 Hey everybody, it's Jay, and there's Pierre. Welcome back to AZ Update. How you doing? Sorry, the video was playing, and all of a sudden all my browsers went, poop, gone. Well, when you do it and you do it live, things don't always go the way you plan. That's right. I think I've learned that from enough live demos in places across the world with hundreds of people sitting and watching. Sometimes you do it live and things just don't always go as planned. No, that's right. And that's one of the reasons why when we started AZ Update, we were doing it using OBS from our local machine. But then we always ended up having the single point of failure. And now, see, my setup kind of blew up and I had to restart all my browsers. But since you were still in the session, it kept on going. I just disappeared and then came back. Big shout-outs to StreamYard. You got one hell of a good product. I really enjoy using it. I would say that the last 18 months of my career would not really have been as easy to complete without a broadcast studio that I could just use whenever I need. Yeah. This week was a busy week. Talking about streaming and StreamYard and all of that good stuff. Ignite, Microsoft Ignite, did you catch any of the sessions? Yeah, I caught a little bit of it. I watched the Windows Update session that there was. It was a breakout about all the new different types of Windows Update services. So I think it's like WOP, stuff like that. It was really interesting because I'm more of a hosting forward person. And to hear a lot about how IT ops people are working around updating and patching was really interesting to me. It's still something that's a lot of work. Oh, it is. And it's super important for our friends. I actually really liked the Blueprint session about Windows patching. I actually enjoyed the whole series of Blueprints. I found they were really well done, very concise. And I've actually bookmarked those. They're in my backpack. So I'll probably download some of that content and watch it again and see what else I can find out as there was so much information I think I missed on a few. Yeah, that's the one of these things that happens, especially with virtual conferences. Is that there tends to be so many tracks, so many different talks that you want to see, conversations, Q&A. It's difficult to get to everything. I know our own teammate, April Edwards, she had a huge PowerShell-based Q&A. Just told me maybe 100-something people showed up after their big talk. And they had some conversations about how the ecosystem around PowerShell keeps changing. Yeah, and I had a chat with Jason Elmick and Snover yesterday. And they mentioned how awesome it was to not just have the sessions that they had and then ask the experts for PowerShell, but to be able to extend this into the after-party was super cool for them. So I can only imagine I couldn't attend. I had some personal things to take care of. But I can only imagine how fun it would be for all the people on YouTube to connect with these guys and be able to ask questions like completely like no old barred, no recording, no pretty cool. Yeah, and there was just so much that came out that it's going to be difficult for us to cover just about everything. It's impossible because there are so many different features that were released, new services that were announced, new ways to work. That's one of the big things that comes out of all of this is that there are new ways to work with technology. And Microsoft is always introducing new ways to improve productivity, whether it's for the person utilizing Office 365 and the suite of services, or it's somebody who's building servers to have Kubernetes clusters. Every time one of these events comes around, you find out all the hard work that people from around product groups and engineering have been doing in order to actually accomplish big, big things. Yeah, and if you didn't get to all of the news for the audience, if you didn't get to all the news and you missed some announcements in the blog post on aka.ms.acupdate, for today, you'll see that there's a link to the Book of News for Ignite November 2021. All of the announcements are listed there with links to the announcement blogs and to everything else that's required or that's offered in terms of information. So check it out if you have not done that already. I love this Azure Updates blog that I'm sharing now. I think that that's one of the easiest ways whenever something new comes out, I always keep an eye on there just to find out what's new, what's different, what's changing about Azure. Yeah, and there's a lot. I know, actually, I use that one. I use their RSS feed in my inbox so that every time something comes up in the morning, I see it right away. I do like the Book of News the way they organize it because if you're an infrastructure, there's a section for infrastructure and all the announcements for infrastructures are there. If you're a developer, all the announcements are there. If you're a .NET person and so on, so it's really well organized that way. Speaking of Book of News and Updates, there was a ton. And there was some really cool stuff, like Microsoft Loop. Is that Microsoft Loop? Yeah, Microsoft Loop. Did you take a look at that? Not yet. No, I saw it was another way of transforming hybrid work. So it's about like a flexible canvas, what I saw, a new way to kind of collaborate together. Yeah, I really think there's potential there. I'm gonna have to dive into it and see how we can get that into our Teams channel and so you can add Teams events into the loop and add other stuff in there and make it so that your work is actually all kind of coalesces into one kind of view. Looking forward to see, maybe I'm wrong, maybe all I've seen was the announcement, so I have to dive into that. Then there's also the metaverse. I don't know what my avatar is gonna look like, but it'll probably be bald. Well, you know what's crazy also is that there was an announcement that now we've actually even turned, we have so many database options at Microsoft. And now Excel essentially is now a database. It's always kind of been that way, but now with that API for JavaScript that they released so that you can actually query, I mean, Excel is a database. So let's not go back to the good old days or bad old days of using something like an access database in a single spot. Don't use an Excel database in just a single spot. You use a real database, but it's neat to know that you can create custom queries maybe and output it into some sort of homepage that you can use for an organization that does everything in Excel. Yeah, the funny thing, Johnny Chips tells me that I can be whoever I want as my avatar, so maybe my avatar will have hair, like big, like Hades Bon Jovi hair. Yeah, mine's gonna be, if I can make mine a gorilla, that's what I wanna be. I wanna be in the metaverse of gorilla. It's my dream come true is to be a gorilla. Yeah. Well, speaking of gorillas and metaverse and so on, how about we go in with the news? Let's get it done. It was really hard for us to actually pinpoint like four or five items for us to talk about today because there was so much. So we've each picked a couple and I'm gonna start with the first one, which is VMM Flex or virtual machine scale set flexible orchestration mode. What that really means is like in the past, we've been using a VMM SS, so a scale set to scale application, but the problem with that, well, none, it's not necessarily a problem, but the keyword in those for VMM SS is that all the machines have to be exactly the same. So homogenous and stateless. So if you're- Because it's set up that if you lose one, one should be replaced immediately based on your scaling and it should auto scale based on metrics that you configure. Exactly. So the way it does it is it basically creates kind of like this gold image with your code on it. And as your scale rules are met, more machines come on or more machines come off, but they're all exactly the same. With VMM SS Flex, as I call it, because say like virtual machine scale set flexible orchestration mode is a mouthful. So VMM Flex, you have some flexibility. First of all, you can have different size machines. You can have different family machines and you can go, let's say you have the three machines in your scale set to handle your nominal load. Sure. The other thing that- Sorry, your scaling can be done with spot instances. So you save a significant amount of money if their spot instances are available. If not, it just picks up another one. Speaking of databases, this is an ideal kind of solution for say open source databases that you may be running on your own that require some sort of replication services. So like MySQL, MariaDB, where they do different types of replication, Postgres or log shipping, the fact that you can have a primary and then secondaries, even like MongoDB is a great example where you have kind of a multi-node set. And if one is to fail, it should just pick right back up. And shouldn't really see much of an outage. So what we want is when these machines come back online, if they're one fails, that it becomes part of a quorum again, part of the application, and it allows you to just continue what you're doing. You shouldn't have to manually make changes. No, and the cool thing is you mentioned databases. So now you can actually have stateful workloads. So if somebody starts a session and the state of the session is stored there and you have that session breaks, then it comes back or you're scaling up to a spot instances, you can actually keep going with that. So different kind of machines, stateful workloads, a lot more flexible spot instances. There's like, there's no real reason now when you're deploying VMs to not deploy VMs in a VMM SS flex environment, because you can just deploy one and then add it later if you need to. So I see this as the next default. Yeah, and here's the thing is that you can set up your disk images as some sort of like a standard image in a vault and then take one of those images that you eventually bring in and have them as your master if you need for one of these VM nodes. Makes a lot of sense to be able to use VM scale sets because you're just adding redundancy by default. Yeah. All right, so we don't take too much time. Let's go on to the next one. And the next one is yours and it is your Azure Chaos. Azure Chaos. Yeah, Azure Chaos Studio systematically improved resilience with controlled chaos. And so if you aren't familiar with chaos engineering, there's a really great website. It's called principlesofchaos.org. And the big idea around chaos engineering is the idea that we can experiment on a system in order to build confidence in the system's capability to withstand turbulent conditions in production. So what does that mean? Someone much smarter than me once said everything's fails all the time, always. And with a product like Chaos Studio, what you can actually do is try to reproduce incidents. You can create game days so that you can say, all right, today what we're going to do is we're gonna go into our production system and we're gonna say if we have any microservices, we're gonna kill a number of different services within a Kubernetes group. Or if you're using VMs, you can say what we're going to do is add CPU stress to a few of our VPN VMs to see whether or not it can withstand an overabundance of CPU traffic on that and still bring up our application. So the big idea behind cycle reliability engineering is being able to use observability to make determinations on not necessarily is the server at green and 100% available. But how many servers do you have in a pool that can withstand not being at 100% in green? If you can have two or three in a 10 pool set and you can still have a quick return to an HTTP call for a web app, then you've got it right. And being able to ensure that you can take that resilience is really a big part of working with Chaos Studio and in Chaos Engineering. So there's a service direct thing so you can work around with Cosmos TV. You can set it up for Azure Kubernetes service. So if you wanna kill particular pods, you wanna add some CPU stress to a virtual machine, you can do that as well because it also supports VMs. It also can attack VNets because sometimes your VNets make it over phone with data and you may need to find out, do I have to add additional routes in order to make sure my data flow works if there's some sort of influx of connections throughout a VNet? Well, that's the thing is back in day, back in my days, no, when we were on-prem and you wanted to test, for example, a cluster, one of the cool things that, well, cool. One of the easy things that we used to do is go behind the cluster and like pull a network cable and see whether or not the workload would fail over. And now you could do that with the Chaos Studio virtually. Netflix came up with this concept a few years ago. The practice is maybe 10 years old now. Netflix started seeing a need to be able to make sure that their application could withstand failure at different levels and still allow them to provide streaming services. And so you can do the same thing for your web applications that you're deploying into the cloud or whatever your business applications that you're supporting. If you wanna make sure that in a particular host pool that it's still working, you kill a couple of specific VMs that your organization is using. Maybe you're using a front end for the VMs and you just wanna make sure things are working. Exactly, that's just one more way to validate that your environment is resilient and test it on an ongoing basis, but also before you actually put it in production. So let's move on to your next item. You, Arth, can talk about Azure Container Apps. Sure, so Azure Container Apps is another way of deploying applications in microservices using a serverless container service, if you will. So you can run modern applications at scale. It allows you to use Kubernetes event-driven auto-scaling distributed application runtime with Dapper Envoy for a deeply integrated hosting service. You basically go into the container app service and you configure your microservice style applications. So you provide disk image or right now you can actually go on to the preview and if you want, there is a quick start image. It's in preview now, so the service isn't fully baked, but there are horizontal auto-scale triggers. You can, like I said, make microservice development easier by using Dapper because Dapper is native to this product. And I know you got a big earful about Dapper recently. Yeah, we had the DapperCon for two days a couple of weeks ago, that was kind of cool. So it's just another way that Azure has given people a container orchestration method. If you wanna use Kubernetes, there's Kubernetes that you use directly. If you wanna use this fully managed serverless container service, you can use that. So you can deploy your containerized apps and without managing the infrastructure. You can write code based on your preferred programming language and you can build it into microservices and- Jay, let me ask you a question because I'm not the really deep on a container and Kubernetes, but so how is that different, for example, than Azure Container Instance or AKS? Container Instances or AKS is a Kubernetes service that's hosting. Essentially, we're giving you a control plane for Kubernetes, but you're still managing the pods. You're still setting everything up on your own. In this case, you're providing an image, you're giving some trigger information and then you are saying, all right, here's my image, build that. And so you're not necessarily needing to manage the Kubernetes cluster, create manifest or use helm. All these things are taking care of for you. You're just providing your disk image and it's going ahead and deploying that. Container instances are good for, say, testing and single instance orchestration. You can use it for multiple microservices if you really want to. But with a service like Flexible Containers, you're going to have a load balancer already configured for you that's gonna work within your containers. You're gonna get that application URL right out of the gate. If you want to go to the product page, I'll put it in our chat. You can take a look at the product page and you can see that there's a bunch of information there that with a little bit of an example that you could take a look. So I recommend, check out the quick start. It builds in just a few minutes and there's integration with GitHub actions and things like that so that you can set up CICD for the process. Okay, which is really cool. Sorry, I now understand a bit more where it fits. And as you were talking about that, I was thinking in my head, well, that means that the IT ops folks that are managing this environment now can just basically set what the standards or what the amount of resources that'll be used. And then they don't have to start managing like the AKS cluster itself. So they're basically freed up to work on other stuff like resiliency and backup and like the real operation stuff, not just tweaking config files in the product. Well, I really look at any way that you could orchestrate application deployment or service deployment, whichever you wanna look at it and reduce the amount of overhead that the administrator needs to do is only beneficial to productivity. And that's the way I look at it. So the more we provide people simple ways to build things, the faster that they're going to be able to deploy. Yeah, I always kind of try to like quicken my head as to, okay, so how I'm looking at a product and I'm looking at a service and it's like, how is that gonna benefit my IT ops? Like what's the ops story to this? And now I get what the ops story is for Azure Container in service. Your ops team may be just working on, determining the right Docker environment, making sure patch levels are correct within all the required dependencies. So then maybe working on scanning dependencies, say for MPM modules to make sure that we have the latest MPM modules. So your IT ops people are still going to be heavily involved in the process. Absolutely, absolutely. That's really cool. All right, so it's, we are going, coming towards the end of the show. So let's look at the last item, news item. And it's one that's near and dear to my heart, Azure Monitor. I've been a big fan of Azure Monitor for a long time. And I'm basically just a fan of being able to know how my resources, regardless where they may be, are doing. Are they all patched? Are they all compliant? Are they all happy? And good little resources, whether they're VMs or other thing or on-prem or in somebody else's cloud. And Azure Monitor gives me that single point, that single pane of glass for me to actually look and kind of see a top-down view of my entire area. And so much more Azure Arc support to get that kind of experience across different places, whether it's in Azure or on another hosting platform. Yeah, and this week, what was announced is, it's not like it's new Azure Monitor, but they've added a bunch of stuff to Azure Monitor. That just makes it so much easier. So one of the things they asked is the ability for if you've got to open source workloads that you're running to use something like open telemetry in your open source application to send data to Azure Monitor. So the Azure Monitor can then ingest that data and show it in that dashboard to show that that application is healthy. But they also added a whole bunch of insights. And the insights, the insights is basically the equivalent of what the OMS solutions were. So the member operations management system and we had solutions. So if you had like active directory, you would install the active directory solution. And they would basically tailor what the triggers and the metrics or how I would show not healthy to healthy. So it was kind of like a management pack for SCOM in the old days, but now they call them insights. So they've added a, go ahead. That network one that they just added, I think it's brilliant because you sometimes need a way to be able to troubleshoot your gateways and your firewall. Absolutely. You know, this provides just more visibility into how traffic is flowing. Yeah, so they've added one for network. As you mentioned, they've added one for Kubernetes that runs on Azure Arc enabled servers. They've added one for VPN, gateway, and Azure firewalls. For log analytics itself is you're collecting all that data into a log analytics workspace, but how do you know that that log analytics workspace is healthy? So now there's actually an insight that looks at the health of your log analytics, which is the basis of Azure Monitor or where the data is stored for Azure Monitor. So it's really, really cool. And I find that more insights get added into that product. The more granular and the more specialized we can create dashboards and graphs that can actually show you the real state of your environment wherever it may be. So I'm really, really happy about that one. So it's cool. Yeah, everything that came out of Ignite this year really reminded us that the idea of adding productivity into everyone's lives that work around IT solutions is just important. So adding ways to keep progressing what you've done already, adding more insight, adding more simplicity to the process, I think was one of the things that I really got and making it so it's easier for people to collaborate. Yeah, it's been a lot of information. I'm actually still processing a lot of it and see how it fits because two days like that, it's just too much. And... Yeah. Well, not too much, but you know what I mean? It's a lot. It's a lot to dissolve into a 30 minute show. That being said, this is what we've caught for this week, but as I mentioned, go to the updates link that we put up earlier. Did you create a banner for that? Yeah, I did, actually. Okay, I don't see it. But anyway, put it up if you can. So go to the updates or go to the book of news. You'll get all of the information. So to close, we have our Learn Module of the Week. And the... Near and dear to your heart. Yes, absolutely. Because it's about monitoring cloud resources. And we did with Oren Thomas and Sonja Kuff, we did a session at Ignite on kind of like the getting started and what does that mean to monitor cloud resources and really hybrid resources. But this one is a fairly quick learning path or learning module, that if you haven't been exposed to Azure Monitor and you're kind of wondering how to set that up, what kind of information you can get and all of that stuff, go and do this Learn Module. It's very quick and it'll basically take you through how to build reliable application monitoring and resource monitoring for your environment, regardless whether it's in the cloud or our hybrid. So I really, really recommend you checking that out. Sounds good. Well, we got another one in Pierre and you got a couple of weeks off. Yes, I am off for two weeks possibly three. I have to talk to my boss and I see that he's in the chat. So maybe we'll have a little discussion offline after. So we'll see you in a few weeks. Thank you for picking up the Slack for the next two weeks, Jay. Have a great weekend. We'll see you next week with Rick. Yeah. All right, cheers.