 Hello, everybody, and welcome to this video. My name is Sarah Lean, and I'm a senior cloud advocate at Microsoft. Today, I want to talk to you about cost savings. Now, when you're running your environment in the cloud or in a hybrid structure, you have to think about cost management. You have to think about how you design those solutions and how that impacts the cost of them. And also, if you're tasked with trying to save some money, you need to know where to look. In this video, I'm gonna talk about 10 things that you can use or do that can help save some money for your environment and your organization. So let's dive in. So the first one is Azure Advisor. Now, Azure Advisor is a free service that gives you a host of information about your environment. It can tell you how to improve your security posture, how it can improve your performance posture within the environment. And the one that we wanna focus on is cost savings. Azure Advisor will take a look at your environment and suggest some cost saving measures that can be implemented for you to save some money. Now, Azure Advisor is turned on by default for every Azure subscription. So it is there for you right now. If you've never looked at it, it will have a host of information and it will give you some recommendations that you can look at implementing or following. And as I said, Azure Advisor itself is free, so it's something that you can utilize again without having any further impact on your cost management. And it could potentially save you a lot of money. Now the second one is right sizing your workloads. On-prem, I was probably always guilty of over sizing some of the workloads or servers that I managed just because I could. I had the capacity and I didn't want all those phone calls coming through from people saying that things were running slow or there was some issues performance wise. So I always just gave them a little bit more. Now, that's probably not practical when you're running workloads in the cloud because you pay for what you use. So if you have a workload that is much bigger than needed, you are paying for that and not really getting the benefit of it. So I'd always say to have a look at right sizing your workloads. Azure Advisor is something that can help you understand those right sizing and getting the sizing right when you're running it within your environment. And also, if you're migrating workloads and servers from an on-prem environment, if you use something like Azure Migrate Server Assessment to understand how your workloads are running, when you actually migrate them into Azure, you can right size them and have their appropriate size attached to them. And again, this will save you some money because you're only using the compute, the storage, the memory, whatever it may be that you actually require for that workload and you're not running something that's too large and not getting any benefit from it. Now, the third thing that you should think about is Azure regions and where you deploy your workloads. Now, I don't think it's any secret that some Azure regions are cheaper than others and others are more expensive than others. And this is just due to the logistics of running those Azure regions and Azure data centers. In some countries and some locations, labor costs for having some people to man that data center or electricity costs are more expensive than in other places. Now, for a lot of organizations, they don't have the choice of what Azure region they can use. They're kind of limited due to compliance needs and they have to pick the one that's maybe in their country. If you do have the flexibility to have your resources in other regions, then have a look and see if there's maybe one that you could potentially use that is cheaper for maybe your development environment or a testing environment. However, it's key that you have a look at how that would impact some of your egress and ingress kind of costs and how that would work within your infrastructure and the management overheads that that would give you as well. But certainly it is something to think about. And again, when I said at the top of the video when you're architecting your solutions, you should have a look to see if there are some things that you could change architecturally that will save your organization some money. Number four is Azure reservations. And before those were only applied to virtual machines. So they were only there for you to take advantage of in terms of virtual machines. But now they're available for much more services, things like Cosmos DB and storage actually have Azure reservations. Now an Azure reservation is you committing to a certain type of resource or a certain amount of resources for a set period of time. So for example, if you know you're going to be running a certain virtual machine for a year, then you can say to Microsoft and the team at Azure that you're gonna run that resource for over a year and they will look at giving you a discount for actually running that service and committing to that service for that year. So again, you can have significant cost savings for that. And like I said, it's not just virtual machines that Azure reservations now cover. It covers a whole host of services. So please do check out what they cover nowadays. And again, how you can apply them within your environment and potentially get some massive cost savings. Now the fifth thing on my use is hybrid use. And this used to be called Azure hybrid use benefit. And what this is, it's allowing you to take advantage of licenses you've maybe previously invested in and taking them up into the cloud. So if you're running say Windows Server or SQL Server on-prem and the licenses you purchased on-prem had software assurance attached to them, you can now tick a box when you're deploying that virtual machine in Azure or SQL Server in Azure and you no longer have to pay for the license because you've already paid for it with that on-prem software assurance license. Because when you run a virtual machine or a SQL Server in Azure, you're paying for the license, the compute, the storage. And if you have already paid for that license, you don't wanna pay for it a second time. So definitely have a look at your licenses and see if you can have software assurance and you can then apply that up into Azure. And again, saving yourself some money in the process. Now we also have the hybrid use program for some Linux versions and some of the Linux subscriptions. So definitely check out the documentation if you're running some Linux versions and see if this hybrid use can actually be applied to those as well. Now the sixth thing on my list is spot virtual machines. Now what spot virtual machines are is they're allowing you to take advantage of some of the unused capacity within Azure data centers. The last thing the Azure team want is any hardware or anything sitting idle. So they've allowed customers the choice of using Azure spot virtual machines. And that's you taking advantage of the unused capacity at a slightly discounted cost. And what they do is allow you to use that without having to pay the full cost of running a virtual machine. However, there are some disadvantages. If the capacity is needed again, then you will be evicted and your workload will be stopped and the other person will take that capacity. Again, you can actually manage how you want to run your spot virtual machines. So you can say you only want to pay a certain amount when you are running spot virtual machines and when the spot virtual machine price hits that limit, your workload will stop and you will be evicted. So it's great for kind of batch processing or dev test environments or labbing something and you want to keep your costs down. You're maybe not precious about what happens to the workload. If it stops running, it's not going to impact your environment and your business too much. It's probably one you have to think very carefully about if you want to run a production workload in. But again, it has its use cases and it's something that can save you some money if you actually apply it correctly within your environment and your architectural design. Now the seventh thing on my list is shutting down your resources when you don't need it. If you're running lots of virtual machines, then you are paying for them every minute, every second, every hour that they are running. And you'll be paying for things like storage, compute, and potentially the license if you haven't applied your hybrid use. And that can be quite costly for a lot of organizations. Now, some servers will need to run 24 seven, 365 days of the year. I fully appreciate that. But there are probably some servers that you maybe don't need or there's some things that could be maybe shut down. Say if your organization shuts down for an extended holiday period, if you know everybody stops and everything shuts down for two weeks in the summertime, you might not want to have your infrastructure running during those two weeks. So you can shut things down, you can schedule them to shut down and you can schedule them to start up when you need them and when you're all coming back into work. And that's a great way of trying to save some money because when a virtual machine is shut down, all you're paying for is the storage. So you're not paying for that compute and you're not potentially paying for the license. So it's again, another cost saving mechanism that you can implement within your environment and save your organization some money just by shutting some things down. Now, the eighth thing on my list is taking advantage of some of the free services that are running within Azure. When you are architecting your Azure environment and thinking about how you're gonna structure some things, it's advantageous if you know the costings of how our services are costed and how much they actually cost. But there are some services within Azure that can help you run your environments that are actually free. So within our, say, Azure app service, the web app service, there are free tiers that you can leverage within Cosmos DB. There are some free tiers. Things like Azure policy are free. Have a look at these free tiers and these free services that you can use and try and implement and make your environment better. Now, of course, with the free tiers or free services, there may be some limitations that might not be suitable for your production workloads. However, they are great for proof of concepts, for testing environments, for even just learning about how these services work and trying to look at them and understand them and how they interact with your environment. So definitely understand the free services and leverage them where you can within your environment because, again, they will help you save some money. If you're just doing a proof of concept, the last thing you want to do is blow all your budget on that proof of concept. And when you get the go ahead, you've got no budget to actually run the production version of the service. So leverage the free tiers where you can and the free services so you don't have to pay additional costs. The ninth thing on my list is Azure DevTest Labs. Now, this is something that can help you in terms of cost saving how you run your entire MIT department. Long gone are the days of, say, one of your developers applying a ticket into your service desk and asking for a virtual machine to be spun up or a separate test environment to be spun up on-prem and having your infrastructure team actually implement that. And it could take days or even weeks for that ticket to be applied, slowing down the process to whatever your developer needs that for. What DevTest Labs allow you to do is you can template certain environments, certain virtual machine setups, and then allow someone like your developer to go in and spin up that template, deploy it within Azure, test whatever they need to test or try out whatever they need to test. Then once they're finished with it and just gets destroyed, doesn't impact anything else other than their testing environment. And then they can go on with whatever process they were in process of doing. And it's great from that management point of view. It speeds up the whole process. It stops your staff having to do something from that they're working on, like a cool project, a new implementation of something to spin up a test environment that a developer might need for five minutes. It just gives that self-service approach for your environment. Also gives people out with your IT department access to Azure to, again, to spin up testing environments as well. So it can be a great cost saver in terms of the whole management of your IT department and how you implement things. And the last thing on my list is Microsoft Learn. Microsoft Learn is a great free resource from Microsoft that allows you to learn new skills both about Azure and other cloud platforms that we run at Microsoft. Part of the beauty of Microsoft Learn is free, but it also has sandbox environments built into the learning paths and the learning modules that we have. So going through these learning modules, you could be learning about new service in Azure. And when you get to the stage where you want to get some hands-on learning, what happens is Microsoft Learn spins up that sandbox environment in Azure for free, allows you to play with that service. And once you're finished with it, it shuts down and it hasn't cost you anything. It's not impacting your production environment. It's not adding on additional cost within your environment. So it's a great place to learn. And it's on demand. It's 24, 7, 365 days of the year. You can use that and train your staff and train yourself on how to use Azure without costing you any money. Hopefully these 10 tips and 10 things have helped you understand some of the cost saving things that you can think about when you're either architecting your Azure environments or your task with having to save your organization some money.