Considering a VDI Migration? No Decision Is Still a Decision | Workspace Weekly

Good day, and welcome to the migration series part one, when change is an optional, confidently navigating your VDI migration. My name is Michael Kent, and I’ll be your speaker for the next forty five minutes or so. We do have a gift to give away at the end of the webinar based on your participation in a couple of polls that you’ll see at the end of the presentation. A little bit of housekeeping. Presentation will run about forty five minutes. We do have a section about ten minutes for q and a. Save your questions. You can post them in the chat. You should have a sidebar that has some options there for you to both view the polls when they come up and to ask questions, and we’ll we’ll table those questions until the q and a section. And I will take about ten minutes to go through some assets that we’ve created as part of this presentation to help guide you through some of the scoring systems that we’ll present to you here today. As I mentioned, you’re welcome to email me after this presentation if you should have any questions that we didn’t cover or you think of a day from now, for example. By way of introduction, my name is Michael Kent. I’m the chief product officer for LoginVSI. Background wise, I think I started in IT when I was nineteen. I’ve gone through data center migrations. I’ve gone through cloud migrations. I worked at a cloud migration company. Been around the block a few times. Have the battle scars to prove it. Here’s my email address should you have a question after. Feel free to engage with me either on socials or email directly. Okay. Let’s dive in. When the status quo is no longer an option, change is inevitable. And so here we are. We’re starting to see some requirements for change. Right? It’s all around us. What we’ve tried to do here is put together kind of some strategy, some specific evaluations, assessments, and a plan to help you get through. Take the shopping cart method. That means absorb what you need, leave what you don’t. We’re really targeting your desktop infrastructure admins, your Windows app deployment professionals, and, any VDI leaders that, will run into this in the future. Today, we’re gonna cover the shifting landscape, compare some of your options, go through a step by step readiness assessment, really spend some time on risk identification. We’ll walk through a strategy, and then we’ll take questions at the end. So let’s get started with that shifting landscape. Something is driving infrastructure change. A lot of it is pricing pressure. Some of it is end of life. Some of it are new options that are now available to us that we didn’t have access to before. Definitely some some preferential pricing in some of those scenarios as well. So we see license upheaval as being probably one of the biggest drivers. Right? There was a lot of consolidation in the industry. This company bought this VDI provider. This company bought this VDI provider, and all of our prices went up. Right? I think I even heard one customer say in their meeting with their VDI provider, the providers actually said to them, let’s be honest. You’ve been stealing from us, which is a great way to start any sales conversation. Right? The the problem is real, and that’s putting a lot of pressure on teams to make a move. Support and end of life issues, are a great checkpoint. Right? This is this is another great inflection point that makes us look at where we’re at and where we wanna go. Oh, if they’re gonna bundle pricing for the next version of Windows, maybe that’s something I wanna look at. Definitely, security posture gaps. We know that the old way was not built for some of the new ways. Right? Zero trust being the obvious one. Yeah. We’re trying to retrofit, and and there’s a lot of products coming into our data centers and coming into our lives to help that. Right? And it seems like everyone has another one, and that’s now why your images have fifteen or twenty security agents. Right? One that does this, one that does that, all trying to address an old way with a new way. And then these cloud make, cloud first mandates, I hear less about those today than I did two years ago, three years ago, five years ago. We’re all in on cloud. Great. You still have to go through a process. Right? There are still things that you’ve accumulated over time that you have to address. So we’ll talk about that a little bit as well. Quick reminder. Remember, you can ask questions now, but we’ll answer them later. So feel free to to use that question box. No actions or what what I like to say to people around me is no decision is still a decision. Right? A decision to stay put is a decision. And, really, we don’t have that that option anymore. Every time we decide to defer, there’s a cost to it. There’s budget pressure. We’re under pressure to deliver a service at the best possible price with the least possible overhead. The longer we wait, the more expensive it can become. Right? It’s gonna come to a crisis. And anytime it comes to a crisis, we have to move fast. And anytime we have to move fast, we need a fast moving vehicle, and fast moving vehicles are expensive. Right? A Lamborghini isn’t cheap, but it’ll get you there fast. A U Haul truck is a lot cheaper, and it can carry a lot more, but it’s definitely not fast. And then we see organizations. A lot of organizations are already impacted. We say forty percent. It’s probably understated. So we ask the first honest question. Is it even right to migrate? Maybe you don’t have the pressure, but we know that renewal is coming for you, and it’s at least twenty percent higher. We know that certain platforms are gonna overhaul in the next eighteen months? What are you doing for that? We know that our ability to support some of the remote and hybrid work is maybe not as good as it should be. Right? I have a great story here. Worked for a company. Pandemic hit. Everyone ran in, grabbed their laptops, ran home. Then they found out that they didn’t have enough space on their VPN concentrator. And they tried to use a cloud service, and they tried this. And two or three months later, they finally were able to acquire the capacity but at a huge cost, a twenty eight million dollar shortfall because of a loss of productivity. You definitely don’t wanna be in that situation. And then, license model mismatches. Right? So named users is the the the way of the the future. There is some flexibility in there, but that is going to change your cost structure. We wanna make sure that we’re on a platform that is under active development, that major refreshes are accounted for. Legacy app dependencies are being addressed. I’ve heard IT transformation for probably twenty years, and still I would say we’re not there. Right? Every technology is the technology that’s gonna get us there. It was web one, web two, web three dot o. Everything is gonna get us there, and we’re still not there. And then insufficient organizational bandwidth. Look. As long as I’ve been in IT, we’ve been the redheaded stepchild. Right? Sorry for any of the gingers out there. But the the idea is cobbler’s children. Right? All the good stuff goes to other people, and we’re stuck with the lowest budgets, the tiniest cost centers, and a lot of the blame. Right? And so that’s not going to change. That’s just a bandwidth issue. Should I have enough people to throw at this problem? It would be easy, but I don’t. And part of that cost reduction is making sure that my staff is pared down to the absolute essentials. And that doesn’t leave me a whole lot of room to do these evaluations, to keep on top of things, and to swing from one technology to another in a safe way. So what are some of those options? Here, my slides got a little munching, and apologize ahead of time. Yeah. You can kind of make out what what we’re going for here up at the the stars, but we have on premise VDI, and I give that kind of a five star maturity. Five star maturity because it grew up with us. Twenty, thirty years of issue solution, issue solution, issue solution. Your on premise VDI has some of the most feature rich capabilities of any option that’s out there, and that’s important to take in account. As we move and we look at vendor cloud DAS, some of them are getting pretty good. Right? But it’s super important for you to understand that you just got single sign on for one of them, like, beginning of this year, four years after you needed it kind of thing. Native cloud DaaS, not quite there. Right? So I understand you might have to be stuck with your last option, very situational, which is this hybrid. Something on the top of something so that I get screen recording, and I get the security that I need, and I get the protocol stack that works for my company or works with my integrations, etcetera. We have upfront costs. Right? High CapEx. We know that drill. We’ve been doing that for a decade or more. We’re hitting capacity, need to purchase another server. That’s a four month procurement cycle. Can we wait? Can we push it off? You know, we might wait until the the end of the quarter. Can we get a deal? Right? Now you’re part of the the negotiating team. Vendor cloud DAS has a low to middle kind of, OpEx focused licensing and compute costs. Same thing with your your cloud and DAS. You’re moving to an OpEx model. Simple, but maybe not as feature rich as you need it. Right? Maybe they don’t have per minute or or on demand pricing for for your shift workers. These are things that you have to consider. And then, of course, hybrid, which is very situational. Right? You’ll have some CapEx. You’ll have some OpEx. You’re gonna get stuck somewhere in the middle. How do you make sure you don’t stall in a bad place? The operations burden shifts, and, of course, finding your best fit. Are you GPU heavy? Do you have security concerns? Are you highly dispersed? Do your users fit in a nice, neat little boxes? And then, how are we gonna deal with multistate, multisite, state being state of being, right, complexity. I moved half of my workforce, but now my file sharing has to change, and we’re stuck in the state that I have part of my company on premise and part of my company in the cloud, and it’s really inefficient. Right? As we do our evaluation, we have some some assets, as I mentioned, that we’ll go through at the end of the presentation. Total cost of ownership. Have we done that calculation? I know nobody cares anymore. You have a mandate, but it’s still important for you to understand and know because it will impact you. Security and compliance, huge, huge requirement. Now, trying to replay all of the security and compliance that you have on prem in the cloud is a trick. Right? Probably means changing products. It definitely means changing, deployment styles. You definitely have to consider that. These would be weightings you would decide. These are my weights. I say that’s twenty five percent. You might be in health care, and your security and appliances fifty percent, and that takes away from other areas. Performance in UX, I’d give you kind of what I think the the lay of the land is there. On prem, you really have land based connections. They’re super fast. They’re they’re great. You’ve deployed an application that has a database back end. They’re all in the same network. It’s great. As soon as you start to break that up, that performance is gonna change. Operational agility, you know, one could one could argue this either way. If you’re a mature organization, maybe your operational agility on prem is great. Right? You have everything automated, your infrastructure as code, etcetera, etcetera. But, definitely, cloud has more APIs to get that stuff done, where you may not have all of those APIs or have coded to all of those APIs internally. And then the big one. Right? We’re really having this conversation because of an incredible amount of vendor pricing pressure is making everyone reevaluate. Right? After all the the trials and tribulations that we’ve had with all of these on premise VDI systems, now on top of insult to injury, we get a huge price increase, a significant price increase. That’s a problem. I really don’t wanna be in this situation again. So can I evaluate that pain versus the pain to do something different? If we take a look at it, just, again, take a step back. Right? Let’s take the the the vendor marketing speak out and take the the future forecasting out and just look at where we are. Probably the top benefits of cloud are are elastic scalability. That’s the one I hear the most. I don’t have to worry about hardware anymore. Yay. You pay for that. Reduced infrastructure ops burden. You know, the that is all taken care of by someone else. Right? If a server blows a drive, I don’t get the call in the middle of the night. That’s very real. Geographic flexibility. Again, the pandemic created a situation where people are highly distributed. When they all came into my office, it was great that I had a system in the office. It was fast. It was well connected. It was well maintained. Now that everyone’s somewhere else, is it more amenable to have some flexibility in geolocations? And that’s also an evaluation criteria for the cloud VDI. Do they have a point of presence that is close to my users and will keep that latency down? And then I don’t know an organization that isn’t being pressured to shift CapEx to OpEx. So those are real benefits. Of course, you have real challenges, and that is latency, both login, and underlying dependencies. I have an application as database. Where’s that database? I know. You’re gonna move the database into the cloud until you find out that five other teams are using it, and then it’s a big reporting hub, and that’s where we’re doing our data marting. And then all of a sudden, it becomes this complex web. Egress costs are a real thing. Right? Nobody really wants to talk about them, but depending, you can have a significant cost for all the network traffic going into and out of the cloud and around. Of course, the cloud vendors will tell you, just put everything in the cloud, right, as will your overly simplified c level. Oh, just put it all in the cloud. You know, not always that easy. Right? There’s always some some stickiness there. Compliance boundaries get complicated. Right? Are we a European company? Are we a US company? How do we keep data in one place versus another place? We didn’t have that problem before because it’s all in one spot. But now as I go geolocational, what is what happens to my data? How do I enforce that and keep compliant with all the different regulations? And, you know, we still have to look at vendor lock in. Right? The more I use a feature set of a particular stack, the more I am beholden to that stack. It’s kinda what got us here, and, I don’t expect that anyone building a product will change that for you. So how do we take a look at our environment and see what we have? We start with a migration readiness assessment. Right? Maybe you’ve already done this. Maybe you haven’t. Maybe you’re gonna wing it. Important that you look at what do I have, what are some of the configurations I’ve had to accumulate over time that I have to account for, validate, or do I still need them? Are they answered by my target solution, what have you? It’s a it’s a very real discovery process that you have to go through. Yeah. Hardware refresh cycles. Well, I need to replace a failing or ailing server before I get through a migration? Right? That that’s a a real, scenario. And do I have all the networking equipment and whatnot in place to support these half states as we’re doing these moves? Workload and app analysis. As I mentioned, you don’t wanna find out when you move the application to the cloud that it has some dependency on some on premise component that is difficult to manage, in in distance. GPU specialization. Right? What is the answer from your vendor on that? How do I get access to it? How is it charged, etcetera? Persistent, nonpersistent delivery. Right? If you go all persistent, that changes your profiling. It changes a lot. User profile and experience, what are these personas? What class of service am I giving them? Where are they located? How sensitive are some of the workloads that they’ll be using? Right? Some applications are fine. A lot of applications will time out faster because they don’t expect long latency. And then do we have baselines on satisfaction benchmarks? Right? Now a lot of us I mean, if we can’t do anything about them, we don’t really wanna measure them, but you should measure them. Right? Because at the end of the day, two things happen. Your users stop complaining, but they still have problems. They’re complaining to someone. It’s just not to you, and you don’t wanna be stuck in that blame. If you’re doing a transition, what you don’t want the answer to be, it’s as bad or worse than where we were. How did we not know this? That’s not a good conversation to have with any kind of management team. Then finally, compliance and security. As we mentioned, what is the basis for your security? Do you have the tooling? What tooling do you need to account for? Where does your data live? How are you gonna deal with users in all these different regions? Identity and access infrastructure. Right? How are we gonna make sure they are who they are? Implement the right controls, sensitivity, data privacy, etcetera. And now as everything splinters or changes, do I still have the same auditing capability? Because my compliance is gonna require that. From migration readiness, we go into a migration scorecard. I’ll show you this asset as well, in just a little bit. And here, what we’re looking at is, is my infrastructure documented? Can I reduce my risk by understanding what I have? Are my apps tested? At least for my critical apps and my critical workflows, have I accounted for something that as I do this transition, I can understand the impact? Network capacity. Yes. We have egress and ingress. What is that gonna mean to my users and by which region? Right? Not all of these options are created equal. Do I understand what that’s gonna look like? Is an activity by my marketing team gonna really impact my developers, etcetera, etcetera? The security control mapping, as I mentioned, it’s very possible you’re going to have to remap some of these to new service lines. Do you understand what they are, why they’re there, if they’re still needed, etcetera? You don’t wanna redo a bunch of work that is unnecessary. And then purse really defining those user personas. Right? These are my my task workers. These are my knowledge workers. These are my engineers. Because for every rule, there’s an exception, and so you at least wanna bucket your users so that as you configure your profile options and you configure their service, sizing, etcetera, you can use those buckets. Understand, though, for every rule, there is an exception. Once we have that readiness scorecard okay, we think we’re ready, we do wanna do one additional check, and that is just risk identification. There’s a lot of things that we know. We know that it’s gonna be a problem because we’ve heard it from a colleague or or, another person or another company ran into it. It’s always the things that are out there to bite us that we’re unaware of, where a larger perspective helps us capture, quantify, and assess. Do I have this problem too? Profile bloat. Right? So if I have persistent VDI and I’m moving, what does that mean? If I’m switching from roaming to something else, what does that mean? If I have my profile sitting on storage equipment that deduplicates, what does that mean in migration? Right? I have something that know that under the covers does a really tricky I see that Michael has the spreadsheet, and I see that Bob has the spreadsheet. So I’m gonna only save the spreadsheet once and put a pointer to someone else, we’re going to get bloat. Right? Concurrent sessioning, licensing traps. Right? So, we probably were able to manage how many people were on the system at one time and get some cost of scale based on that. That goes away with per user pricing. Yeah. There are some options out there, but you still have to evaluate those options. It’s a risk. Printer redirection. I told a story earlier about people running in, grabbing their laptops, running home, and a quarter later, them finding a twenty eight million dollar shortfall. That twenty eight million dollar shortfall was printer and profile redirection. What used to take five seconds to print a packing slip to a remote warehouse for the product to ship took five minutes. They had the same number of orders. It was a medical company. They never lost an order. They were a critical medical device. They just couldn’t ship fast enough. Nobody caught that it went from five seconds to five minutes when they were home over VPN for a quarter, and they shipped twenty eight million dollars less because of it. Let’s not undersell the importance of redirection. Identity provider chains. Right? So, oh, yeah, everything’s working off of my ADFS on my on premise site. How is that going to change, and how am I going to account for it? We’ve already, I think, whipped the dead horse of the egress costs, so we won’t, touch that anymore. It’s just a risk. And then, now we start getting into some of the regional challenges, and that is Image Drift. Right? It’s a Herculean effort to have one gold image. Of all of the companies I know, I know one that has done it really, really well. And if I were to show you their map of of the minefield, it is you you’d need an entire wall to see all of the ifs and thens and and what ifs and that kind of stuff. They actually are not on one single image anymore because they they have a a Chinese language requirement, and it was just too hard to get one system, one image, etcetera. So now they have one more image. But Drift is real. It’s a it’s a risk. It’s something you have to look for as you start regionalizing. How am I gonna manage that risk? Okay. So we’ve done an assessment. We have an understanding of our readiness. We’ve looked for some risks and mitigations to those risks. Now we move into, we think we’re ready. Let’s build a strategy. First part of that strategy we covered. Let’s do a discovery and assessment. Let’s just understand the lay of the land. You’ll have a readiness scorecard. It will it will, help guide you along that that path and help you have conversations with other teams and, managements as to here’s where we think we are from a readiness standpoint. Here’s what we think we need to do to be more ready. And, ultimately, this is what it will cost in time and dollars to be really ready. Do we wanna be successful, or do we are we in a hurry? And let’s balance the risk that way. Designing and planning. Right? So once we understand what we have to account for, then we look at the service, and we start designing that service to answer those questions. How am I gonna do profiles? Am I doing frontline? Do I have the shift workers that will nicely fit into three blocks, etcetera, etcetera, etcetera? If we are not using a a particular cloud, are we doing an evaluation process comparing a versus b versus c? You do still have options. What’s that process look like, and what are those questions that you need to answer? And then architecture blueprinting, not just for run rates. Right? So, step zero, but also for, Doctor. Right? I know how I do that on premise. I just don’t know what that looks like if I’m multi region out to my users. Right? We go into pilot and validation. One of the benefits of Login Enterprise as a product, LoginVSI’s product, is that we do things the way a user does. So instead of sending your sacrificial lamb into an area and letting friends and family get beat up by the configuration and and the issues and whatnot, we have a virtual user. It acts as your users act, and so it can take the lumps. So you can send it in as the canary, and it can give you a consistent baseline so you understand what the performance is even as you load up that ring deployment. So ring zero. I send in my canary first, make sure it’s okay, then I send in my friends and family. Right? Then I send in my power users. Then I send in my, subject matter experts. But in each ring, I send the canary in first. And I leave that in there because I want a qualitative measurement as I load up that system. Right? Performance at stage zero may be fantastic. Stage one, when nobody lives there yet, everyone’s moving in, could also look awesome. But as I get to fifty percent capacity, what does it look like? Am I still good? If I have a continuous set of eyes rolling these baselines and letting me know how performance is, I can move faster and more confidently, and that’s often important. It’s also an easy way for me to test my rollback plan. You will have to roll back somewhere, somehow, someway, for some reason. Right? If you wanna reduce your risk, again, by using this this concept of testing in a virtual user, you can actually roll back the plan. I move forward. My virtual user goes red. There’s a problem. I leave that problem looping. And then as I roll back my system, I’m waiting to see that virtual user turn green because then I know I’ve restored the service per my service line. We go on to four. We actually do the phased rollout. Again, using some concept of a a ring deployment or or similar, we can start moving groups of people, send in a canary first, whatever canary you have, but send in a canary first. And, ideally, it’s continuous so that you understand the moment something peaks or flips. And then finally, you go through and start doing that decom process. You’re in your new environment here at step five. Now we care about kind of constantly watching it as products change, as updates hit, as applications upgrade, and as my vendor provides new SKUs that aim I might want to evaluate because they’re closer to what my persona needs from a mapping perspective than the last one. Okay. This is our five phase migration road map. We can talk a little bit about, key elements of the vendor. Right? What is my profile, my protocol performance? Right? Do I need to to use another location causing duplication, causing image drift, etcetera, more locations, more things to manage? How am I gonna handle the GPU workload? You’re not getting away from it. It will come to get you somewhere. You have to understand what that plan is. Persistent versus nonpersistent, old oldest time, that one. And then, support for your hypervisor and, I’d say, more importantly, your storage stack. Right? Big difference when you get in the cloud and you start using cloud IOPS. Big difference. Everyone has a solution how to make it faster, but it’s usually not the default solution. If I, over time, have tuned a set of infrastructure for my users that are reading and writing to the disks heavily, and I move that to the cloud, I could have a big problem. It changes. If you don’t believe that, just try to unzip a file in a cloud PC Just see how that goes, and you’ll you’ll understand the need for IOPS. It’s just a different animal. You have to understand it. API and automation. Well, now that we have all this automation, what can we do with it? How can we minimize risk by using, CICD pipelines or infrastructure as code or what have you? And then absolutely wanna know what your options are from a disaster recovery perspective from that vendor because they will influence what you do. They’ll definitely influence how you build, and it may also be important, from a pricing perspective. Right? The one thing about cloud infrastructure is you pay what you should have paid in the data center, but you overextended or or, did some tricks to to solve. Well, I’ll use the old server as my backup server. It’s out of warranty, but we only use it in a disaster scenario. Doesn’t happen in the cloud. You pay for the the a and you pay pay for the b, right, from that perspective. Commercial fit, do they have the licensing strategy that I that my users need? Right? Maybe I don’t maybe I don’t have three shifts that can really take advantage of frontline. Maybe I just have an evening shift, so I’m paying still twice what I need. Right? Migration assistance. Right? What can they throw my way to to make this transition, less costly? License flexibility. Right? How does that play into it, especially based on my old infrastructure, compliance references? Right? How have other hospitals done this? How have other financial services institutions done this? What other government agencies are in there? Because this becomes important. Right? What other logistics companies are using something like this, and what were some of the challenges that they saw? And can I talk to them? Do they still have all their mental faculties, or did it drive them crazy? And then we can talk, SLA, but you’re probably gonna get what you’re gonna get. Right? The the target provider will have an SLA. You’ll have to decide if that SLA works for you. If not, you’ll have to change your deployment architecture to account for something better. Take more responsibility. That plays against your your OpEx. It plays against your financial modeling and your total cost of ownership. Okay. So key takeaways kind of by role. If you wanna start this migration, you can start the readiness scorecard. You’ll have it. You can actually probably start it this week. Start documenting your applications and dependencies. Of course, you wanna know them all, but I would start with your noisiest. And even inside of your noisiest applications, your noisiest steps inside those applications. If they’re gonna complain, they’re gonna complain when they hit that button in Excel that tries to load a million lines of a spreadsheet and do a pivot table. That’s where they’re really gonna complain. So you can still be surgical here and get a really good understanding of kind of your your highest risk areas even while you go through the rest of that documentation. Test your bandwidth. Test it from target source to target and from target to target. Right? So as we get these complex network infrastructures and how this user is gonna come in from this region, they’re gonna hit this this VNet, and they’re gonna go, you know, drop into this VPN tunnel. Test it. Please test it. And then identify your canary group. One of the I’ll shameless plug. LoginVSI has a virtual user that is a perfect canary. Infrastructure manager, you need to get your stakeholders aligned, build that charter, and get that sponsorship. Right? So the when a user says, why are we doing this? This is stupid. Everything was perfect. The the charter can say, look. This is a this is a top down push. It’s important for the solvency of the company. Yada yada yada. Right? But having that executive sponsorship in that manifesto is important later when people begin to squawk, and they will squawk. Map your compliance requirements. You don’t wanna be on the wrong side of that, and then define your success mind metrics. Right? You can’t please all the people all the time, but you should as a team, as a management team, and up through your executive alignment, have we are successful if these things drop in these tolerances. And then from a VP and director perspective, you’re the one to look at the TCO analysis. You’re the one to establish the the new governance models, communicate the strategic rationale for the the move clearly to both your organization and your leadership. This is the right thing to do. This is the right way to do it, and here’s why. Okay. Let’s see. I’m gonna go through some assets here in a minute, and then, Richard, I will answer your question. Okay. So what I had put together was just a a handful of assets. Again, if you have a full PMO office and you can manage that, fantastic. But to make sure we cover everyone that might have come in, I created just some simple assets to help. First here is a simple migration readiness scorecard. You’ll see instructions on the first tab. And on the second tab, we have just a a rating system. Is your infrastructure, ready? Score one to five. Your scoring guide is over there in column j, and it will calculate for you kind of a readiness as you go. You have access to the spreadsheet. You can change the weights. You can change the, the, formulas if you want. So if I am a three, I’m moderately ready for a change. If I am a five, I’m in great shape. If I’m a one or a two, I probably have an issue. Right? So here we go through app compatibility, network readiness, user experience, security and compliance, budget, governance, vendor readiness. So very important that, you have some sense of this. Am I ready to start this move? You can add to this. Right? Copy a formula. Get yourself a rating. It gives a great perspective to everyone working on the team, and it’s something that can even be debated. You can talk to your security teams. Hey. Do you think we are ready? Is the cloud ready for our security and compliance requirements? Because if, I mean, the the security officers are ruling the roost right now. Right? Whatever they say, everyone, you know, bows to to their to their word. Right? So if security and client says we’re not ready, then it’s not your fault. Right? They have some work to do. Great. Your director will target them and say, great. You have to get in here. It’s a mandate from the CEO. You have to get in there. You can’t tell him no. You just gotta tell him what and how much. Right? Gives you a great way to get a visual perspective of where you are. Migration rating and scorecard, that’s number one. K? Number two, we have a TCO calculator. Again, it has some assumptions. Right? So you can use this to try to get some understanding of the total cost of ownership of where you’re going. It does calculate, you know, do I wanna do what’s my on premise? What’s my cloud? How do those compare against each other? It’s important for you to have some sense of it. Right? What we tried to do here is list a bunch of those. You have partners like ITQ that can help if if you need. But the key here is that we’re trying to surface those costs so it’s not a surprise. And everyone knows what we’re getting. So later, when your CFO comes to you and says, why is our cloud usage so high? You can say, we we talked about this. We we we put it all on well, we put in your favorite tool, a spreadsheet, CFO. Like, how do you not speak Excel? So that way you have visibility and transparency. Super, super important. K? Number three asset that you’ll be able to download here in a moment is just a vendor comparison matrix. Again, we created a scoring system. You can change your scoring system. We left vendor names off at the top to, protect the innocent, if you will, but it just gives you some framework, some way to compare x versus y versus z. Is it a strategic fit? What is the operational complexity? Do they have the features I need? Am I gonna have to have an overlay to be able to get the same security that I have on prem today? And then some quantitative analysis and notes. That’s incredibly small. I apologize. Again, it just gives you some qualitative assessment notes, hey. Some some key strengths. So it’s not measurable by a percentage, but it’s still important to factor into your calculus. Okay. Moving along, we have a risk register. Again, I have some instructions for you. I have the actual risk register. Five through what what do we have here? Five through nineteen. These are risks that I’ve run into in the past. You can then add your own, but they give you some sense. Again, you decide the likelihood and the impact, and it will calculate a risk score and let you know if it’s critical. If you’ve closed the risk or comfortable, we have it handled. You can mark your status. It’s just to give you a framework for understanding what’s out there to go bump in the night. This also includes just a little heat map, if you will, that kinda helps you understand where you have potential issues. Right? As you close those, the criticality will change. Moving on. That’s number four. And last, but certainly not least, is your change management playbook. Again, it’s a guide. I’m sure you have marketing and communications teams that are willing to bend over backward to help you. I never had it. Right? I had to generate my own emails when I was doing these migrations. I had to come up with the language. I had to get a buddy to check it so I wasn’t, you know, making a bad joke or something like that because I think I’m funny and, you know, no one else does, that kind of stuff. And so, again, just as a framework, we’ve tried to put some initial email, the go live week reminders, go live day, maybe even a survey after. Be a good citizen. Do a survey after. Training. Hey. You’re now using something different. Can you complete this course so that when you move into your new home, you know how to live there? We have some, adoption metrics. Right? How is it looking now that we’ve moved over there? Did I deploy, Windows three sixty five to my organization, but no one’s using it? I should probably know. Right? Going back to agreeing with my management team on what are the key metrics, here’s some great ones. How is my login success? What are my tickets looking like? Do I have active users below my my license count? Eighty percent. Because I’m burning licenses, and people aren’t using them. And then what’s my satisfaction score? Just common areas that you can look to to get some sense of were you successful. And, of course, you wanna generate an FAQ. So I have some prepopulated, change them to your needs, but the idea is have a set of FAQs so that people can self medicate. Right? We don’t want all these things hitting our help desk day one. So if we can do something to help keep some of the ticket deflection, right, from the tier one help desk, those those poor souls that are in the the line of fire, let’s do that. Start with an FAQ. Let’s get it out to people early ahead. Right? Maybe even create that link, put it on the web, create that link, and send it as part of your premigration readiness, communications so that they have some understanding of what’s about to hit them. And if you set an expectation with the user, they’re a lot easier to work with later when, you know, risk has been realized. That makes sense. K. These will be available for you, and, I’ll drop the the link to that in the chat here in a second. Let’s get back to our presentation. Loading. Loading. Okay. So we went through the migration readiness scorecard. We went through your TCL calculator, a vendor comparison matrix, a risk register, and your change management playbook. Again, shopping cart method. Here is the link. I just threw them up on a public GitHub. So you can GitHub forward slash login v s I forward slash migration underscore series underscore assets, and you can download all five of those spreadsheets. If you have any trouble, you can send us an email, and I’ll happily make any adjustment I need. Okay. So I’m gonna start a poll, and I’m gonna ask answer Richard’s question. So let’s start the poll, And let’s get to this question. Are there any assessment tools that can be accurate, analyze the current geographical workload of an environment? I’m particularly interested in tools that can provide insights into metrics like GPU, CUDA, and overall GPU usage. Depends. NVIDIA built one. It’s called nVector, and nVector is designed to do what you’re looking to do, but it only works for NVIDIA GPUs. We’ve integrated into Login Enterprise, so it is an option. You can actually track your GPUs with Login Enterprise. If you’re unfamiliar with the platform or some of the folks, you can go to our website and look at our blogs link. We do a weekly series called the workspace weekly, and there is a workspace weekly where Joshua Kennedy goes through how to set up Login Enterprise to do GPU testing with Nvector as part of a partnership we put together with NVIDIA. For non NVIDIA, we do not have an option, but, honestly, I’d love to talk to you about it because we could probably put something together that follows a similar path. But as far as existing, tools, the only one that I’ve seen does it particularly well is Nvector, but it is only for NVIDIA GPUs, and it is already integrated into the login enterprise platform. Let’s see how our poll is going. So, Richard, hopefully, I answered your question. Let me know if there’s a follow-up there. Where is your organization right now? Are you actively evaluating? Are you, aware you have to move, but you haven’t started yet? Are you in the middle of a migration? And how is that going? And are you satisfied with current platform? Are you being forced by your vendor or a contract deadline or or or and we’ll start that poll. Let’s see what you have. And I’ll just remind you folks, participation in poll is get you entered in for the giveaway instead of, AirPods will be selected after the presentation is, complete. Alright. So far, we’re split. Oh, here we come. Evaluating migration options. I would expect that’s a lot of folks. Last year, I I got to a fair bit of trade shows, and it was the talk of the town. Give everyone a couple more seconds. Okay. One third are evaluating options. About twenty percent, know they need to move but haven’t started yet. And I expect they’ll go to evaluating options pretty quickly. So that means probably almost half. Right? I have a handful, ten percent, already in my mid migration. I feel your pain. Methodical. Methodical. Methodical. Visibility, transparency is going to get you through, where you can have an automaton that can do some of this, rigor testing for you, will help you move faster. I found Ring deployments work really, really, really well. And then you do wanna pay attention to that kind of day zero, day one, day two model where day zero, is pictures. Day one is brand spanking new, infrastructure. And day two, people have moved in, and now change management becomes a problem. So, keep an eye out for that. Use your rings. Pay attention to how performance goes as you fill those rings. And then for the folks that are currently satisfied, awesome. Right? I think it’s important for you to understand if it moving is, the right decision. Hopefully, those assets that you can download will help you show that. Look, folks. I know you really want your cloud first mandate, but it’s gonna cost you another million dollars a year. Is that what we’re saying? Oh, it also means we have to change all our security tooling, and we have to change this, and we have to change that. Do you want that kind of upheaval right now? It’s always good to know even if you don’t go. Right? Awesome. Okay. Poll number two. Feel free to ask any other questions. What are we worried about? What is your single biggest migration concern? For those of you that are migration or planning a migration, for those of you that are happy, namaste. But for everyone else, are we worried about disrupting users? Are we worried about unpredictable costs? I mean, these are all valid app compatibility issues, skills gap, or compliance and data residency. Alright. Currently, fifty fifty hidden and unpredictable costs or app compatibility. Give a couple more seconds here. Alright. Last call. Anyone else wanna weigh in? Now is your time. Okay. So we end up with a quarter of the folks are worried about disrupting their users. We’ll touch on that in a second. A quarter are worried about hidden costs. Yeah. Yeah. I would be too. App compatibility issues comes in at fifty percent. The good news is that it’s all mitigatable. Right? And we’ll talk a little bit about that while we have the screen up. First and foremost, disrupting users, that’s the canary. Right? We use the canary to make sure. And, ideally, you can loop that canary so that you’re getting a regular pulse of that wave, that migration set, what have you, so that as soon as you feel it change, you can roll back if you need to, or you can pause. Hey. We got half of them in. We can fix the capacity next week, but they’re good for right now. So you absolutely with with what you have here today can mitigate that risk. Hidden and unpredictable costs, that’s the big challenge. Right? That is the big challenge. I’m gonna address that one last. App compatibility issues and legacy dependencies, testing, testing, testing. Right? Ideally, automated testing, and, ideally, automated testing that gives you the same outputs every time. Right? So it’s not like Joe tested it on Tuesday, but he was late for a date night with his wife, and so he kinda cut some corners. Right? Or, you know, Michael was supposed to get in there Saturday morning at six when we finally finished that that monster of a change, but he never even logged in because he was at a soccer game watching his kid play. Right? Whatever those are, we do want some consistency. So automation and consistent reporting is really gonna help you with those app compatibility issues. You know what they are. Right? You know what your top line applications are. Start at the biggest. Start at the noisiest, and not even every step. That’s a daunting task for anyone. Start with the steps you know cause the most problem. If you don’t know, talk to your support team. They know. Right? They know when it when you did the last upgrade, how many people complained that Salesforce broke or that this broke. Or now in Excel, I can’t do this or I can’t do that. That plug in that we added to Outlook is causing all kinds of trouble. Right? Whatever that is, they have a a rank stack of tickets that will help you get in front of that so that, again, as you get toward the lower end of that, yeah, there are issues, but they’re not showstoppers. Right? Cover your showstoppers first. And then let’s talk about hidden costs. Hidden cost is a very real issue. Piloting, doing their rings, understanding and and testing early will help you uncover those costs. Do it small. Right? Start with a small pilot so you can see, okay. What is the total cost of ownership of this now that I have a proper configuration in place? Do it for a subset five users, something like that, and let it run for a billing cycle. That will help you uncover those hidden costs. Alright, ladies and gentlemen. Thank you very much for spending your time with me here today. We will have a migration part two, spotting those gaps before they come become problems. You should have the I will post the the link right now in the chat. Give me a second on that one. We have people you can talk to. You we have partners that you can work with if you’re stuck or if you’re worried. You have some resources, hopefully, now that you can use to to put together a good plan, have it be well reasoned and shareable with a larger audience, run your migration readiness scorecard this week. And if you need a deeper dive into anything that we talked about here or how we can help you, we have a bank of resources. Pull up your phone. We also have an ROI calculator that is, something you can look at. We can also do a customized demo of what our platform is capable of and what it can do for you. So feel free to leverage that and use us for for that piece. Richard, absolutely give us a give us a shout, and we could take you through what we’re doing with NVIDIA and Nvector and discuss how it can help you or if there’s more that you need from the product. Are there any other questions now at the time? And if not, I thank you, and let me get you a link really quick. And in the chat we go. Hopefully, everyone can see that. It’ll definitely be part of the video, if you will watch this on demand. Again, GitHub dot com forward slash login v s I, no spaces, forward slash migration underscore assets. Migration underscore series underscore assets. Thank you very much.

Good day, everybody. Thank you for joining us for the migration series session two, spot the gaps before they become problems. We’re gonna hang tight for two minutes and let some folks file in. We’ll be back shortly. Okay. Let’s go ahead and get started. Again, thank you very much for joining us. This is the second part of our migration series, really focused on some of the gaps that you want to identify before you go into a migration effort. There is going be a part three we’ll talk about at the end of this presentation, where we go through hands on moving something from one side to the other, just to help everyone understand what entails accomplishing the move. A little bit of housekeeping. My name is Michael Kent. I’m the chief product officer for LoginTSI. I’ve been here six years. I previously worked in a migration company, so migration, unfortunately, is the thing we never wanna do, but often have to do, and I’ve done a lot of it in my career. A little bit of housekeeping, we’re gonna try to keep this to forty five ish minutes. We do have some Q and A on your screen. You have a Q and A button. You can pop those questions in at any time. We won’t answer them until we get to the bottom of the presentation. And of course, there’s a chat window if you have anything you want to pop in there. We will talk about some platform specific evaluations. I do have some assets that we’ll share with you as well. Just like we saw in the first of the migration series, they’re all in the same repository. So if you have that from the beginning, it still works and we’ve uploaded the new documents to that. If anyone has trouble, just let us know. And let’s talk about what we’re going to see here today. So platform specific evaluation for Citrix and Horizon shops, if you’re moving, some things you have to consider that are specific. I also threw in some physical to cloud, so that’ll be in there as well. Really, the architectural decision framework, it’s the framework that you need to focus on. The data points will change, they’ll update, new options will become available, but it gives you some framing around how to make that decision. We talked through the migration tool chain. I expect that there’ll be some questions there. Real world deployment patterns, so we’ll talk through a few of those. Where we see migrations stall, and then we’ll go into Q and A. Before we get started, let’s talk about where are you in your migration journey, just to kind of level set so later when people ask questions, we’ll have an understanding of kind of where people are at. We’ll give you a thirty or so seconds to pop an answer in. All right. Anyone else wanna participate in the poll? Okay, so it looks like most people are still evaluating, that makes a lot of sense. Some of this technology is very new, and stepping into it cautiously is always a smart choice. Even they’re committed to migrate, I understand that too. And and without saying it out loud, we all know why why that is. The world has changed, and it only takes a couple of acquisitions to really put some new pressure on an old decision. Right? So, we feel you. We hear you. We see you. Let’s talk about how, again, we can get you to the other side quick and clean. So we’ll start with some differentiated migration paths, something that is probably more specific to your platform wherever you’re starting. If you’re coming from Citrix DaaS or CVAD, there are a few things that you want to pay attention to. We covered a general framework in in session one. If you haven’t seen it, it’s on replay. You can grab it. But there are a few things that are specific to Citrix that we wanna pay attention to. The NetScaler dependency is a real thing. It’s not part of Citrix. It is part of Citrix. That’s been its own roller coaster, but it does play a role, and it plays a role, especially if you’re doing the SSL offload or the ICA proxy. It plays a role in performance. There is no real direct equivalent in AVD or Windows three sixty five, and so you have to account for it. Whether you have to have software based load balancers or terminate those certificates a different way, what have you, there’s two things that really jump out, and that is protocol. Nothing you can do about that. Microsoft has done a great job around protocol efficiency, but some of the load balancing, some of the offloading that Netscaler does, as well as security, right? So it provides a security barrier, and that’s not trivial. Your policy migration, they have to change, right? You’re gonna be moving to some other policy system, most likely Intune, and again, not trivial, and you want to account for that early in your planning and early in your migration and in your testing. You don’t want to undersell that or understate that and then get caught in a lurch later on online. And then the profiles. Right? We have to move profiles into FSLogix. That is its own beast. Its performance is its own animal, and you want to make sure that you understand it, not just the performance once you’ve migrated, but what it’s going to take to get you into FSLogix, that whole migration path. There is time that you need for that. And again, this is an area where you’re going to want to know ahead of time, well, what are my profiles look like? Am I going to do any profile cleanup? If I’m moving them as is, how long does it take to do that move? And what does it take to to keep those up to date if I need some coexistence and and and that kind of stuff. The the insider note is really people tend to under classify the NetScaler dependencies. And so they budget a couple weeks, but anything, especially if you’re dealing with SSL certificates and network routing, these are different groups. They have different procedures. It’s going to probably take you a bit longer. Just understand it in phase one. Don’t wait to phase three. We don’t want you stuck waiting for something to be turned up. On this side, right, again, different technology. Persona management is different. They have to migrate to FSLogix as well. That size audit is super important in phase one because of the time it takes you just to see data. Workspace ONE UEM policies have to be converted. Again, maybe you have a lot, maybe you have a little, but there is a conversion that that you have to to manage as part of that. It’s not just the movement of the workspace itself. VM sizing, right? None of your VMware sizing is going to map directly to an Azure SKU. CPU memory ratios differ. Your D series, your E series are all optimized for something different. And I would even say be careful about the region because there will be minor, not major, but minor differences in how those systems run-in different regions, and you wanna know about it. I just ran into this this week. Provisioning AKS in a European region was really, really different than provisioning that same service in US East two, right? So you want to understand that ahead of time. As you’re doing that SKU validation is a great time to understand the density to performance ratio, and we’re going to talk about that. The last customer I did this with, we were able to make almost a sliding scale that said at a ten second log on, you’re using D8s. At a nine second log on, it jumps, right, because it’s a stepping function. It’s not a linear, nice, smooth linear regression. So you will have stepping functions, and then when you hit that step, you’re not optimized until you hit seventy, eighty percent of the utilization at that step. So these are things that you look at is because these SKUs have stepping functions and it’s not easy to slice and dice those CPUs and disk IO and GPUs, etcetera, you do have to pay attention to the steps. Okay. And then there are some differences if you’re going physical, and some people do. If you have a lot of remote work or if you’re pulling in Teams from other time zones, the world has moved away from shipping laptops. Right? It was it’s expensive. Getting hardware is expensive. Maintaining the hardware outside of your geo is very difficult. So now it’s use whatever you have and will bring you in through a virtual desktop. So there is a physical to traditional endpoint transfer. And again, no broker to retire, so so that’s not your problem. App behavior absolutely changes in the cloud for just the complexity of the stack. Right? So don’t assume that you’re gonna have the same behavior. It is testable. Right? It is something that you can at least set an expectation around, and we’ll talk about that a little bit. Then the security agent validation, going from physical to any EUC platform, whether it’s on prem or cloud, there is a dramatic difference in the resource scheduling of a shared environment. Even if you’re doing persistent, right, there’s still an underlying hypervisor that can get busy, that can shave off performance that you don’t really see, and it just it it comes back to you in slow starts or intermittent long IO streams, these kind of things, right? Then if you add a security agent on top of that, that was mostly coded and written and designed and tested for physical hardware into a virtual environment, the characteristics change dramatically again. So if a security agent is checking a file and it expects really fast IO and does some blocking, this becomes an immediate problem because the scheduling will slow down, the blocking will slow down. It goes from zero to untenable very fast if you’re not paying attention. So security agent testing specifically, you should pay attention to. You’ll prob you might have to go back to your vendor and say, okay. We we are moving to this environment. Do you have something that’s already optimized for this environment? Or you might go to your target environment and use something that that environment recommends, a Microsoft recommendation or something like that. But definitely, it’s something you have to consider, and especially if you’re going from physical because most people have spare processing on their laptops or their desktops. And the whole idea of moving into Windows three sixty five or into AV pooled, even AV persistent, is to try to share some of that and recoup some of that cost. So don’t over provision. If that’s going to be the concept, then these security agents really do leave a mark, I guess, is the best way I could say it. Okay. So I wanted to talk a little bit about a decision framework. Again, it’s not the end all, be all, but it should help frame your mind around the things that you want to consider as you drive. So we talk about choosing your destination. Windows three sixty five, and there’s flavors inside of Windows three sixty five, right? So you can share that desktop for up to three if you’re doing shift work, etcetera. But it is persistent. It is always on. It’s a flat fee, which a lot of people really, really like. Very low ops burden, relatively compared. And then probably not good for elastic scale. Spinning up, spinning down, you know, quickly, I have four hundred students coming in. They’re gonna be here for a week. They’re all interns, and then they’re gone. Right? That elasticity is probably better suited for another service. Pooled, Knowledge Workers, of course. You can get Knowledge Workers on Windows three sixty five. It’s my primary PC here. I run on a Mac, but my product is mostly Windows, so I use a Cloud PC every day, all day. I know several people on my team do as well, because we have to test and verify everything that we say on a Windows system. I’ve tested some of the lower end SKUs. I’ve tested I’ve had to work on some of the lower end SKUs for for some banks and whatnot, and that’s how they bring in their contractors, and you definitely wanna know what they’re good at. So at the low end, you’re getting browsing. So if everything in your life is SaaS, woo hoo, you’re a great space. As soon as you start layering on the Office Suite, then you’re looking at a bigger SKU. And if that’s all they do and they just crunch numbers, great. As soon as you start dealing with specialized plug ins and whatnot, you might need to move to either AVD Personal or AVD Pooled. And then also, this might be a preference, Right? If users don’t like starting fresh every time, right, don’t like waiting for everything to pop up, right, I I have one cloud PC that nothing to do with the cloud PC, but because of all the security and all the rules that they layer on top of each other, I sign in and I go get coffee. And the idea is that by the time I come back from coffee, Teams has finally opened, right, kind of thing. And, it’s just all the layering of all the different pieces. It’s not Teams per se. It’s just all of the policies and security agents and everything else that it has to run. That might skew you toward a personal. And then definitely use, right? GPU workloads, people that need to be able to get in and have their environment exactly the way they left it, developers, absolutely, designers, the artsy people in the world. I say that with air quotes because I’m one of them. Traders, like, if they’re trading on the floor, they want as little friction to getting into those work spaces to get effective quickly. Right? Seconds matter in some of these. Right? And then the relay the the realistic notion that there may be workloads you just can’t move, just they’re too sensitive. Right? They have compliance requirements that are just very difficult to either solve or evidence in a cloud environment, and, you know, the time it’s gonna take you to go through and and get all the answers, you want to get a bunch of other stuff moving. So there is something to be said about keeping things on premise and keep that as an option, an arrow in your quiver, as we like to say. Do have, again, as I mentioned, we do have some assets available to help. So the deployment pattern comparison is in the GitHub repository that you’ll get as part of this. We also have Cloud PC Troubleshooter and some other assets I’ll share toward the end of the presentation and walk you through and leave behinds, right? Leave behinds. As we look at our architecture decisions, probably the big one, I put that note down there that SSO on Cloud PC is not magic for a reason. We run into it all the time. You’ll find it all over the forums. You’ll find it all over Reddit. We see it all the time. PRT, PRT, PRT. That primary refresh token causes so many challenges as it moves through the different components and pieces of your environment. So you wanna start there. You wanna start with identity, and you really wanna pay attention to every time you take an iteration around that identity. And I’m talking about conditional access, talking about MFA policies. Right? So, you you said an MFA policy is as soon as you change a duration or a session time out, you’re gonna be challenged to make sure that PRT works. When you change conditional access policies, same thing. If you switch from SAML to OID, you’re going to have the same problem. And I promise you that through this process, you will have an idea and you’ll change that idea. It’s just the evolution, new features will come out, and so you want to pay attention to your identity first. Network architectures, this is as much getting back to resources that are on your premise. So, okay, I can move the workload out to the cloud, but the database is still on prem, and from a compliance perspective, that’s the easiest for me to evidence with my compliance ratings, etcetera, it’s going to stay for now. So we want to make sure that we get the network architecture correct, and this is about the time I tell a story. I worked for a medical device manufacturer, and pandemic hit. Everyone ran into the office, grabbed their laptops, and ran home. They were a critical device. They did not see a downturn. Patients still had the same problems they did, so there was no effect economically on the company. They still had as much demand in the pandemic as they had outside of the pandemic. But the quarter after the pandemic hit, they had a huge financial shortfall, and they didn’t understand why. And it turned out to be that as people ran home and took their laptop and connected back through VPN, the first problem is the VPN was absolutely not sized for that workload, for having all of those people come in. And they tried a bunch of different solutions, finally ended up going back to a hardware based solution just so they could manage the capacity. Oops. So that was number one. Number two was that somewhere downstream, they were printing to a remote warehouse, and the printing, when you’re in the facility, the networking was such that it would print in five seconds. For a company that ships a product, I can’t charge for the product until I can get the invoice out. I can’t invoice you if you didn’t receive the equipment. So there’s a supply chain capability here. They found that when they were going through VPN after the VPN was fixed, printing to that remote warehouse went from five seconds to five minutes. The VPN routing was routing differently than when they were on prem. They just didn’t ship as many devices, so they couldn’t collect as much money, and the shortfall was in the millions. So when I say network architecture, this is what we’re talking about. Let’s make sure that we are measuring the process your constituents use to get to their business processes, and that we’re not making a network decision that bounces them all over or gives them a different path and slows that path down, affecting your line of business. Last here, I’ll talk about image strategy. It is a big deal, right? Ephemeral versus persistent, you know, you can battle that inside your company for years. Regardless, there will be some caution and some misgivings, and they’re gonna fear uncertainty and doubt. This is where good measurement ahead of time, what do they have now, and where you can show this is what you’ll have in the future, right? Here’s the differences, they’re pretty close, Is is gonna really help you get through that migration with the least amount of complaining and and cruft as possible. We know that people are much more tolerant when you can set an expectation and justify a decision. Hey, it’s not your laptop that’s blazing fast because it’s a gamer laptop and it has a GPU and it’s going crazy. But realistically, that’s overkill for you using Salesforce, right, kind of thing. This is a bit more right sized for Salesforce and you’re in sales, you don’t update Salesforce anyway, so what are we really complaining about? That’s also a joke for all my salespeople out there. So definitely, you have to pay attention to that PRT. We do have in the assets a Cloud PC troubleshooting kind of a branching tree that helps you understand where PRT might be bugging you. Okay, let’s talk a little bit through the tools that you’re going to be using to get you from point A to point B. When you’re doing your assessment, and we’ve learned we’re going to do the assessment, The saying is that if you fail to plan, then you plan to fail. Right? So let’s not let’s not stick our foot in it. Let’s make sure that we’re doing the planning. So you can use Azure Migrate to help you do your assessment, you’re going use your DEM tooling if you have it to do part of that assessment, and of course, Login Enterprise is designed to do exactly that. Look at where you are, run something that is not EUC specific. You can run it on hardware, you can run it in EUC, you can run it in cloud, and just capture the current state. How long does it really take for my users to log in? How long does it take for my business applications to fire up? How long does it take my finance when they click that plugin to load them the Monte Carlo diagram, right, or process a hundred thousand lines of invoices and build the charts. Understanding that before you do your move is super important. Your design, right? Hydra is a huge help here. It helps you not just design your, you know, figure out what you want to do, it will also help you move into it, build them out, test them, destroy them, build new ones that are better, etcetera. So your iteration in EVD, hides as your friend. Entre ID, we have to make sure that we’re getting our identity correct, and that includes, you know, hybrid joined, etcetera. Right? Every step that Entre has to go through and every network connection that Entre has to go through because of sync or because of something is a challenge to PRT, and we wanna make sure we have an efficient path. Or that we made the adjustments to the token length, etcetera, so that we’re not getting caught with another authentication prompt. Nothing will kill your momentum like a repetitive prompt asking you the same question and not figuring out why it can’t get through. Azure Network Watcher, again, this just goes back to, are we overloading some link and not realizing it? As we know that the workload characteristic is going to change, no WAN link is as good as a LAN. Right? Oh, I shouldn’t say that. I’ve seen some pretty bad LANs, but the idea is the chances are that local is faster is high. Right? So we have to account for that to keep our latency and our access times in line with what the applications need if we’re moving people into a cloud and expecting them to come backward onto prem and access resources. That should just make sense. Design. Compatibility. The packaging, repackaging, the shims, whatever it is you need to do. We’ll go into that in a little bit. Getting a passfail, just a smoke test. All right, I went through some repackaging. I moved this application. I dropped it into an AVD configuration. Does it work? Does it have DLL dependencies I didn’t account for? Does it have something did the packaging work? And most importantly, outside of passfails, is it performant? Right? Does it take thirty seconds to load? These kind of things. That’s also part of pass fail. And then your FSLogix migration tooling. You’re removing these, so you want to make sure that MS Logix is in good shape and can handle the load that you’re about to put on it. That takes us to execution time. Again, as you’re moving into AVD, again, hydrate your friend, whether it’s provisioning, image management, profiles, policies, etcetera. Giving you visibility into what’s happening helps you make decisions near real time to get you into and out of problems, gets you to and through. The OneDrive movement tooling, so you will have to plan for that move and how long it’s gonna take. So that’s one of the tools that you’ll you’ll pull in, you know, how do I get all my OneDrives populated and ready to go because we don’t want those calls coming back to anything on premise. And then, again, Login Enterprise for performance. When I’m using this application, is it laggy? Did a business process that used to take five seconds turn into five minutes? That’s a big deal. Right? Am I getting fifteen prompts as I go through this process? What used to be single sign on is now single sign on, accept a certificate, single sign on, we couldn’t find you, single sign on, retry, single sign on, one more, single sign on. We wanna make sure that we smooth out all of that before we subject our users to it, really, right? And that’s all part of this migration tooling. You have tools to help you understand that. Now, the world today, nothing is static. Everything is changing, and it’s changing rapidly. We come out with new tools. Hydra comes out with new versions. Microsoft comes out with new versions. Your app vendors come out with new versions. Your internal apps come out with new versions. It’s a lot of new. Any time there’s new, you have a change management issue, right? Great, you moved a population of users over there. We want to do two things as we operate this environment. One, we want to make sure that no second wave or third wave makes performance for our first wave worse, and we want to make sure that as we do these changes, we don’t end up bumping into a compatibility issue because it worked on my machine or it worked in one environment and not another. So we have a way to actually determine that with Login Enterprise. Hydra can help you with your policies and your governance, and then of course you still have DEM tooling. I know Microsoft is working fastidiously on their endpoint monitoring. DEM tooling has a head start. Just to be honest, it’s probably something that you’re going to have to hang on to for a little bit while Microsoft catches up, and that’s just the reality of it. Okay. From an assessment compatibility and validation, if we’re in phase one, we’re probably spending our time in Azure Migrate. If we are moving from phase one through phase five, we’re gonna keep Login Enterprise warmed up. Login Enterprise is going to help us baseline where we are today. And as we take any of these phase one through phase five steps, check and compare. Hey, we just repackaged the app. Let’s run it on premise. Does it work as good as our old installer on premise? Yes, check NextGate. Does it work Windows three sixty five? Does it work in AVD pool? Does it work in ADVD persistent? Then you have a decision matrix. Okay, it works in all these areas, which is the best for us, which is the best for the user, which is the best for the company. Knowing your options is huge. You will have applications that will break that, applications that were single threaded. So no matter what you do from a scale perspective, you’re limited. It will never take advantage of the CPU cores that it has. It just isn’t built that way. These are things you want to know through that Phase one through Phase five. Your MSI packaging, we already touched on that. Hydra, okay, I have it over there, how do I move things around? I don’t want to have to drop down to the CLI, pull everything out, rebuild everything, or more importantly, I’m no longer at zero. Have people who have moved into the apartments. I can’t do these kind of swings. So how do I manage an existing environment that my Wave one users already moved into, but I need to make some changes to support my Wave two users? FSLogix, can’t stress that enough. As you start putting data in there, its performance will change, and you want to know what that means and what that looks like. And then, of course, as you’re in that phase three, prepping everyone for that that big move, that pilot move, having your OneDrive to cloud files conversion, you’ll be working pretty heavily in that. Okay, next poll. What is the biggest platform specific challenge to date? Don’t be shy. Even if you’re planning. If you’re planning, something’s keeping you up at night, hopefully it’s on this list. Ah, yes, yes. Okay. Here we go. Currently split. Let’s see if anyone else has an opinion on the matter. And it’s neck and neck. Anyone else? They’re battling for first position. Policies. Alright. Ten seconds. Yeah. Good. I would I would say there there we’ve stressed the importance of the NetScaler. Right? Maybe you don’t have a NetScaler, maybe you’re you’re you’re okay with that. Just one point validation, you know, that’s actually pretty easy to to deal with. I ran a migration project for three thousand users to ABD, and at the very, very end of the project, the team that I was working with said, well, how does this compare to what my user like, my on my laptop? And and I said, well, the best part is we don’t have to guess. Right? You can run this piece of Login Enterprise on your laptop, and it will answer that question for you. We ended up getting pretty close. I think the difference was one point one seconds or something like that. And again, they had the ability to say, well, if I can handle one point two seconds more on the login time, I reduce my cost by sixteen hundred dollars a day. So these are real numbers. Right? These come down to real numbers, and understanding what that breaking point is is super important. Profile data migration, no doubts. Policy translation, I’ve never run into a whole lot of trouble there. I’d be more worried about capability. Can I get as granular as I need to get? But you have a lot of tools there, and ultimately, it’s iteration and testing, iteration and testing. If you have a lot of policies, that should keep you up at night. If you have a lot of data in profiles, that should also keep you up at night. Data profile migration is probably less complex overall because you’re you’re moving data from one place to another, yeah, there’s always nuances, and And I’m sure no sooner do I say that than someone says, You forgot about this. But policies, because it’s walking through a set of configurations, you have a parity issue for sure, and you have just a sensitivity issue. And then when you go from on prem to cloud, latency can play a factor in there as well. So very good data. I would not sleep on the app repackaging. It always sounds easy. It always ends up you will end up with more exceptions than rules I have found, so don’t sleep on app repackaging. Okay, back to the slides. So now let’s go through just what we see as some of the real world patterns. Realistically, if you look at phase one, and I don’t know why my slides are compressed there, I’ll just talk you through it. Phase one is two to eight weeks, and what we’re really looking at is assessment and architecture. You wanna have a basic mapping. It will most likely change. You’re gonna iterate over it. You want to iterate over it because as soon as you commit, Microsoft will come up with a new advanced feature and you’re going to want to incorporate that. So build yourself for that flexibility and for iteration. Phase two is three to six weeks, and here you’re really looking at application compatibility. It’s a lot of times for a lot of customers, the long pole in the tent, a lot of apps. I gotta get through a lot of apps, and then the profile prep. A lot of data needs a lot of prep. Phase three is the wave migration and coexistence, exactly what does it look like. Find your sacrificial lambs, your guinea pigs, your canary testers, the people that are going to go through all the bumps to make it easier for everyone else. Who’s got the machete and whacking their way through the forest? That’s your phase three. And then step four is really that ongoing phase where we’re looking at steady state, very important steady state, and then safely decommissioning those things that we can start pulling out. Once we clear capacity, we can start pulling out infrastructure, etcetera. We do want to make sure we keep our eyes on it. The number of times I’ve seen a migration, even from a data center to a data center, we moved it and it was too bad, we had to move it back. And we moved it again and we still didn’t get right and we had to move back. We don’t want that to happen. So my recommendation is you put continuous monitoring in first, first thing. And of course, it’ll look beautiful. There’s no one in there. It should be happy and healthy and just green across the board. Then as you add five users, you want it to stay green. As you add ten users and twenty and then one hundred and then one thousand, you already have your litmus test. So if you hit a wave of one hundred and all of a sudden you see your previously green dashboard start to turn red, we know we’ve hit a limit, and we can go back through design. Is my PRT taking too long? Do I have a network issue? What’s going on? Before you run into a big issue, right? And you can step it up from that. If you use the tooling right, you can move with a great deal of confidence because you’ve checked and double checked and triple checked, right? Okay. If you’re doing your lift to Windows three sixty five, we have some specifics there. Understand why it works. Let’s focus on what makes the pattern hard. Data migration makes it hard, policy translation. These are the two worries that everyone had. App behavior validation. Yes, yes, yes. And then user expectation management. That comes because of the app behavior validation. When I know what expectation I can set, I can then present it to a set of users. And when they say that’s not good enough, I go back to the drawing board. Or I pull in my executive sponsor and say, look, we have a cost for performance challenge. How are we going to deal with this? Also, why you might have that that’s a harder group to move, so we put them at the end. Let’s do the eightytwenty rule. Let’s get the easy stuff done first. Deal with the hard stuff later, right? We hope the technology will catch up, and maybe there’s some cleanup or some other activity that can make that better. Definitely get lowers the noise so that we can dig into that group, really understand what makes it so tough, and come up with solutions. Hybrid end state, right? Things that we have to pay attention for. Governance challenge and hybrid architectures. Well, great. I have all these policies, but they look different between what I have on prem and what I have in the cloud. How do I marry those? I’m still using this technology. It doesn’t quite translate to this technology. And now my infosec team is mad at me, and my compliance officer is, like, ready to jump off the roof. Right? So we try not to stress these people out, but it is a real thing that you have to pay attention to and a real thing that you have to manage between. Governance is probably the hardest part in a hybrid scenario. And then, of course, all the dependencies. As I mentioned, when you’re in a hybrid, it’s more complex. Troubleshooting is harder. Everything is more difficult. Still think you want to focus on what you can move cleanly, but yeah, hybrid is a you don’t want to be split in two. It’s like having your foot in two different canoes, and they’re going in different directions. It comes a point where it has the potential to rip you right in half. And so try to minimize it and try to be in that hybrid state as little as possible or reduce the impact of being in that hybrid state. Right? So if I have fewer critical workloads that are in a hybrid state, I can give them more love and attention and not take on a huge administrative burden. Ah, yes. Where migrations stall. The easy answer to this is when user expectations are not met, If you’re not setting user expectations, you can’t meet them. App compatibility is a huge potential problem. Profile migration, huge potential problem. For obvious reasons, app compatibility, how do I find people to to do the application compatibility? What are all the nuances? Do I what are the dependencies? I can’t have a compatible app until it’s been converted. It has to be tested. And we all know that UAT is somewhere between the hardest thing to do and some sort of fable that we’re told exists but we’ve never seen, like a unicorn. Right? UAT can be very, very difficult to track down. Again, from a very simple program perspective, it’s got to be formal. Right? We we make the joke is that it feels right or it feels slower is not a criteria. It’s it’s not a measurable thing. So you always wanna punch it down to something that’s measurable. Feels slower well, wait a second. Right? How long does it take? Time. Time is measurable. Well, it takes the same time. Yeah. But it feels slower. Okay. What part of it feels slower? Well, the screen doesn’t update as fast. Okay. But the application’s there, you can click a button, you can enter an order, you can check-in a patient. Yes. This is how you will set your expectation to make sure you meet an expectation. And as you layer on changes and layer on technologies, those might change and you want to know, oh wait a second, I have to set a flag. You just turned on the security agent and checking in a patient went from ten seconds to thirty eight seconds or two minutes or what have you. Doctors need to pull up an x-ray in a minute and now it’s taking five, right? Doctors are mad, management is mad, patients are mad. We don’t want that, right? Or it might even be critical care, right, where speed is life, right? So we want to make sure that as we layer on these changes and we do these moves, we have something, the canary in the colon, that’s telling us, Woah, I don’t know what you did, but it is not better. It’s worse. And then we can track it back to a change, right? So we’ll start with triage. Simple. No go. Does it work? Yes or no. Then does it meet some predefined thresholds? Even if they’re new thresholds for your future state, you still have to measure them. Right? You’re not gonna get a one second log on time. It’s gonna be a ten second log in time, but we still have to honor the ten second log in time. Because if that hits twelve seconds, yeah, it feels slower, and we’re wrong. We we we did something. We either have to go reset an expectation, or we have to fix a long log in time. You know, you get the point. And then remediate. So as we get tier one, things either work or they don’t. Great. Move them on. Tier two, do they meet their thresholds? Yes. Move it on. It ends up being a funnel that’s larger at the top and smaller at the bottom. Right? Everything goes in the triage. Some things get to step two, some things to step three, some things to step four. Right? At step four, you move on to the ones that can’t get past step four. When you clear a bunch of those, you go to the test, the the test step. Right? What can’t meet the latency threshold? And then upward. Right? That way you get an efficient funnel through an actual program. And the other important part about a compatibility program is you can get help, right? It’s not subjective, it’s objective, and you can get help. You can fleet this. I have people I can give this back to my app owner and say, your app is not going to run well. Go check with the vendor. Go check with the software developer. And you can delegate some of this because you have a formal program and formal expectations. Okay. The other one, of course, is migration, profile migration. We want to make sure it has some dedicated ownership because it’s its own animal and because it’s one of the longer transitions, you have to have someone that is sitting on top of it and paying attention to it. And I know there’s not enough resources for it, and I know it’s not easy. But such is the work that for some reason we all got convinced we wanted to do, right, IT work. So keep it in the back of your mind. Explain to your management team, look, this is this is a very sensitive topic, and we can’t just expect to to do it at the last minute. It really has to be a first class citizen in our work stream, in our thought process. Let’s make sure we give it its due. No webinar would be remiss without some capabilities to help you along your journey. Again, Login Enterprise for both baselining and comparing. The application compatibility test runs are super important. You’re going to have multiple changes. Let’s make sure we understand before you have a large impact what’s happening. That includes everything from a network change to a policy change to a new security agent, a new version of the application. If I had a nickel for all the times, a new version of the application that was supposed to be better was delivered, but only if you configure it a certain way, I’d be a rich man. So we have to pay attention to that. And then again, as we’re moving in to AVD and you decide, oh, this might be a better workload for Windows three sixty five, Hydra’s Management Console can help you there too. It can help you with your image management, it can help you with your policies, and it can help you with your migrations. With that said, I’m going to share my screen and walk you through a few resources. We do have a third series where we actually do execution, right? So stay tuned for that. You’ll be able to sign up for that. And while we do that, I’m going to quickly share my screen and walk through some assets that are in the link that you just saw. Okay, so first one, we have the deployment pattern comparison, different patterns, some decision factors to help you understand where you might want to use your users, deliver your users. You will customize this to you, your opinions, your thoughts. We give you something to start with, but you’ll want to validate this and make this your own. So we go through usage patterns. We go through horizon. Are you going hybrid? What are the architectures? Etcetera. So that’s one asset that we have available to you in the GitHub repository link that you just saw. We also have that toolchain reference card, Azure Migrate, Login Enterprise, the Dem tools if you have them, Hydra, what it does, what its deliverables are, Azure Network Quadrant, etcetera. So that is all here for your reference as well, bi phase. So this is another asset that we’ve put together for you along your journey. The third one I have for you is a Cloud PC login prompt troubleshooting kind of branching script, kind of a what’s my question, what’s my possible answer. If a prompt appears before the desktop loads, look in this place. If it’s after locking or reconnecting, check here. If your prompt is inside a session, check here, etcetera. So it goes pretty deep into it. I hope this is incredibly useful for you because there’s nothing like moving someone having a surprise in a fire drill. So let’s hope that this helps put out fires. Ideally, you don’t have them because you’ve pretested, but something changes, and that’s another joke I tell. I tell other IT people, you know, nothing ever changes, and it’s never the network guy’s fault. We give a lot of grief to network guys. The last item I have for you is platform evaluation worksheets. If you’re going from Citrix on, and I realize that’s really small part, we just give you some criteria, right? Some things you can look at, the management layer, here’s your risk level, what’s your status. You can add, you can take away, This is yours, of course, to customize, but we wanted to give you something to get you started based on the things that we’ve seen other customers go through. It does have one for Horizon as well. So again, very small, apologies. But we do have some important key items for you to take a look at. Those four resources are available in the GitHub link, and I’ll stop sharing my screen. As I mentioned, migration series three is coming in probably thirty, sixty days. Keep an eye out for it. Register for that session. Pull down your session two resources. You can also grab your session one resources. Book an environment assessment. Let us get in there and take a look and and help you out, and or start a platform evaluation. Right? So we have the ability for you to start these evaluations and at least get used to these tools, understand what they can do, ask questions, and get some of your answers ultimately. Some additional The presentation will be available, but we do have some resources available, so free training on our platform, logging in VSI products. We have a Slack community that you can join, ask questions, get answers. We do have an ROI calculator, I should say an ROI study, right, where we talk through how users are finding value with these platforms and products. And then templates, so we have our own we curate our own set of application templates. These are things that we’ve collected over time, thirty or thirty one of them, that really go into, this is how I test this application, this is a reference architecture for here, here’s some things to expect, resources that you have to help you down the road. And we can open up for questions. Anyone still with us, there’s a Q and A button on the right hand side. Go ahead and click it and you can pop your question in and we’ll do our best to answer it. No questions means I’ve answered everything so well in the presentation that it leaves nothing to the imagination. If you have something, please, please, please let us know. We can hang on for another minute. If you’re just too shy, then we’ll just say thank you for joining the webinar, and feel free to you will get when you click out of the presentation here, we’re going to take you to our EUC playbook. If you have not already downloaded that playbook, it is an incredible asset in conjunction with migration webinars that we’re putting together to help you understand and go through your migration process. Super valuable asset. Highly recommend. Highly recommend. And with that, I’ll just thank everyone, and enjoy the rest of your day.