Grey Beards on Systems - 095: GreyBeards talk file sync&share with S. Azam Ali, VP Customer Success at CentreStack

Episode Date: December 20, 2019

We haven’t talked with a file synch and share vendor in a while now and Matt was interested in the technology. He had been talking with CentreStack, and found that they had been making some inroads in the enterprise. So we contacted S. Azam Ali, VP of Customer Success at CentreStack and asked if he … Continue reading "095: GreyBeards talk file sync&share with S. Azam Ali, VP Customer Success at CentreStack"

Hey everybody, Ray Lucchese here with Matt Lieb. Welcome to the next episode of the Greybeards on Storage podcast, the show where we get Greybeards Storage bloggers to talk with system vendors to discuss upcoming products, technologies and trends affecting the data center today. This Greybeard on Storage episode was recorded on December 12, 2019. We have with us here today Syed Azam Ali, VP of CenterStack Customer Success. So Azam, why don't you tell us a little bit about yourself and CenterStack? Thank you, Ray. Thank you, Matt, for inviting me to this podcast.

I am the VP of Customer Success at CenterStack. What we do is basically we provide a cloud file sharing solution like Dropbox. And also, you know, the most important use case that we solve for the pain point that we resolve for customers is allowing them to provide remote access to their on-prem data or the internet without a VPN. A lot of enterprises are looking to provide easy access to their users without users connecting via VPN into their internal network before they can access the data that they need. And we also provide cloud backup. You know, we integrate with Active Directory natively if there's a need to basically do Dropbox alternative type solution to make sure everything is centralized and IT has full control over the data instead of giving control to companies like Dropbox or

ShareFile, Citrix, ShareFile and stuff. This way, we provide a medium for them to keep the data under full control. That's right. And also have some security around the data. It's interesting that you indicated that you can do this without VPN. Do you still provide security for the, I'm not exactly certain what the terminology is, the connection?

Basically, it will be over a secure access point. That's right. So you will deploy CenterStack on a clean virtual server in your network. All we do, basically, it's a gateway to your on-prem data. I see. And then users have the ability to access it as long as they have internet access,

they have access to the files that they need. You know, I've had problems where I've been on different corporate offices where accessing certain websites were a problem, but you wouldn't need that in this. I would have to use VPN in those environments. For CenterStack, you don't need to use that sort of solution. You don't need to have any VPN connection into the network. As long as

you have internet access, you have access to the files. That's right. We have clients for Windows, we have Mac client, we have mobile applications, that's right, for iOS devices and Android devices and stuff. And it's Dropbox-like functionality, so I can update the file on my Mac laptop, for instance. It'll show up in the corporate data center as updated, and it will also be deployed down to any location that's synced with that. Is that how it would work? That is correct. Well, you know, whenever a user makes a change on a file, it will be synced.

The changes will sync to the central location, and all the other users will see the most up-to-date copy. If companies have a lot of branch locations, they can use our file server agent technology where you can deploy that on a local machine in that local branch on a local file server, for example, and that will act as a cache appliance. No need to purchase hardware appliances or other infrastructure. All they need is a local file server. Or if they have less than 10 employees, they can even use windows 10 and convert that into a cache appliance

and sync the folders that that location needs for those employees in that location and our software agent will basically keep the changes in sync in both places does it do versioning yes we have versioning enabled by default that's right so versioning is very important because if if a folder gets attacked by ransomware, you as a company can really go back to a previous version right before the attack. And because this is, I'll say, a corporate deployed service, it doesn't require cloud infrastructure or anything like that to be added to it?

Is that how this would work? There's no cloud infrastructure required. That's right. So if you have a data center and you want to keep everything on-prem, you deploy CenterStack on a clean virtual machine in your own data center, and then you have to install the SSL certificate, get the public DNS name, do the NAT rule. That's right, the firewall rule and stuff, and that's about it.

All we need is 443 inbound on the firewall to be open. God, you need to support Macs. That's the first thing I'll have to say. Windows 10 is great, but being a non-Windows user, it's a problem for the gateway. Oh, for the gateway. We talked to a client a while back that supported actually hardware. It was like a Linux drive and stuff like that, and it offered sort of file sync and share, but it was sitting on your desk or something like that for a desktop environment.

This is a bit more sophisticated than it ever was, I would have to say. Yeah, we have Linux client on the roadmap that we will be releasing hopefully, you know, sometime in the future. I don't know when, but if users are using Linux, they will get the drive, map drive on the Linux machines too. Right now, if they want to use Linux, they can use a browser, log in using the AD credentials, and they will see all the folders they have access to. Right. So it supports SMB file servers as well as NFS file servers? That is correct, yes.

And there's really no, I mean, is there, you know, the question is just probably a dozen file servers that I can name off the top of my tongue. You support any file server because it's NFS or SMB access? Yeah, as long as it can be access or SMB. That's right, on NFS, we support. We even support data that's on NAS devices, for example. As long as it's a CIFS folder, share, you can use a Windows 10 machine, connect to that NAS folder, and then you can attach that NAS folder to the center stack. There's a whole bunch in the high performance computing space, special purpose file servers and stuff like that, some of which are not NFS and stuff.

You don't have any support for any of those guys, right? Not right now. Huh. Well, this is very interesting. Is there any limitation to the amount of storage that that you would i'll sync and share i guess i mean this basically the storage is not required on the server center stack server itself basically the backend storage is the file server shares there's no limit you can have terabytes of data as, you know, all we are doing is connecting to your

folders on the file server and publishing them as team folders. So you are not really copying the data anywhere. The data is staying on your file servers. Metadata is in your control. We are not taking the metadata because center stack server is going to be in on the center stack server or in the database, basically. That's's right so it's not like some other solutions where they take the user metadata in their control plane in the cloud in this case even metadata is staying in your own private cloud so you don't actually replicate the metadata for the NAS servers or the file systems that that that are under sync and share on the gateway storage per se or anything like that and no and no yeah so you mentioned the database and that sort of

stuff does it does it does it have an internal data structure I must have something database wise and that sort of stuff. That is correct. We basically, you know, when you use our installer, you choose the MySQL database, which comes with the installer, or you can connect to any external database. We support MySQL and SQL right now. Right, right, for your internal structures and stuff like that. So somewhere in there you mentioned cloud backup and that sort of stuff. How does that work in this environment?

So basically if you have companies who want to copy the file server shares or allow users to copy, for example, my documents folder or videos or pictures on their phones to a storage repository in the cloud, like AWS S3 or Azure Blob Container, we allow that. That's right. You can enable the cloud backup feature, and that will allow you to basically take those file server shares that you have on-premise and copy them into, for example,

Azure Blob or Wasabi storage, along with all the permissions. So our cloud backup is different than other cloud backups. The other cloud backup solutions are like raw backups. That's right. When we are making cloud backups, we are also taking the permissions and the folder structure and backing that up into the cloud. So it's very easy when you want to restore,

you are restoring from the cloud to your on-prem file server, you know, the data as well as the permissions. You can even go very granular and go into a subfolder and restore subfolder to your local file server or even restore a file for a user. So in this environment, let's say I had multiple corporate data centers, I would have a center stack, I'll call it virtual machine gateway, running in each of them?

You don't have to. You just need one center stack server and then the data center, if they have the file servers in both locations, center stack can and then the data center, you know, if they have the file servers in two, both locations center stack will connect, can connect to both. Yeah, I guess it depends on how traditionally diverse your data sets will. So if you needed to have two separate file servers

for two separate locations, then it might make sense to put center stack in both. Yeah, I mean, it's, I guess there's different hub-and-spoke kind of models that could be deployed here, one of which is, like you say, one central data center where all the satellite data centers connect to for information. Yes, we have file share synchronization. That's right.

So it's very easy to basically sync the shares to the branch locations. And that will simplify collaboration on branch locations. That's right. The users in those locations don't have to travel all the way back to the central file server. They might not have a good internet connection back to the data. Oh, so they could actually be sort of disconnected for a while, like my laptop when I'm out of internet access. So for that, you have to enable offline access. We have offline access,

which you as a user can enable it yourself from your cloud drive, or the admin can enable an offline for, you know, enable offline access for like road warriors who are always on the road, they might be in a plane, they need access to the files in a folder. They, you know, they can add, if they enable offline access, then they will have access to the data in that folder, even in a plane without a Wi-Fi, They can make changes. As soon as they get off the plane, get the Wi-Fi back, the changes will sync to the central location. They can even attach a local folder from their local machine. That's right. And then when they are in the office, they can continue working on that local folder on the C drive. But when they

are away from the office, they will be able to access it from the cloud, from internet basically. So it's kind of like a Dropbox on steroids. Yeah, I would say, you know, it's from a Dropbox perspective. It provides offline access to specific folders or files that you designate and that sort of stuff almost

on as a normal default I'm not sure I've ever seen just online access to to Dropbox but so explain what non offline access would actually look like if I was sitting here on my desktop and I'm I've got access to a folder on some corporate headquarters in Chicago and I'm here in Denver, I would open the, you know, just access a file like I would normally do. But in this case, Centristack is going to bring that file to me in real time. Is that how it works? In real time. Basically, for example, you open a browser on your laptop, you type in the URL, that's right, the secure URL, and then you'll be presented with a login page. You will put your credentials. As soon as you put your credentials in, they will display all the folders that you have access to.

And if you are using iPhone, for example, again, you will download the cloud application. That's right, CenterStack Cloud for iOS. And then you'll put the access point, you'll log in, and you'll see all the folders you have access to. Yeah, so on a Mac, for instance, after I've done the login and that sort of stuff, does the, and I'm not sure, does the center stack cloud storage device appear on my desktop at that point? In the finder, you'll see the center stack cloud drive. Oh, okay.

So it's just yet another external drive from that perspective. That's right and then and uh and you know on the top you'll see the cloud icon and all it needs are active directory credentials and the security uh and i imagine it supports two-factor authentication as well yes we have our own built-in two-factor authentication which you can enable for all the users that's right in which case they have to use the microsoft authenticator or google on authenticator besides you know they have to put in the code besides their ad credentials one important feature we have is on-demand sync so a lot of a lot of solutions out there they do a full sync down of all the data to fill up the user's hard drive. With our solution, we will display all the folders users have access to on, for example, in their cloud drive, but nothing is downloaded to their C drive.

Once they go into a folder and double click on a file, that's when that file will be downloaded to a local cache folder, which is hidden from the users. So next time they need to access, and that location is there for active data. You as an admin can define the size of that folder, or you can disable it completely. That's right. The hidden cache folder sitting on the desktop? On the desktop somewhere for faster access. So next time if user needs to access the same file, that's right.

If they work, they don't have to travel all the way back. They will be accessing it from that local cache folder. And then our engine, client engine will pick up the changes and sync to the source. And if somebody else is accessing the same file, does that present a lockout situation or something like that? We have distributed file logging feature enabled by default. If one user has a file open, he will place a lock on the file. And if another user tries to open the same file, they will be displayed. They will get a message.

Basically, user one has the file open, editing, you can have a read-only copy. Right, right, right. And that happens regardless of where you're accessing that file from. So I could be sitting in Denver accessing a file in corporate Chicago, and somebody in New York could try to access it the same time or sometime later, and he would see the lock at that point. That is correct. That is correct. And you mentioned that the cache file, the hidden cache file sitting on the person's desktop, for instance, is configurable in like terabytes or number of files? only use 10 gigabyte of the hard disk space or one gigabyte that's right after if it fills up our software will automatically clean the old files to make a space for the new ones you as a user don't have to do anything it's all automated and god forbid if the user would open

you know let's say his configuration is 10 gigabytes if he opens 20 gigabytes of files in one instance and starts editing all of them, he'd get some sort of an error message? Error message. It would expect, yeah, yeah. So, Azam, how is this licensed? It's licensed based on user. It's user-based licensing.

If, you know, a user can use 10 different devices, but it will only consume one license. And can the user have, let's say, you know, I'm a consultant. I work with a number of corporations. If it's, you know, potentially all of them could have center stack, shared file structures and that sort of stuff. Could I have access to all of them from my desktop? I mean, presumably one at a time, right? I mean, I would be working with corporate A, corporate B, and then corporate C, you know, over the course of many months or something like that.

Yeah, I mean, if they give you access, if you have access, you will be able to access the files. Yeah. I mean, we have customers who have some data, let's say, in Azure. That's right. Or in AWS. And also, we have customers who have some data, let's say, in Azure, that's right, or in AWS. And also they have hybrid environments. They have some data in Azure, in cloud, and some on on-prem. So we present that data to the users in their MapDrive too or in a browser.

So they will see the on-prem folders as well as folders in the Azure container or S3 buckets. And they can perform all the operations. That's right. They can share the files. They can share a folder. This is also very easy, you know, for if you have to share a large folder, you cannot really send in an email. That's right. Like some other solutions, you can just send a secure link yeah so when i first came up on one of these services you

know moving the terabyte of data across uh uh you know a home oriented internet took took days to move all that data so you're saying that if I had the stuff sitting on the cloud someplace, I could just provide a secure link to it? Yeah, you can just provide a secure link to it. That's right. So actually, Ray, you brought up an interesting point. What about seeding large portions of data? If you're an organization, can you copy, you know, of data. If you're an organization, can you copy a half petabyte onto an Amazon Snowball type device? You can, that's right. It's your own S3 account or a Snowball. You should be

able to. You will have to send, I guess they send you some sort of a device where you can copy the data and send it back to them. All we do is basically, yeah. In that case, you would have an S3 object storage with that half a betabyte. Now, accessing it through center stack would take another step of some type, right? That is correct. You have to basically, you know, we have a concept of team folders. Basically, you know, you will create a team

folder and instead of connecting to your file server share, you will choose cloud storage. And in the cloud storage, you have all these options where you can connect to AWS S3 bucket. That's right. Azure blob container, Wasabi storage, or you can even connect to SharePoint data. We allow integration with SharePoint too. So if you have folders already in SharePoint data in SharePoint, you can present that to the users too in a map drive or in the browser.

So it's like centralized access for all the data that's out there, whether it's in the cloud or on-prem. Before you go on, so does that include exchange mailboxes and that sort of stuff? No, no, not exchange. You mentioned SharePoint, so exchange would be the next logical step in my mind. Yeah, you're asking a lot there, Rick. I guess.

Am I dating myself here? Maybe. I don't know. Theoretically. What about OneDrive? OneDrive, yeah, can be integrated as well. So essentially, it's a customer-owned sync and share service that can be deployed on-prem, can also make use of cloud storage as well?

Do you supply any sort of replication capabilities where you replicate data from one gateway to another gateway that might be used as a disaster recovery site? Yeah, basically, a lot of companies, what enterprises, what they are doing using our solution, they are copying the data that they have on-prem into cloud storage as a backup, as a backup for disaster recovery and business continuity. For example, if they lose access to their on-prem data center, users will never have any downtime. They'll be able to access the files directly from the cloud. The question about the applications running with the data,

but accessing the data is probably the most critical step. I agree with that. Yeah. Yeah, yeah. And then is it possible to set up a replication? And it's probably outside center stacks per view from a an azure block to say error a wasabi a three block now what we allow basically if you have data in the cloud for example and you want to bring a copy down to a local file server that's right we provide file server agent, which you can use.

So if you have data in the cloud and you want, in a location, you basically create a new file server and you wanna bring the data down, all you have to do is use our file server agent software, and that will allow you to copy the data back down. If you wanna have a local copy. Right, right, right. And then you have some sort of WAN optimization software to make sure that you're utilizing the internet wisely

and that sort of stuff? No, not right now. Usually internet is getting fast every day, so latency is really going out out the picture right now. But, you know, for locations which have bad internet connections, they can use a file server agent as a cache appliance. That's right. You mentioned Wasabi a number of times. Is Wasabi becoming a popular cloud storage target? The reason is the cost. It's the cost it's not as good as aws or azure that's right but

economics is one-fifth the cost of aws so a lot of times if you you know it depends if you are already a aws shop or azure shop that's right most likely you're going to stay with aws i'm talking about small businesses who can't really afford AWS, then they go with Wasabi. Speaking of businesses, I'm not sure what the right term is. Let's say, how many gateways have you currently deployed? We have millions of users using it all around the world. We have a lot of businesses, a lot of partners. That's right. And it's a multi-tenant solution. So we have a lot of MSPs who have deployed this.

And they can create unlimited number of tenants from a single console. And then they can integrate into their customers on-prem Active Directory and file servers. That's right. And so this would be, it's a software solution. So I can effectively download the CenterStack gateway software and deploy it on my environment. And once I pay for the licenses, I can enable users to access the data. Is that how it works?

That's correct. That is correct. Yeah. If you are a MSP, for example, that's right, you can deploy center stack in your data center. That's right. And you can offer your own fully branded white the permissions to the cloud storage or the storage that you have and then retire their on-prem file servers. If they want to get small businesses, that's right. They don't have the resources to manage the on-prem infrastructure.

You as an MSP, your goal is to basically take over their data, take over their IT. So this can be, you can use center stack for that. And then you provide like 7x24 support. Is that part of the license agreement or is it something they would purchase on top of that? You'll pay 15% additional per user to get the premium 24-7 support. And you can open unlimited number of cases in our support system. That's right. Do you have any sort of, I'm not exactly certain what the terminology is, but trialability so I could bring it down and load up like, you know, maybe a half a terabyte of storage and try it out to see if it works?

Yeah. You get 30-day free trial. You can try 30 days free trial. And a limited number of users for that, I would assume? No, unlimited number of users. That's even better. Oh, gosh. I use FileSync and Share all the time. I've got data sitting on my iPad. I've got data on my laptop. I've got data on my laptop. I've got data in my desktop.

I've got data in the cloud. I use these sorts of services all the time. I guess the question is, so it's licensed on a per-user basis. That's interesting because everybody else, I would say the big guys are all licensed on a per storage capacity as well as user basis, right? Matt, is that the case? Yeah, I mean, we have a SaaS offering too. That's right. If you are a small business and you want to have a copy of the data in the cloud,

we have a SaaS offering which we host in AWS data centers. That's right. Which comes with the storage. So you have to sign up for a minimum of, for example, 10 licenses and you get one terabyte of S3 storage. That can be shared across those 10 users, regardless of the number of endpoints they have.

I mean, like iPads, iPhones, laptops, desktops. So it's anywhere. Yeah, yeah. Yeah, it's user-based license licensing that's right it falls under two different categories right the first is all these cows that that you buy consumer staff but in addition to that you've got this this whole concept of how much storage you're consuming on either us or whatever provider you're using. So you're actually paying not just for the client access license,

but you're paying Amazon for their storage as well. Is that a correct statement? In that case, if you want to go, you know, copy the data into the cloud storage also. If you want to keep everything in your local environment, then you're not paying for the storage. The other thing that's interesting here, and I'm not sure exactly how to say this,

is that you're sort of passive to any of the file services that are out there. A lot of the on-prem sync and share services, you pretty much deploy their complete storage system, and it supports NFS or SMB file shares and that sort of stuff. But you're actually storing your data on their backend storage. We are a storage agnostic.

We don't care as long as it's a safe share accessible over SMB, NFS, you can access it without a VPN. And it's seamless, which I think is really compelling. The user doesn't know or doesn't actually have to care. Where the files actually physically reside. Exactly. Yeah, for users, it's like they're accessing from their local machine because of the on-demand sync feature.

And then for, let's say, a corporate NAS box or something like that, if I'm accessing the data in the corporation, it's going directly from a server to the NAS box. You're not anywhere plugged into that data path. No, we are basically going to be connecting to your file server. As another connection to that server. So if I update a file, let's say, from a corporate server, how do you get notification that the file has changed so that it could be synced to, let's say, any users that have copies of it offline, for instance?

Basically, it can always sync. As soon as a change is made to a file, it will sync those changes. That's right. So the next time a user opens that file, they'll see the updated copy. Yeah, Azam, I guess the question is, how do you get notification that, you know, a data center could potentially have, let's say, a million files, right? Here I am, a data center user. I go update file X or something like that. How do you know it's been updated?

I mean, you can subscribe to notifications. As a user, you can subscribe to… So the gateway effectively is subscribing to notifications of the NAS server that files have been changed. Is that how this works? My concern is you're not in the data path from a corporate perspective. So you're sitting there passively behind the, let's say, or in parallel to the data center accessing these files that have been changed and allowing them to be synced and that sort of stuff. That's the key thing. As long as you are using our clients, if you are making changes to the files, that's right.

So anybody in the corporation that wants to access CenterStack, well, do we have to use the client software in the corporation where we have direct access to the NAS files? You will be accessing the NAS files through a share on a Windows machine. That's right. Yeah, directly. There's a group policy setting that you can enable, which will, even if you access the share directly, it will, you know, our file locking will work. We have many granular group policies of our own that you as an admin can, you know,

configure to basically configure the environment based on your needs and requirements. Yeah, so is there like a, I hesitate to say, an operating console for the CenterStack gateway that tells you kind of how it's being used and that sort of stuff, or how many files shares are being deployed? And that file share, file sync, I guess. We have, yeah, basically we have a console it's a centralized management console that you can go in and monitor everything we have auditing

and reporting built in you can look at the audit reports on what's happening on the data who's sharing the data externally to whom and then you can also look at file change log to see who has deleted the folder by mistake. That's right. Basically, even if a user deletes a file or a folder, it's a deletion marker. The file or folder never gets deleted from the backend storage

until it hits the retention policy that you have defined. So users can always go back and do So users can always go back and restore. And that works for any of the users accessing the data via your client or gateway service. Yeah, yeah, on a browser. Yeah, yeah, yeah, exactly. That's right. But, I mean, if I was accessing the data directly from the corporation, I could delete and do all this stuff. And your logging and that sort of stuff

would not necessarily be active for that. Well, gosh, this has been great. Azam, is there anything you'd like to say to our listening audience before we go off? Yeah, I would say that they should take a look at CenterStack. It's, you know, it's, uh i say this this is like the best kept secret they have uh you know it's very feature rich it's enterprise ready and uh it's

just you know it's not known out there so i think it has a use case especially the direct access use case for a lot of enterprises who want to keep the data under full control and they want to have security around the data and still provide the flexibility for the users to access it on any device over the internet without a VPN. And also, yeah, that's it. No, for business continuity and disaster recovery also, we have the backup. That's right. So they can have a copy of the data away from their data center in case of a disaster,

if their data center completely goes down or the file server completely dies, all they have to do is spin up a new virtual machine and convert it into a file server, use our file server agent software, and copy the data back down. Well, this has been great. Thank you very much, Azam, for being on our show today. Thank you for inviting me, guys. Next time, we will

talk to another system storage technology person. Any questions you want us to ask, please let us know. And if you enjoy our podcast, tell your friends about it, and please review us on iTunes, Google Play, and Spotify, as this will help get the word out. That's it for now. Bye, Matt. Goodbye, Ray. And bye, Azam. Goodbye, Ray.

Until next time.