 Hello, and welcome back to another episode of As It Updates. I'm Anthony Bartolo, joined today by Pierre Roman. Pierre, how's it going? I'm good, how are you? Good, good. I'm a little bit upset. I made a big mistake. So apparently, only North America moved forward with the time zone change in terms of daylight savings time. And I forgot to update the schedule for Sarah, who is supposed to also be joining us on the show today. So Sarah is unaware that we're actually an hour earlier than we were last week. Because in Europe, they haven't done daylight savings time. They changed at a different time. So definitely find mistake. Apologies to Sarah. She's unable to join us for the show today because the scheduling is different. But we have producer Pierre on the show, which is awesome. Yes, I know. It's a big mistake on my fault. I did not realize that you're- May I call Paul? May I call Paul? May I call Paul? Europe does not change in terms of time zone change. So apologies. We're actually an hour earlier now in Europe for the next... When do they change? Do you know when they change in Europe? I have no idea. The ST, or daylight saving times, is different across... For us, different countries, across different regions, for us in Canada, across different provinces. So it's a bit of a crapshoot to try to figure out who's doing it and who's not doing it. There's actually a lot of talk about stopping the DS, they like leaving it to daylight savings or leaving it to... One or the other? Yeah, because really, it used... It was something that was set up for to get the farmers more daylight. Well, just get up earlier because the sun doesn't change. It's like... Anyway, I'm missing the point of what daylight saving is. Other than to screw up with our schedules and our sleep patterns. Well, again, big apologies to Sarah Lane. Apologies for that. Should have known in terms of scheduling that Europe is different. Jay Shock mentions in the chat that it's in two weeks. And then the European Union will be in line with us in regards to daylight savings time so that hopefully this won't happen again. Sarah's actually gonna be on assignment for the next two weeks. She's got a couple of projects on the go with the CAP team. So we'll actually have a couple guest hosts for the next two shows. And then Sarah will join us back on the show. So thank you, producer Pierre, for stopping in last minute. Let's just say a quick... Oh, sorry, go ahead. I was gonna say, I might have to rely on you. I didn't go through all of the show notes. It's okay. We'll wing it. We'll wing it. We'll wing it. And the people in the chat room will help out as well. So hello to Robert Jr. We have Jay Shock, aka producer Steve. We have Paul Jensen in the online. A lot of people chatting on there already in regards to the excitement around the show, what we're gonna be talking about. I quickly wanted to cover your new show that's in conjunction with Jared or producer Steve that you have both put together. No, it's producer Pierre and Steve, the audio guy. Is that... Oh, we're going by Steve the audio. But he does the visual too though. Yeah, I know, but he's always been known as Steve the audio guy. All right. And he's reminded me of the choice. And he's gonna correct us in the chat room whether or not it's correct. And yes, Jared, we are testing in production. We are testing in production right now. Yeah. So tell us a little bit about the show. Well, the show is about many things. It's about streaming. It's about recording at home. It's about technology in general. I'm sure at some point we're gonna talk about Windows. We're gonna talk about Azure. We're gonna talk about IT Pro stuff. But it's not necessarily just about the IT Pro business. It's, I helped produce this show. Jared helps produce Patch and Switch. We've done a lot of watch parties for Ignites and pre-ignites and post-Ignite shows over the past few years. And every once in a while, we wanna try something like a virtual whiteboard, like the video from Hanselman on using a tablet or a Surface Book in my case. And Microsoft whiteboard to basically turn an OBS feed into a virtual white lightboard. But sometimes you wanna test things with, like during a stream, because something sometimes on your own machine doesn't look exactly the same as when you're going across the wire. So we decided to have a show where we can talk about the different things that we're trying, trying about gear for people who are interested in terms of like lighting cameras set up that we have and try some stuff, but try it in production. I thought it was really cool. You know, I thought it was, you know, in the fact that you were going behind the scenes, showing how a lot of the producing is done, a lot of the technology, and it's, you know, applicable to a lot of us right now that are speaking in virtual conferences and online conferences. The tips and tricks that you shared, both you and Jared are applicable to, you know, if you wanna jazz up your presentation, if you wanna provide that umph or impact in terms of your delivery, what was shared last week, and you'll have the show this week is patch and switch. So you'll have the show again next week. It's really impactful in regards to your delivery mechanisms when you're delivering in a virtual conference. Yeah, well, this is what we're hoping for, and hopefully people, it resonates with people. If it's not your cup of tea and you're not planning on streaming, we will be talking about other things at some point. So let's get started with the news and the first item to talk about today is the Azure availability zones coming to more regions. Yes. Jared, you wanna take this one? Jared. Are you gonna share the screens? Yeah, I'll do all the screen sharing. So you just talk away. All right, so Azure availability zones. I've had that conversations with customers in the past where not every region had availability zones, and availability zones is within a data center or within a series of data centers within a region where they're configured where there's three, at least two or three distinct zones within, so a separate power, separate network, separate everything so that when you're having an application or a workload that is split across availability zones, when you pick your availability zones to deploy your cluster or your workload, if one of those like network power cooling, et cetera, goes down or has some issues, you can fall back onto the other and now we've committed to expand that availability zones to more region. In fact, they've committed that by the end of 2021, every country in which we operate Azure regions will deliver Azure availability zones for their customers. So it's a big deal. It's kind of like we've committed because it's been asked by customers for a long time. So we've said, yep, all right, we're doing it. We had been doing it on newer data centers. So the new ones that we are announcing, we've been basically taking that, but now we're going back and all the ones that we had missed. So for example, Canada East and Canada Central for in our case for Anthony are now going to be by the end of the year, have the availability zones that they're looking for. And what's exciting is that our friends south of the border in Brazil have received availability zones. And I know they're a big user of Azure. And so that availability made available to them, I think oil and gas is big in Brazil, are taking advantage of that now and more and more will be made available. We'll keep an eye on it as availability zones become available into specific regions. We'll share that with everybody in terms of what region is being lit up. Definitely excited that it's coming to Canada. I know we don't have it yet, but it's something that a lot of the financial sector here has been very eager for. It's interesting that we're doing this now and I'm thinking that going forward, I think you mentioned this, any new data center that we're launching will automatically now include availability zones. Yeah, it's part of the baseline if you will now, but at the beginning it wasn't really so much because we were more talking about the zone redundant services. So if you're doing something in Canada, the East, for example, your redundant region would be Canada West. But now within regions, we're gonna have availability zones. In some cases where in a zone that has multiple data center, it's a little easier to implement because you already have that physical separation. But in regions where it's a large single data centers, now you have to reformat the data center itself to kind of introduce that separation within a single physical data center. So it's not as easy, it's not as straightforward as you would think, but we are committed to getting it done across all of our regions by the end of 2021. So let's continue on with the news. Next up is the availability now of alerting in terms of budgeting for your Azure spend. We talked about this before in regards to governance and controlling in terms of what people can set up in terms of resources and you're governing over the ability to spin up specific VMs or specific resources required to do your day-to-day operations. Obviously, everybody wants to do the most expensive one, thinking that's gonna best fit their application, but with governance you can control that. Now there's a new functionality being available to cost management on Azure cost management that you can set up alerts. And what the alerts do is that you can now set up budgets for your organization in terms of spend and then be alerted in terms of a threshold that you set when somebody is about to close in on that spend or that allotted budget for your cloud instance. This is really, really well done in fact that I can now be notified to say, hey, this individual's gonna be at this spend very shortly and it's gonna cause the budgets to go out of whack. Also allows you to do the triangulation in terms of, well, who's spending what and why is there spend going overboard and do I have to adjust my budgets based on needs? So if accounting during a busy season requires that additional umph in terms of processing power to have that allotted in the budget so that you can go through and make sure that the allotted budget for compute or resources available on Azure is specific to each tailored need of each department inside of an organization and have that alert mechanism made available not only to the IT professionals but also to the business decision makers to have that understanding in terms of how the spend is going and how close we're getting to the threshold of what's been budgeted for the spend itself on for cloud computing services. Pierre thoughts? Well, I really love that and actually, I wish that I had been available when I was running my last demo subscription. It's not just about the budget itself because the budget is great being able to allocate some money. It's the alerting where because sometimes you would either forget about a part or as you set the budget and your SREs or your developers or IT pros deploy the workloads. Sometimes it doesn't actually deploy. There's changes that happen in production where I don't know, like another machine jumps up or whether or not as you're deploying if there's a quota issue, you say, well, I'm just gonna get a bigger machine and we'll address later. And then all of a sudden your cost estimates are out of whack and then you get a notification. It doesn't block anything, but it basically tells you that something is off versus what you had planned. So you can go back and check and cover it. In my case, when I did a demo where I deployed in a scale set of machines and just for shock and ha during the demo, I picked the G2 machine, which at the time I think or G5, I think at the time was like $8,000 a month. And I forgot to turn them off after the demo and I just closed my laptop. So thank you very much. And I walked off the stage and left 10 machines at $8,000 a month running in my demo subscription, which was not free by the way, it was still billed to our cost center. So at the end of the month, I got a fairly nasty email from, well not nasty, he was really good about it, but it was like, hey, what the hell is there a $80,000 bill on your account that's supposed to be $400 a month? If I had had notification, I would have found out. So Andrew McCollum joined us in the chat and made a good point, maybe a good to budget the Azure Sentinel. I bring that up only because in scenarios like Azure Sentinel, when you're doing your security spec for your organization, you should have that budgeted in terms of a separate bucket in terms of your organization's security. It shouldn't just be as part of the resources spun up based on your applications or the services that you run. You should have that understanding. Okay, so if I deploy Sentinel in my organization there's a reason as to why and the budgeting spend for the Sentinel being utilized is X and it should be separate. And it should be, the big thing here with this is specifically Azure cost management and the learning capabilities made for it is to keep the business decision maker in the mill. So perfect example is your example in terms of spinning up the G2s and it's an $8,000 a month budget requirement. And then there's the sticker shock when it comes up at the end of the month saying, whoa, what is this bill, right? Can you imagine the business decision maker looking at that bill going, what is this? I had no idea this was coming through, what's going on in cost and using Azure cost management with alerting. You can now keep the business decision maker in tune what's going on in terms of budget spend. So stuff like Azure Sentinel doesn't become a surprise when there's a spike because there could be an attack that's happening at your organization, which is incurring additional costs, right? The business decision maker would be aware and there would be no sticker shock to the end of the month saying, well, how come we spent all this money and then you have to do all this investigation, right? That applies to all services. Correct. Because I've also had discussions with log analytics because we recommend a single log analytics as little number of log analytics workspace as possible because the ingestion of the data is the cost that's the expensive part. And if you're doing a hub and spoke environment, you might think, okay, I'm ingesting a terabyte of data that's gonna cost me X, but really because it's hub and spoke, you're doubling that amount. And if you're not realizing that, oh, because I'm ingesting it twice, there's gonna be a charge for that. So now Andrew brings up a good question in regards to what habits that should he then look at in regards to his use of Sentinel. I'd have to investigate further, but I think you can actually manage your log analytics piece in terms of ingestion because that is the most expensive part is the ingestion of the data itself. You could lessen the frequency of that ingestion, but that does then provide a bigger attack plane for your organization, right? It's a balance. So you have to make sure that you're still adhering to what your organization security requirements are without going overboard in terms of how much data or how frequently are you consuming that data for security analysis? Yeah, and Sentinel uses log analytics workspace to store the information. And I think is in April sometimes, actually I just had a blog post this week and a video with the PM of the Azure Monitor agent, basically how we're collecting the info for Sentinel for performance monitoring for Azure Monitor for all these, the model is changing where you're gonna have more granular control over how and what information you're actually ingesting into your log analytics. So go check it out and you'll be able to tweak because you don't need every performance monitor on a machine if you're only monitoring, let's say CPU and disk. So you can check that out. All right, next up in the news. Make workloads on AMD back virtual machines. Pierre, do you got this one? Yeah, sure. Actually, you have the wrong. No, that's actually the screen, but it's okay. So the big thing here is the Mark Rosinovich announcement. Yes, so we are now providing a confidential platform that you can deploy your code to that is AMD back. So it's the new Epic 7007,003 processor series. And it's a new approach that complements what we've already got in terms of confidential computing solutions, such as confidential containers for Azure Kubernetes services and that opens a possibility to create new applications without requiring code modification to turn them instantly into something that is a lot more secure. And that's the big thing, right? There's no code modification required. This is inherently secure to the virtual machine itself based on the 7003 Epic processor provided by AMD in partnership with Microsoft. This is something where Microsoft is trying something new based on security requirements have been put forth by organizations. And so there is a current private preview being made available for people to test this out. We say this on the show numerous times. Hey, if you wanna have your input and the services that are being put forth by Microsoft, definitely jump into opportunities like this to have your voice heard. So this link is being available on ITOps talk. Pierre is showing the link right now, aka.ms4 slash azitupdateshow. All the links for today's show is provided in this blog post. We have one every week, every Friday that shares all the links that we've talked about on the show. Sign up for this private preview. It is limited. There's not availability for everyone because it is a private preview. But if you have something in your organization or there's a service organization that requires this type of enhanced security, definitely check this out and get into the private preview early. Well, if you can't get into the private preview then keep your eye on it because the next step is gonna be the public preview and then you can basically jump onto that. Yes, very interesting. So I got a quick question for you, Pierre. In terms of backup, back in the DLT days, what was the longest time that your backups would survive on those tapes? Oh, hard to tell because back then, we didn't do as many test restores as we should have. So at the end of the week, you would take the box of tape or on Monday morning, after the weekend full backup, you would take the box of tape and you would take it to the bank or to your off-site storage, in some case was like my basement versus the office downtown. And then you would basically store those tapes and you had the grandfather, father, son kind of rotation. How often did you actually go back and say is last year's backup good? Well, that's just it, right? I'd have, from my backups, I would have a backup for the year, a backup for the month and a backup for the day, right? A daily one would be doing a test infrequently. It wouldn't be done every day. It would be done every other day or what have you. The monthly one would be checked and the yearly one would be checked, but only checked once. And it wouldn't be checked again. It would be in the safe. If you've seen me speak about backup, we've probably heard the horror story of one of my friends who used to take the tape, put it in the safe at their home, which was magnetically locked and then the tape would be erased, right? The reason we're talking about this, I know, I know. Put the magnet on it and see if it works. But it's not the first time I heard that, right? It's not the first time I heard that, you know, then there's introduction of Azure backup and there's, you know, taking the information, backing it off site, making the data available if you want to spin it up in the test area for production applications to be tested out before being put to production. So much goodness in terms of Azure backup. Now there's a new functionality in terms of long-term backup retention for Azure SQL. Managed instances currently in public preview. And what this is is you have the availability to store Azure SQL managed instance in a separate repository for up to 10 years. So this is long-term storage. If you need sales data, if you need, you know, specific technical diagrams, if you need architectural data, all that information in long-term retention backup for 10 years held up in the cloud in a separate repository from your production or testing environment or even your Azure backup environment so that you can have and go to at any time. Obviously you can still can do your testing. It is only currently available or in general preview for SQL managed instances. Microsoft is working to broaden this capability across a plethora of SQL offerings that Microsoft does have. Kara shaking his head, seeing a drink. So it's coming. But again, it is currently in general preview. If you have a requirement or need for this, definitely go check it out, test it and let us know what you think. Yeah, that's a really, really important long-term, especially for compliance issues and for historical data type of, especially even in an area where for GDPR for example, somebody calls you and said, I want you to remove my information from your environment. So you got to go back to all your backups and be able to do that. Also the fact that it's in the cloud and you don't have to worry about storing it, like the physical storage part of it is also really, really important. And the fact that now with Azure Backup Center, which eventually will surface that environment, you'll only get one place to manage all of your backup, whether they're the long-term storage, which are basically on like deep cold storage, meaning performance is out the window. You're not looking at the very high throughputs, but it is there for a long, long period of time. So really cool, really important. I used to say that we used to be the tag on my personal blog, relax, but make a backup. Paul Jensen on chat makes mention, he works in a very regulated industry and longer retention is key to us. So it's something, you know, Paul, definitely go and check it out in regards to the general preview, provide your thoughts in terms of this capability. I was thinking, you know, 10 years worth of backup on, you know, financials, I'd hate to see my numbers for device spend. Yeah, and all that so depends on your industry. So Paul worked in a highly regulated industry. You're talking about financials, so financials in Canada, it's seven years worth of data that you have to retain. So it all depends that you really, IT pros and operations folks really need to be in touch with if they're working for a large corporation or any kind of corporation that has compliance, they have to be meet regularly with the lawyers of the company to figure out what your compliance footprint looks like. So you can actually manage it in a way that you retain compliance across all your workloads. Let's continue on. So next up are events that are coming up. First up, Hello World. Hello World is a new daily show put on by our team. So all of us were cloud advocacy team. It is on at 10 30 AM Eastern standard time every day on Learn TV aka.ms4 slash Learn TV. It's a new show that talks about everything that's happening at Microsoft, everything that's happening in terms of Microsoft's impact around the world, how the world impacts Microsoft, everything from what we're doing in technology from an IT pro developer and end user, machine learning, governance, you name it. Every show is different. Every show has three to four segments, which talks about everything that's going on. And they have a plethora of hosts. And a plethora of hosts. Also today, as we mentioned earlier, it is a fortnight. So that means that patch and switch is on today, which means that testing and production will happen next week. But we all jump on the patch and switch. So we always have fun in the chat room. So definitely join us for patch and switch today. Last but not least in terms of events, the digital event for the Azure VMware solution, digital event is next week. So on the 23rd of March, if you haven't registered yet, if you are using Azure VMware solution, do check out this event. Jason Zander is the keynote speaker for this event, but you do need to register to be able to participate. You can watch the event without registering, but if you have burning questions that you need to ask, definitely register for the event so that you can have your voice heard and possibly have a question asked to Jason Zander. Yeah, and if you are using our VMware solution on Azure and you have questions, you can always go to our Discord server because we have Shannon and Jeremiah Dooley, which basically run or cover that area of our team on our Discord server. You can ask questions directly to them. And now Puma and Schmidt will also be answering questions as well, and she was doing that before, but she's now gonna be focusing a lot more on VMware and the Azure VMware solution, which we're very excited about because of the amount of history she has in the background. Let's talk about the Microsoft Learn module of the week. We have one minute left. Create an enterprise scale architecture in Azure. If you're looking to migrate portions of your organization to Azure, to the cloud, here's the best way to do it in terms of learning the uptick of the tool that's immediately available. This cloud adoption framework tool is something where you can go through and hands on labs to learn about the functionality to move your organization to the cloud in a hybrid or full cloud implementation. Woo, Pierre, it's been a busy show. If people wanna get ahold of you, what's the best way to do it ahold of you? Wired Canuck pretty much anywhere on socials. And if you wanna get ahold of me, you can find me at wirelesslife on Twitter. Hey everybody, thank you for joining us on the show. Have a great weekend and we'll see you next week. Cheers.