool for that." But if I can move towards one tool that will provide me everything I need, then that would be great.

Our most important criteria when working or selecting a vendor: Of course, there's always cost. Stability is another. The amount of time they have been in the market. Then we do a PoC and see if they can meet the use cases that we have. We have a standard, some 125 use cases that we put out there. We see they perform based on those use cases. It's completely agnostic. I don't care who the vendor is. Can they meet my use cases? Do they have stability? Do they have the reputation? And then we negotiate cost.

Right now, the solution we have in place, I would give it about an eight out of 10. Things are changing, technology needs to change. It's been doing a job, but I mentioned the things I asked for. I'm looking for more, so that's probably why an eight. It's doing the job, but I'm looking for it to go further. 

They have a new product that's going to go further, so that's how we do it. Now, the new solution, I'm not sure how we're rating it yet. I have to see what it's going to do. 

But would I recommend it? Yes. Because we have the experience, it has done this and this for me. Can I depend on it? Yes.

If a colleague at another company was looking to implement a similar solution I would need to know they were looking for before I could advise them. I would tell them from my perspective, this is what the tool has done for us. If they ask me the questions, I'll tell them exactly what I think. At this point, would you recommend this solution? Sure, because it does this for me. If you're looking into looking at code, at that part, yes it does that for me. If you're looking for monitoring, yes it does that for me. I would recommend it. 

We currently use AppMon. It has been performing great compared to what we replaced.

The main use case is application performance monitoring and availability.

The main benefits will be more proactive, high-quality performance monitoring and availability with more insight into the application. It is very impressive.

It has helped my business move forward probably just by taking the next step that everybody needs to think about, which is AI and the possibility of cloud. I do not know what direction the company will go with that yet.

• PurePath
• Being able to drill down and detect what the problems are.
• More insight into detecting what the problem is with some of our applications. 

We are still struggling a bit with finding an answer quickly. It is all there, but there is so much that it is still really hard to figure out exactly which way you want to go:

• What do I right click? 
• Where do I go? 

It is a little overwhelming with the amount of data that we have. I think it is more of an issue on our end. We just need to get more familiar with it. We just started implementing it last June, are still relatively new, and are currently struggling with some performance issues.

We have already seen the next release. It is doing so much more than what we ever expected. 

Right now, we are struggling a bit with stability. 

We have outages. We have not pinpointed what the problem is. We have data loss. We have had to restart and have not narrowed down exactly what the problem is yet. So, that is in the works.

Scalability is good. 

Because we are having stability issues right now, it depends on what we need to do:

• Do we need to add more memory? 
• Do we need to add more collectors? 
• Do we need to bump up something? 

So, this is a little uncertain at the moment.

Technical support is good. 

There was only one time that I had to contact someone at two o'clock in the morning and the response was not as quick as I thought it was going to be. However, we addressed that with meetings. I think they have done a lot better job of their support.

We have used siloed monitoring tools, specifically CA APM. The challenges were probably the end-to-end and getting the full picture of what the problem was. There always seemed to be some major piece missing and Dynatrace has allowed us to get as close to that big picture as we possibly can. Then, with OneAgent, it has to provide the problem to you. It does not really get any better than that.

We picked Dynatrace because of the value that it provides.

Our environment is very complicated anyway, so the initial setup was a bit of a struggle, but only because we have so many applications and JVMs that we have been working on for long time. We had to move everything from CA APM to Dynatrace, so it was a big conversion process.

We found a tool that can be utilized by testing, DevOps, marketing, software engineering, and monitoring. Before, we always had everybody doing their own thing. Now, everybody's utilizing one tool, which is huge. That is a huge savings.

Definitely look at Dynatrace if you are looking to purchase an APM solution.

We did evaluate other solutions, but I was not a part of that process.

If I had just one solution which could provide real answers as opposed to just data, the benefit to my team would be time savings. We are scrambling all the time to try to figure out when there is a problem. If we could have something else telling us what the problem is, we could spend more time fixing it, providing valuable dashboards. and other valuable monitoring, then have a better proactive monitoring environment. So, it is huge.

The role of AI when it comes to IT's ability to scale in the Cloud and manage performance problems is huge and really important, especially after seeing OneAgent. We will probably be moving to that, then probably be upgrading to version 7. I think that is the direction that the company will be going. They do not say that they are not supporting it, but it is highly encouraged that you jump onboard the OneAgent train.

Most important criteria when selecting an APM solution: Something that was less overhead for us. We were finding with our old tool, which we still support because we are still not off of it, to get the same thing that we get with Dynatrace required more servers and effort on our part to address everything that we wanted to from the performance side. So, there is less infrastructure to support and it is a little more consolidated.