PurePerformance - Azure Updates with Product Manager Patrick Thurner

Episode Date: January 31, 2019

Coming to you from Dynatrace Perform in Las Vegas, it's Pure Performance! Hello from Dynatrace Perform 2019 in Las Vegas. I'm Andy Grabner, and this is Up Close and Personal with Product Management on Pure Performance. I want to introduce you to a guest that I just bumped into, Patrick Duna. Hi, Patrick. Hey, how's it going? Pretty good, actually. We're just walking around, and Patrick, for people that may not know you, what's your role in Dynatrace? I'm Product Manager here at Dynatraceise and I'm mostly driving the Microsoft stack,

means primarily Azure as well as some agent stuff like.NET, H&N and so on. So yeah, a lot of things going on here. Pretty cool. I have done a couple of interviews with your colleagues from product management, asking them about their breakouts. I think you have not actively had a breakout? No, no breakout. Several hot days, hands-on trainings, sessions around Azure, like the Azure monitoring stuff going on, supporting Sophia as well on your unbreakable pipeline on DevOps, which was great.

So yeah, looking forward to the session. Yeah. Now, obviously, there's a lot of stuff happening in your world around Azure. So from obviously there's a lot of stuff happening in your world, right, around Azure. So from a product management perspective, what are the things that people should be aware of? Like the folks that are listening in are most likely people that are interested obviously in Dynatrace, but also about cloud. I mean, now Azure specific, what is there that people need to know about Dynatrace,

the capabilities and the use cases, and maybe what's planned? Right. So there's a lot of things going on. So Azure for us is really a first class citizen on Dynatrace and also Azure is a premier partner when it comes to monitoring for us. And the partnership here is really core because Azure comes with its own monitoring platform like the home Azure Monitor services and so on. So with that in mind, with all our capabilities, we have with the one agent and all the integrations we have in Azure, like the extensions to make it really easy to put our one agent on the services to monitor them.

It's really key to have also the integration of the existing monitoring metrics and alert stuff into Dynatrace to provide better root cause analysis within our product. So that comes really handy with the close relationship to Microsoft tying all these things into Dynatrace. And we just short before GA of our Azure dashboard, which is the out of the box integration of Azure Monitor Metrics.

So that provides you deeper insights into the past services where we cannot run our one agent there, but it's heavy, interesting to see in the root cause analysis if any, for example, service quota or throttling happens and to see if any, for example, service quota or shortening happens, to see if any impact on your services is related to that. That's pretty cool. So that's similar. I mean, I look back a while ago when, I think, from a cloud perspective,

we started with AWS, right? And we implemented integration with CloudWatch, and then we had the AWS dashboard, as you said, that gave us all the overview. So now we have the same thing for Azure. Exactly. And even more. So we started with the new metrics engine already for the Azure metrics.

So that provides automatically all the metrics flowing to our new Davis AI engine to provide this analytics out of the box in Dynatrace. That's pretty cool. Now, how about rolling out agents? So is there, I mean, again, I'm not, I mean, I used to do a lot with Microsoft, now I think I'm not as close as you are anymore with the technology. What are the most common ways to roll out the one agent

in an Azure environment, and what do we provide to make it easier for customers? So the most common one is using the extension mechanisms. So everything in Azure is provisioned through the Azure resource managers. Means that's an API and you have the PowerShell, you have the Azure CLI, you can do it through the portal. And with that in mind, we have the VM extension, for example, for the VM based services we just install the one engine through the extensions on whatever mechanism you use so that works

in in automation it works for the portal manually and it's the most easiest one the same for example for for app services we have a site extension over there it's just within the portal a few clicks to do there and set up the agent monitoring. And then there's certain services which are building up on the same extensions and some of them, like the previous old cloud services, we have some out-of-the-box scripts available to run the one-agent installation and tied to your automation. So it's pretty easy.

So that's pretty cool so besides like say monitoring the stack I know that Azure DevOps is a big topic right I mean we had just recently excited workshop here in Linz with the large team also like Sophia was here Rob the people that are driving this from our perspective anything you want to mention on the Azure DevOps and know you also said you did a hot day. So just some thoughts on that, how we integrate and how we enable. I think it's the native tooling for Azure.

A lot of customers are still using other automation toolings like Jenkins. But especially in my role, it's important also to have a good integration with Azure DevOps. And we provide this extension. We're doing further enhancements to that one, making it much easier for customers tying Dynatrace to their monitoring in the DevOps pipeline. It means you have a really smooth rollout into your Azure ecosystem. So there's a lot of things to come, so we're still figuring out what we can do here.

But with your Unbreakable Pipeline that's already working on Azure, I think it's a big step to the next. Yeah. Cool things. That's pretty cool. Yeah, I think we also learned a lot from that, right? I mean, we're going through different iterations. So that's why you said, you said we learned from the first version that was thankfully provided

by Donovan Brown and by Abel Wang. Yeah, they did a great job in showing us what's doable now. It's on us to make this tie good into our product, make it easy, as good as possible, and drive this path. That's pretty cool. So, looking a little bit into the future, because people are always excited to hear what else is coming. Now, you just said the Azure DevOps are going GA.

Looking further down the line, is there anything you can reveal of what's coming in the next couple of months? Absolutely. So, with the Azure monitoring integration, which we provide this out-of-the-box dashboard, we're extending that far more. So we're integrating all the PaaS services which we're monitoring with Azure Monitor into SmartScript, the whole topologies. We then see also your PaaS services in the topology and the whole service flow. So that's cool. We're providing the charting capabilities of all these metrics. We're working on integrating the alerts of Azure Monitor in Dynatrace

because some of the use cases you cannot just fix by monitoring the metrics, for example. If you get into throttling of services due to specific tiers, you just get alerts for that. So for that it's important to also tie alerts which then get elevated in Dynaprace as events, whether it's problem events, warnings, information, to provide really the information on these things. So we're just extending also all the service support, coming all these data, big data services like HD insight get the metrics from these Azure Functions support we just released a beta support for that that's working pretty cool on on dotnet core compiled applications so cheese then just the next thing to provide it make it much easier with

automatic service detection and so on and so forth. So, yeah, working also close with the Service Fabric team. Shravan from Service Fabric team is also here to perform, helping us on the hands-on training sessions. So that's a great opportunity for all attendees to meet with the Microsoft guys. Yeah, and there's a lot of things in the pipeline. That's pretty cool. So for the folks that are listening and those that might actually be here in Vegas, if they want to meet you and

talk a little bit more about Azure, I believe you will be around in the marketplace? On the marketplace, yeah, on the towers for the distributed tracing. So just pay me a visit and have a nice chat. That's awesome. Looking forward. Cool. Hey, thanks, Patrick. Thanks.

It was a pleasure, as always. Yeah. For Pure Performance, I'm Andy Grabner. Thank you.