ArticlesBlog

Azure file shares and Azure File Sync overview

Azure file shares and Azure File Sync overview


Hi everyone. My name is Fabian with me here as my colleague Mike. We are both program managers for Azure Files and Azure File Sync and we’re here to talk to you a little bit about our product. So Mike, maybe you start us off with talking a bit about Azure Files as a type of storage in Azure and give us an overview. Yeah, absolutely. So as Azure Files is just a fully managed file service in Azure. With a couple of clicks in the portal or with a few API calls you can get a file share and there’s no VM to manage or anything like that. It’s just a fully managed file share. You can use that file share for anything you normally use a file share for so that’s like, a broad range of applications ranging from applications that need like hard drive level of performance all the way up to applications that need like all SSDs so you can choose between the different performance tears in Azure Files. Also, it’s good for like general purpose file sharing. So just information workers that want to store documents and things like that. You can connect to these shares from within Azure. You can connect to them directly from on-prem over SMB. Like reaching up into Azure to connect to your share. And we also have a hybrid mode where you can actually connect to your Azure file share through a local Windows file server. That sounds very interesting. So hybrid is sort of the term of the hour. Lots of companies start or have their file services on premises today and they’re looking at how do I modernize? How do I solve some of the issues that I have on premises today? So tell me more about hybrid. Yeah, so our hybrid story is really interesting especially if you already Windows file servers running on-prem, it’s super easy. All you got to do is install an agent on these Windows file servers, connect them up to Azure Files and we will actually sync the data between on-prem and the cloud. Now. The great thing about that is that your on-premise users don’t have to change anything. They still go through that Windows file server they’ve been using all the time. Active Directory still works. They can mount the drives they can connect directly. Whatever they want to do. And Azure Files, through a product that we call Azure File Sync, will actually synchronize that on-prem server with Azure Files and you can then turn your on-prem server into just a cash so you don’t have to store all the data on-prem anymore. You can grow in the cloud, as you need to and the on-prem you don’t have to grow your footprint anymore. [Fabian] I think that’s very important: fixed storage footprint on premises just storing the hot data that’s really relevant on-prem is a very important aspect of the whole hybrid equation. What happens if I want to access a cold file that I don’t have on-premises anymore. How does that work? [Mike] Sync takes care of that, all transparently. So sink will actually go to the cloud fetch the file and give it back to the user. They won’t really be able to tell the difference except. Of course, we have to go to the cloud to get it so it might take you know a second and a half instead of a half a second. [Fabian] So most of the access requests go against the hot data anyway, so it’s already cached on the server which makes this just the same experience that customers used to have on your file server. That’s great. All right. So let’s let’s keep switching gears. Let’s let’s talk about how do I protect my data? How resilient is Azure Files? What can I do to offload some of the other on-prem tasks that I have? [Mike] So there’s a few different aspects of that. One of it is the redundancy level that you’re going to choose. So Azure supports a few different redundancy levels, you can have locally redundant data where we store three copies of your data. You can have zonally redundant where we store Nine copies of your data, all across different areas in the same region or we have geographically redundant where you can actually have data in two different Azure regions. Now on top of your redundancy level you can also take snapshots. All right, so point in time snapshots of your data. And we integrate with Azure backup as well, which will actually manage when those snapshots happen like set up a schedule and also orchestrate restore so you can do item level recovery or restore entire shares and that sort of thing. [Fabian] Oh, that’s excellent. So I no longer have to worry about on-premises back up because all of my data is sinking to the cloud so cloud backup in the cloud takes care of all of my data right? [Mike] That’s right. [Fabian] So no more additional storage on-prem for backups. No more backup licenses. No more nothing on-prem. I can completely handle this in the cloud. That’s great. Excellent. [Mike] Yeah, and like I said super easy to set up and I’m ready to see your demo. [Fabian] Yeah. Absolutely. Let’s go look at how this thing actually works. [Mike] Okay. So what we’re going to see here is we are going to see connecting Azure file-share to an existing Windows file server. So to do that, there’s a really three steps. One is creating the Azure File Share in the cloud. Another is installing an agent on the on-premise file server and then we create a Storage Sync Service and that actually does the synchronization between that Azure file share and your on-prem Windows file server. [Fabian] Exactly. So, let’s dive right in and take a look. So what we have here is the Windows Server that we want to enable Azure File Sync on. So all we have to do as a first step is we have to get the agent just as Mike set: download, install an agent. We can get it from aka.ms/AFS/agent and there are also more places in the Azure portal where you can see that link so you don’t have to write it down right now. So when we are on the Download Center page, we can see that there are multiple different agent versions available. They support each different Windows Server OS so we have something for Server 2012 R2 all the way up to Server 2019. So the one that I’m using is a 2019 server and to not make you watch a download. I’ve already pulled down that MSI right here. so I can double click and start installing it so it’s a fairly quick installation just using the default directory if I want to set up a proxy that is specific to just Azure File Sync I can go ahead and set that up in this case I don’t need that, yep Microsoft update sounds good but there is also another update option which is actually the worry-free, most convenient option that you can imagine. There is this “Automatically Update” option here. So every time a new version becomes available of the Azure File Sync agent, every time a new version GA’s the server will automatically notice that and will then start to incorporate that new update. It will download the bits and it will start making sure that the agent is patched to the latest version so you never have to worry about this agent timing out or not being supported anymore. It’ll just keep updating itself. [Mike] And what about restarts? Because this is a live server. Correct? [Fabian] Yes restarts are sometimes required so we can’t always avoid them even though they’re rare, but if they happen the way this operates is it downloads the agent and it installs it to a specific point and then it waits for you to restart on a normal schedule. [Mike] So it’ll never restart automatically? [Fabian] It won’t restart your server. It’ll sit there it’ll wait and still operate on the old agent version until you’re ready to restart the server and then once you do that in your ordinary patch cycle, then the next time your server comes up. It’s just the latest agent version. Alright, so let’s go ahead install that. Install that agent right here [Mike] and as this is going: just reminding people that so we’re installing this local agent. And then we’re going to Azure and create a file share, [Fabian] Correct. So we set up storage and Azure and then we set up sync in Azure. These are the two steps that are coming after we installed the agent. There is another step in there as well, which is we’re actually creating a trust relationship from the server. So if you can manage the server and all sync relationships in the cloud, in a secure way. [Mike] And even though you’re installing this agent right now, this local file server is still able to serve files, [Fabian] Correct. Yeah, it’s fully functional. I’m currently installing it, I’ve actually just finished installing it but I can access the local name space. Here are. the different volumes that I have available. Here is my permanent storage and here’s just my ordinary directory that we’re going to sync in a minute. And right now the agent is installed, but I can go ahead and access any file on here. That’s that’s not a problem. So now that we’ve installed the agent this update agent here automatically pops up and kind of looks for updates. So just in case I have an old MSI lying around this one will automatically notice that there is a new version and pull that down for me in this case. It’s a very recent and I don’t have to apply any updates. Now after a couple of seconds, we’re actually getting into the server registration utility. But to give that server registration that trust creation that I’ve just talked about. We’re actually going ahead and go back to the Azure portal and we’ll set up a couple of resources now. It’s about setting up storage and setting up sync. So let’s start with storage first if my Resource Group where I wanted to put stuff in and so an Azure file share lives in a storage account. So just as blobs and files and tables and queues do as well, file shares are being created inside of a storage account. So I’ll create a new storage account here. And now that I have my storage account selected, I just need to answer… I just need to answer a couple of different questions and I basically give it a name. I can give it a location. We should pick a location that’s close by to my file server so that the file server has low access latency to the Azure Data Center and then I can select whether or not I need performance like a premium performance of my file share or not. It’s really not needed in this case if it’s a general purpose file share. If you have a performance app in the cloud that wants to access the file share natively. Maybe you need the IOPS and throughput to to do that. If all you want to do is Azure File Sync, you’re fine with just standard performance. And here’s what you mentioned earlier where you can select locally redundant or geo-redundant types of storage and we can set that up. So I’m not going to go create that now, because it takes a moment to deploy. I already have one created here and we’re now go ahead and select that one. And we can see inside here. We have the different types of storage. I was mentioning earlier and in the file section. I can go ahead and create a new file share. So I’ll just going to call that this study file share. I can give it a quota or not in most cases that’s not needed. So I’ll just create the file share so it can grow to the maximum size. There are five terabyte file shares their a hundred terabyte file shares and you’ll get hundred terabyte file shares globally in just a couple weeks. [Already available at this time] [Mike] That’s right. [Fabian] So, all right, so we have our storage created and now it’s up to the next step which is setting up sync. So I’m going to add a new resource to my Resource Group and this time we’re going to search for Azure File Sync. So and with Azure file sync will see that we can create what’s called a Storage Sync Service and that’s rather simple to do: we can give it a name. Just going to give it that name and we going to pick a location and that location must be the same location as the storage account that we selected. So that sink is always in the same region as the storage that it actually syncs with. So now we can go ahead review and create. So it’s deploying that Storage Sync Service right now. So in just a couple of seconds, we’re done deploying it and when we go here we can now see two main sections here. There is a convenient Getting Started section where you can learn a bit about Azure File Sync and get to the documents. There is a Registered Server section and this is by the way also where you get the link to download the agent. So we’ve already downloaded the agent and now we need to register the server. So we need to make server appear here. To do that, I go back to my little utility that popped up and I say, yep, I want to login to the public cloud. And now I need to provide my credentials. [Mike] This is actually logging into Azure, right? This is like registering the server with Azure? [Fabian] So what we’re doing right now is we’re logging into Azure to get a list of all the available Storage Sync Services that are in my subscription that I could register to server with. And so as this is logging in I have to do the second Factor Authentication here. All right, so that should get me in. There we go. Now what it’s doing in the background is it’s fetching all the resource groups that I have all the subscriptions that I have and is going to allow me to so I’ll pick the subscription that I’m interested in then I pick the resource Group that I’m interested in and then I picked the Storage Sync Service and we just called it “test”. [Mike] So that’s the one you just created. [Fabian] That’s the one I just created. So now I hit register. I have to authenticate one more time in the future we’re going to take that out. [Already no longer necessary] And now we can register the server. Oh registration is successful. There we go. So now I can actually go back to the Azure portal and I can hit refresh here. And here is our freshly deployed Windows Server 2019 where we just installed the agent. So the next step is to tell the server to sync something. Yeah. So what do we want to sync? Here on the server we talked about this particular path. So that’s what we want to sync. So we go to “Sync Groups”. We say: “New sync group”. And we’re going to give that sync group a name. I’m going to call that study. We’re going to select a storage account. So here is a storage account that we can choose from. I’ll choose that here’s our study as a file share that we’ve just created and so now I can set up that sync group. What we’ve actually done now is we’ve created a sync group. If I refresh in a second. We’re also currently creating a cloud endpoint and that cloud endpoint is the location in the cloud. It’s the Azure file share that we want to sync with. Okay. So now we need to say what server and what path on the server should sync with it? So that’s the “server endpoint”. Now that we have that, we can go ahead add a server endpoint. And here’s our registered.server. That’s the VM I was on. We can decide if we want “cloud tiering”. And in this case I say yes, I want cloud tiering and I want to keep 20% of my disk space free. So users can operate and applications can paste new data and as I’m getting less than 20 percent free space, then I’m starting to have tiering kick in and kick out the cold files that nobody’s interested in. So with that set, I’m going to specify the path that we want to sync. That path is this one I’ll just going to grab it and put it over here and I’m going to say go ahead create. [Mike] Cool. Now you’re adding one server here, but it’s pretty clear in the UI, and I just want to point out to people that you can add lots of servers. [Fabian] Exactly. So if you have branch offices and you want to share the same set of data to many other branch office servers, absolutely you can do that. [Mike] All syncing to the same Azure file share. [Fabian] All syncing to the same file share, getting the same data. Yeah, everything works very seamlessly. So this is currently provisioning as we’re going through the individual steps. We can see that we’re all the way down to enabling sync. So it’s actually talking to the server right now to tell it: go and start sync on that location that we specified [Mike] And the local file server is still up and running this whole time there was Zero downtime. [Fabian] Correct. So there is absolutely zero down time if I refresh I can see that We’ve enabled sync now and as that server is now starting to sync, we can actually go back to the Azure file share. So here’s our list of sync groups. We can actually go back to our Azure file share. Here’s the storage account where that file share is in and we can look at the file share is right over here. And even before I was able to… [Mike] data is already flowing in. [Fabian] …data is already flowing in. It starts with some directories, then it goes into subfolders and fills them out. So it’s really really quick to start moving some data. [Mike] Yeah pretty awesome. All right. So I think the main takeaways here is if you have existing Windows file servers, it’s super easy to connect them up to Azure Files using Azure File Sync. It’s not just for Windows file servers: You can migrate from other places too. But hope you check out Azure Files!

Comment here