 We're on. Here we go, Swiftstack. My name is Chris and this is Doug. We're gonna spend the next 20 minutes or so going through Swiftstack. Specifically, we want to talk about ways that Swiftstack is used in the world, in our customer base. I'll introduce Swiftstack real quickly and fundamentally, as you can read, Swiftstack is object storage software. We are intentionally focused on providing massively scalable, massively flexible object storage and we're built, as you might guess or you may know already, we're built proudly on the foundation that is the OpenStack Swift project here. We've been a part of the OpenStack community from the beginning. We employ the Swift project technical lead, several of the core developers, and we work with a number of the companies here to move that project forward on a regular basis. But Swiftstack then takes this solid, extremely well-proven, fundamental, open-source Swift technology and builds on that a management tool, what we call a controller, which takes that engine and turns that into something that is ready for the enterprise, that has the integration with things like authentication tools, that has management and monitoring and a handful of other pieces and parts that we add on to what is the open-source OpenStack Swift project. So why does Swiftstack, or why does this the Swift really object storage platform or concept exist in the first place? There's a couple of things and if you go back to when I started in storage and my background is as a developer on the traditional storage side of things, SAN and NAS type of platforms, that's where I started. From that point forward until now, what we have seen is a change in the industry, a change in the infrastructure, a desire as you see on the screen for storage and really for all of infrastructure to be more agile, to be more flexible, to be more scalable, to be less about lock-in, more about transition, ease of transition and what has driven specifically in the unstructured space of the storage side of the industry is a push toward a new method of access and a new architecture and so what that looks like for Swift, for Swiftstack in particular, is rather than mounting silos of NAS or SAN storage, you now access everything through an API, right, flexible. It could scale, you could literally have a Swiftstack storage platform that scales continents and have a single mount, or not mount point single API for access and the architectures I mentioned could run from everything from a couple of nodes, a couple of virtual machines, if you're just trying this out for the first time, to a couple of bare metal nodes, three of them likely in your first data center deployment out to 10 data centers across 20, or 20 data centers across all the continents and anything along the way that flexibility is there. Specifically, and we're going to talk for the next few minutes about use cases, specifically we see them kind of fall in two buckets and that's what's on the screen. One is, as you could expect related to the OpenStack world, is storage for private cloud and we won't spend much time on this because it's kind of a no-brainer. If you are standing up, whether it's through a distribution or roll your own, you're standing up an OpenStack cloud and there's going to be a Swift component in that, probably as the back end for Glance or for your Cinder backups. Swiftstack is hands down without question the simplest way to get that up and running and you can grow that from your simple Glance repository up to whatever else. But if you're starting to go from the private cloud or even independent of that private cloud, a massive percentage of our customer base is actually just on the object storage side of things. So they're saying we need that scalable platform for our unstructured data and it may or may not tie into private cloud for them and that's where we want to spend the next few minutes talking about some of those use cases. We'll run through back up at Swiftstack as a target for backup. We'll talk about file sync and share, kind of that enterprise drop box experience, if you will. We'll talk a little bit about some other use cases that fall in there as well and a way that we can bridge the legacy applications to the object storage as we move forward. So I'll hand over to Doug to talk about backup and I'll switch back in a minute. Hi, everyone. So this is probably the most obvious use of open stack Swift storage is backup. So we already know that Cinder and Glance, those projects use Swift as object storage backup, but there are a number of commercial backup applications that work as well. So if you've got, you know, Commvault, Netbackup, IBM, TSM, which is now rebranded Spectrum Protect, they're going to be able to backup as well. And the key factor here is why not just use your existing NAS? So people have used NAS, filers, block storage forever to backup to object storage, obviously commodity hardware less expensive, but the the neatest feature out there, and that's what we're in demo in a second, is the DR scenario. So when you're using object storage as your backup target, you have a single namespace across all of your geo clusters. So you can have New York, San Francisco, Texas, same namespace, same data, a replica in each, you can backup to any of them, you can restore to any of them, fully automated, no reason to figure out replication, anything like that. In fact, you can back up immediately to New York and access that data in California. Whereas if you're backing up to a NAS or saying you would do your backup, you'd take a snapshot, you'd have to wait for that entire snapshot to sync to the other side. In fact, it would be then a different namespace, a different mount point, you'd have to tell yourself for how to access that mount point. And so the simplicity is what really matters. So let's see that in a quick demo. Alright, so in this demo, we're going to use convo. This is a talk I give. And so what I've done is I have a backup in New York City. Now I'm in our San Francisco office, and I want to do the restore. And what I'm going to do is I don't have to tell convo anything differently. I'm not pulling this across the wire. I simply go into convo and I pick what I want to restore, which happens to be somebody's desktop with cap photos, because we know that's the best use for open stacks with. And what I'm going to do is I'm going to just run against the media agent that happens to be in my San Francisco office. Nothing else needs to be done. I don't have to tell it's a different namespace. It says if I was going to the rack space cloud or any other cloud, we're presenting as a cloud on site, simple, done, my restore is running. And this would have been much more complicated. Had I used block storage, I would have had to replicate. I would have to replicate the database. I'd have to mount it. I'd have to import it. Instead, I'm done in less than a minute. In fact, I'm done faster than I can talk through it. So Chris, why don't you talk us through file syncing chair? Sure. And just an obvious statement, all of this is being recorded. So if that screencast went through quickly, and you'd like to slow that down and see what happened, either watch for the recording of this and you can slow down and pause or go to our website, Swiss stack comment, we have demos of all of these things that you can watch in more detail as well. File syncing chair, kind of an industry term that's growing now for, as I mentioned earlier, this idea of the Dropbox or box.com or Google Drive experience, right? It's really the new way in many, many senses to do home directories. We all have worked around and had this idea that we have a place where we put some data, whatever kind of scratch stuff we want to put now in the new way, the new world, right, the new flexible scalable back end. We all have come to expect that repository for us. I should be able to collaborate on the things that are in there. I want to be able to share a file or a directory. I want to have access to that from my laptop, from my mobile device, from my tablet and from my desktop or workstation at work, whatever that may be. And so there are applications that sit in front of Swiss stack. And you can see some of the names of them on our screen right now that provide that functionality, that provide that interface to the end users with everything, all of that data actually living inside of Swiss stack on the back end. So we work with a handful of them, specifically as Doug did with Commvault on the backup use case. I'm going to go through a quick little screencast here, a recording of Storage Made Easy, who's been a partner of ours for a while. In fact, if you come to our booth, they're literally right around the corner from us. And it'll go through quickly here as well. But what you're watching is essentially you're going to see three interfaces. And what you see right here is the Storage Made Easy interface to configure it to .2 Swiss stack object storage. And what we'll build then is a container or the place where objects live inside of Swiss stack. And then Storage Made Easy understands what that is on the back end and provides an interface, their browser interface, which is what you see right now. Now you have this new container, my containers, is a place where your data is going to live. The interface you're looking at right now is the Swiss stack web console. It's another view into that same space. And you can see my container there as well. We're going to put some data in from a couple of different places. Right here, we'll drag it into the Storage Made Easy interface. We'll flip back and see from the Swiss stack web console that data is there. It flew by quickly. And the last piece, last interface, this is the one that you probably would expect, your end users would expect. This is configuring the Mac Finder, right? If you're a Mac user, then you say, I want to have a directory that looks like everything else I use. And I want that to sync automatically to that back end. I don't want to go to a web interface to put my data in and out. So due to a little bit of configuration here, and now I have this Mac Finder. I can drag data right into my shared directory. And we'll flip around a little bit and see that sync up again from the Swiss stack view of the world, the Storage Made Easy web view of the world. And it slows down again. So went through quickly, but a little bit of configuration gives you that interface. Likewise, if we had had the time, you could have seen that same picture of the world from a mobile device or from a tablet or from a Windows laptop as well, right? So that front end, that experience that you've come to expect is one that's very common for us with Swiss stack on the back end. All right, so we all know that you have legacy applications out there. And those legacy applications don't speak Object API today. In fact, they speak SIFs or NFS. And the reason you love Object Storage is it scales out. It's simple to administer. It's commodity storage. So it saves you money. And you have it in all these different sites. And you want to be able to use those legacy applications with it. And so what we're going to advise is using one of these best of breed NAS gateways and put that in front of your object storage. Are there other projects where you've got file systems in front of it? Are there other object storage companies where you have some sort of gateway built in? Yes, but these are best of breed. These companies just do it. In fact, we're going to go into a beer in their use case. They've been around for a long time. They've been accelerating sands and NASs out in the media and entertainment and genomic worlds. And now they can accelerate an object store. So one of the first things that we see is people looking to move off their traditional NAS. Well, with Swift stack and a beer, you can put a scale out of beer cluster in front of it. It will import all of your namespaces. It will accelerate your current NAS and it will move all that data over to your Swiss stack object storage. Now you could run your current NAS and Swiss stack behind it. And that way if your NAS has a single controller or single points of failure or you have to do an upgrade to it with a touch of a button, it automatically switches over to running against the object storage as the primary. And if you're looking to get rid of your NAS, you hit a button and this thing called flash move completes the process. And now you're running 100% off object storage. And as long as you're running 100% off object storage, what about backing up to the cloud? So there are multiple clouds out there, many of them here today, and they're offering extremely affordable storage in the cloud. Now maybe because of that distance in the bandwidth and whatnot, it doesn't make any sense for you to run against it. But you'd like to know that you have a second backup of your data. And so you can do the same thing. You could have your on-premise object storage or Swiss stack presenting SIFS and NFS in front of it. You can have that data mirrored to one of the cloud providers. At a touch of a button you'd be able to switch over should you have any type of outage on site. And you've got a backup of all your data. And now let's take that concept one step further. Hybrid compute. So I've got my on-premise data. I can mirror that to the cloud if I want to. But with my on-premise data I can now run with a VIR cloud compute against my on-premise data. So say you don't have a giant compute farm. You're running Hadoop or you're running Apache or Spark. You're running something highly intense in your HPV HPC environment. And you want a huge amount of compute you can scale up immediately. But you don't want to transport all that data to the cloud. No problem. Spin up in a VIR instance. It will offer SIF and NFS. It will connect back to your object storage on site. Run the jobs. Shut the whole thing down. Get a small bill. And keep control of all of your data. So we really see this as a best of breed approach. That NAS from a VIR and some of the other partners scales out. Object storage scales out. It's going to be far more durable and reliable than a lot of the other NAS solutions today. Much more so than some of the NAS solutions just bolting a object storage API on top. So Chris I'm going to let you talk about the world of other use cases. Sure thanks Doug. So we have we talked about several already and we're obviously moving quickly but these are hopefully planting ideas. Certainly we're interested and willing to talk about any of them in more detail. We talked about the obvious choice of Swift Stack being the back end for your OpenStack cloud right if you're using if you need Swift for a glance or for Cinder backups. It's a no-brainer. We talked about a backup target not just for Cinder but using it at a third party backup application. We talked about File Sync and Share and we talked about putting a gateway, a file system gateway in the picture to connect to some of your legacy applications. There are a whole slew more that we could go through here today. We put a couple of them on the screen for you. Customers of ours and partners of ours in the genomics life sciences space. So you can imagine the amount of data that's being generated today in the research area of things. Massive amounts of data coming off of genetic sequencers and whatever else going into Swift. Swift Stack as a back end for that allowing for the scalability that they need and even a pseudo CDN if you will. Researchers generating data in one place and then using the built in replication functionality of Swift Stack to have that spread out to the consumers of that. The researchers, the universities and medical facilities that are looking at that data as it's produced. Media and entertainment kind of similar, very different world but similar workflows as it pertains to data. Generating all of these cameras are producing a massive amount of data this week alone and this is being cut, you know, produced into to replays and smaller versions and it's going to go up on websites all over. All of that stuff is landing in Swift Stack in many cases working closely with a variety of media companies in that space as well. Doug mentioned a little bit about the HPC, kind of big data, a natural place for Hadoop rather than scaling out HDFS which could be on costly disk for data that you may not be actually analyzing. There can be a Swift Stack object storage tier that is the place where much of that lives in long term and you're just using HDFS for the actual analysis of data when it's relevant. So a variety of these that we won't spend much more time on today given the short window we have but a lot of good stuff that's going on in our world. Swift Stack just to recap. Software product, object storage use cases. We made some news this week and the launch of our release of our 3.0 product so feel free to swing by or check out our website with some additional detail on that. More integration with authentication, even enhanced management functionality and pieces there, more tie-ins to the S3 interface as well for applications that know S3 that haven't moved to Swift API yet and more and more and more. So let us know if there are any questions that we can answer or other work that we can do with you around these use cases or if you've got other cool ideas we're always open for that as well. Doug anything you wanted to add? No, thank you very much for your time Chris. Alright well thank you for your time Doug. Alright we'll talk to you guys after if you have questions we'll be on the side.