In this podcast, Jeremy Adair, Sr. Systems Engineer, discuss how Nutanix’s software-defined scale-out file storage solution, Files, designed to address a wide range of use cases.
Speaker 1: On behalf of Nutanix and Carahsoft, we would like to welcome you to today's podcast, focused around eliminating standalone NAS silos with Nutanix Files, where Jeremy Adair, senior systems engineer at Nutanix will discuss how the software defined scale-out file storage solution, Nutanix Files, is designed to address a wide range of use cases.
Jeremy Adair: Hey, good morning. My name is Jeremy Adair. I'm assistant engineer for Nutanix. I wanted to take the time to thank everybody for taking the time to join. I've been with Nutanix for about three and a half years. Prior to that, I ran infrastructure for joint special operations command, managing 3-tier architecture FlexPod and prior to that, I worked for NetApp. Today we'll be covering Nutanix Files and Files Analytics because that's something that Nutanix offers natively. In the past, a trip down memory lane, when I was at NetApp, and even when I managed FlexPod, I also had to manage a storage area network or a NAS, and with that comes complexity. How do we secure? How do we manage that? And then the maintenance of it. How do we upgrade it? Provision lawns and volumes and create file shares.
And then how do we track what files are where, and far as quotas and so forth and so on. Nutanix makes that really, really easy with Nutanix files. So if we look at data life cycle management, to automate the process of managing data flow from creation to deletion, right? Nutanix tries to make that as invisible as possible today. So as you can see, Nutanix is more than just HCI, right? Converging the three tiers into a one single interface managing your entire stack. We provide block services, file services and objects services depending on what your use case is. And this is pretty much what I used to manage. I manage for 10 years, right? That's not my cabling. I have OCD and my cabling would never look like that. But just the complexity and we look at SANs, right?
I'll look at a SAN as a 1995, 94, I had a Motorola back cell phone, right? And that back cell phone only served one purpose. That purpose was to make calls, right? So to take a picture, I had to carry a camera and to take a video, I had to carry a camcorder. So Nutanix took the smartphone approach where the intelligence of software could enable these functionalities without needing all the hardware and cabling. You still had to have hardware in cabling, but I'm talking about the silos of infrastructure everywhere. And then how do we scale a SAN or a NAS? And say, add a HA pair, right. And do a bunch of 2nx cables and a bunch of SAS cables. Right? So we do that. We scale based off of adding another node to the cluster and scaling out through prism.
So if you notice NetApp on command and these different interfaces, just to manage storage, right? So we move all of that, the whole stack, as well as all these new additional capabilities like block through OSCA as the foul SNB and NFS and object to a single interface called Prism. Its HTML5 interfaces that manage your entire stack. And a lot of top five reasons of why customers use Nutanix files is simple, it's flexible, it's easy to scale, it's highly available and resilient and it's API and data-driven, and it's also integrated with the cloud. So in order to create a file server and Nutanix, which we'll walk through shortly, it takes about 20 minutes to create a file server cluster. I don't know if you know, but I'm assuming if you're a Nutanix customer, maybe you're not, in order to form a Nutanix cluster, we start with a minimum of three nodes to create a Nutanix cluster.
And there's a reason for that. We've maintained the integrity of two copies of data all times the same thing goes with file servers. You start off with three file server cluster that creates a cluster quorum. And then we create shares off of that. I'm going to go ahead and dive into the demo. So this is the prism interface that manages our entire stack. Like I said earlier, we did not rely on complex stands and NAS devices. So if you dropped down here and got a file server to create a new file server in Nutanix, you click file server add. Now for this example here, I'm not going to create a new file server cause we have about 30 minutes and it takes about 20 minutes to build the file server cluster but if I wanted to build out a file server cluster, put the name of the file server, domain name, how large you want the file server, as far as capacity. You can customize the capacity as well, right.
Just depends on how large your environment or how small. And you can configure it based on performance. Once the file servers build out, we can integrate that with active directory and the old app and all that stuff. We support both NFS and SNB and to create a share and Nutanix click add share, name the share. What file server are you using? What protocol are you using? Now enable multiple protocol access. For the old school SAN users, it used to be called multi mode, but we can create a share that supports both SNB and NFS, can be accessed by both windows and Linux users as an export. For this example, I'm just going to show you SNB for windows. Here we can create two types of shares that the distributed share or standard share. What a distributed share means is that you do not need another load balancer. That share is load balanced across all FSVMs or file server VMs in the cluster. So we'll create this.
Standard share is a share that is designated to one file server VM in the cluster. Self-service for stores, when those previous versions, see do not lose that functionality even going up with Nutanix, you can enable windows previous versions here. We also do In-Flight Encryption on right with SNB three grip SNB three messages. And we also can do block file types. So if I wanted to keep someone for putting the.movie file in here or .FLV file, that's where I would put it to block that file type from being placed in that share. Once that's been set, it's going to give you your mount point, the name of your share, just click create. You could use a search functionality up here for the share that I just created. So I created that J six dev home Lao file server. And there's too many hands in the cookie jar here.
Normally you would just have one file server cluster and your organization, and you would scale out on that. If I wanted to connect to the share I created, I would go to a windows machine, I would grab my mount point, which is down here. And that's how easy it is to create a file share distributed file share on Nutanix. Now, if I had set restrictions on what files could go in here, the Colonel used to call me all the time and say, "Hey, Jeremy, we only have a certain amount of space. I want a file share. Can you determine what files, what type of files are saving and how old and so forth and so on." And with the old traditional tools, I really had no way of determining that, right? So I'd have to go through each share and look at it.
I'll show you shortly how Nutanix handles that. Also, with shares too, when you're having a performance related issue, we kind of take the guesswork out of it. We can look at how many files are in that share, we can look at the performance of that share. Now some of these shares are not in use. This is a demo site. And then I can look at what quotas. So we have two types of quotas. We have soft quotas and hard quotas. Soft quota will send the user an email. These are for departmental user shares. It could send a user, an email saying that you need to clean up your data. Soft quota allows them to keep writing to that share or a hard quota. If we set a quota policy hard limit, that means that they'll get emails and they will not be able to write to that share.
These quota policies could be taught to a user or user group from items directory. This is where you specify the size of that user. And this is where you put your email address or SMTP addresses in here so it notifies the users that they need to clean up their shares in order to maintain, keep the capacity down. If I wanted to create an NFS share, it's the same thing I create Add share here, choose the size of it or choose NFS. And then I'd click next. And I'd mount that as an export on a Linux server. If my file server was... Our software's intelligent enough to know when performance optimization is what they call it and needs to take place. So remember I talked about standard into distributed shares, distributed shares, load balance, right? If our software identifies that the performance of the file server is lacking, it will pop up with a performance optimization and recommend you either scale up the resources for that file server VM or Scale-out adding another file server VM. So I can scale in or scale out, right?
You can't do that with traditional architecture without adding another HA pair. So I could add this, I could change this to four file server VMs, or I could update the CPU and storage with the capacity of the existing. I can manage roles to my file servers. I can also manage roles and users for my shares. If you look at your share and you wanted to update your share, you can go in here and modify the share. If you want to, if you decided you want to block certain file types. When we create a file server cluster, we protect that cluster with what's called protection domains. So as soon as you add file server cluster, it creates a asynchronous replication schedule.
That schedule, there's a schedule that you can set up hourly, daily, weekly, or monthly, however meets your organization’s requirements. And then that, the snapshots are sent a file server VM information to another location. So now I'm going to pull up File analytics. File analytics is a drone view into your environment and I really wish I would've had this when I was an onsite engineer to... Colonel comes up to me and says he needs to identify what file types, who's accessing and why? And I couldn't provide any information without manually looking through average year. So this is Nutanix file analytics. This is the home screen. We can look at a capacity trend up in the top left of what capacity has been added and removed in a net change. So we can change that to 30 days up to one year to see, you know, pretty much how our data is growing or shrinking.
We can look at the data age, anything less than 12 months old. And where that capacity is at, and how old it is. Anomalies are for stuff that's out of the ordinary that we won't have set up. We can set up anomalies. So you have a upset employee who goes in and deletes a hundred files from a directory that you need, that would pop up here.
If we set it an anomaly up to pick that up. Permission denials, that's good for a ransomware and folks that are trying to access a file that they actually don't have a need to see. You'll be able to see your permission denials here. And then we can see file distribution by size, how large they are. File distribution by type, what kind of files are we storing? And this is going to give you a detailed view and what type of files, you click here for more information. We can look at our top five active users, 24 hours up to one year, top five access files up to one year, and then file operations up to one year, permission denied... Used to have engineers coming up to me all the time and say, Hey, somebody took ownership of my file. And I had to track that down. And that was nearly impossible because they could go in and set permissions to themselves and take ownership of that file.
Audit trails is a great one for that permission change. Who's adding, deleting and creating files. If I go back to the dashboard and I've looked at the permission denials, Ryan Green had a permission, but now, so I could look at the audit trail. I can look at the audit trails by client IP, by user folders or files. And I see any on operation date. I can view an audit trail of everything Ryan Green has been accessing. I can filter my operations by what he was actually doing. Permission change, permission denied. This gives me the ability to go back and actually set the permissions back. And first of all, lets me know, ask him why he changed the permissions anomalies.
When we spoke about anomalies, I could add a user. I can do follow them an anomaly rules and I can do create actions, events, delete permissions, change permissions to nods and file blocking, read and rename. I can set several different anomalies. That anomaly is something out of the ordinary is going to show on your dashboard right here on this demo cluster nobody's actually set up anomalies. This information can also be exported into a CSV or a file that's readable. If you go into tie and directory service, you can do that here or Mac users to particular file server. You can do that as here as well. Your DNS information is here.
Remember I told you about windows previous version. So if you click on the share, self-service restore, those snapshots will be listed here had any snapshots been taken out at the time. Virus scanning can be done also whether use of adding an ICAP server. I'll send you the link over for supported. I kept servers like McAfee and some of the others where you can actually scan your SNB and NFS shares for viruses.
Well, thank you for joining us today. We can dig much deeper if you need any additional information, let me know. I can be reached at jeremyadair@nutanix.com.
Speaker 1: Thanks for listening. If you'd like more information on how Carahsoft or Nutanix can assist your organization, please visit www.carahsoft.com or email us at nutanix@carahsoft.com. Thanks again for listening and have a great day.