 The Cube at OpenStack Summit at Atlanta 2014 is brought to you by Brocade. Say goodbye to the status quo and hello to Brocade. And Red Hat. Here are your hosts, John Furrier and Stu Miniman. Okay, welcome back everyone here live in Atlanta for the OpenStack Summit. This is The Cube, our flagship program. We go out to the events to strike the signal from the noise. I'm John Furrier, the founder of SiliconANGLE. I'm joined by my co-host Stu Miniman and also at wikibond.org. And our next guest is Guillaume Obachan, CTO of Digital Film Tree. Welcome to The Cube. Thank you very much, glad to be here. You're up on stage there talking about OpenStack. First tell the folks out there what you do. Very interesting business. We love it because it's The Cube. Obviously we love media, we love what you do. So explain what you guys do. So we're a full-service post-production facility for TV and film. And that means we do the whole finishing process from camera acquisition to distribution. And we do shows like NCIS Los Angeles and Modern Family. And we did her last year with Spike Jon. So that's what we do. And about six years ago, seven years ago, we really started to see that we needed to write our own software in order to further our business and continue innovating in post-production. And that's sort of what led to our involvement in OpenStack. I love going to the NAB show when Intel's there. They show up all the fancy geek stuff going on. You know, I bet people don't think about the tech involved in Hollywood these days. What's it like there now? I mean, how much tech? I mean, obviously you see all the blockbusters out there. It's all special effects. Everyone sees that. But there's some really good innovations that are under the covers. Explain kind of some of the cutting-edge stuff in Hollywood. Yeah, I mean too, in this day and age to be a great filmmaker, you really have to be a superlative technologist as well. We're shooting digitally. We're doing visual effects digitally. We're pushing content around. We're shooting television shows and films all over the place, across the country, across the globe. So really to do the whole workflow and to get the best creative experience, you have to have your hands really deep into all aspects of technology. So as an artist, and you also got to be a coder possibly too. So knowing some Python, doing some data analysis. Is that the kind of the scene there now? Absolutely. I mean, we like to say that we're entering this era of software-defined post-production. So to manipulate that pipeline and to serve the creative process in the most efficient way possible and to really maximize your dollar on a television show or a feature film, you've got to be in there. You've got to be writing custom code. Every TV show and film is a beautiful snowflake. You're starting to see in high school programs and now even college, just the evolution of journalism and now film is converging. It's not just journalism anymore, film making. It's a lot of science pieces. So art, they're kind of calling it media arts and sciences. That's kind of a new genre. Explain what that means for folks out there who aren't in the business. Big trend towards where the technology intersects with the creative process. What's going on there? Why is it so killer right now? Well, I think video is a beautiful way of communicating. And as video has sort of been democratized into the hands of more people through technology, they can communicate with video just like filmmakers have been doing for 100 years. And it's really been pushed into the masses. And the tooling is out there. What's the state-of-the-art on tooling? It's getting in the hands. And obviously, people think of vines and whatnot. But high-end stuff now is coming down to the footprint size for kids and adults to actually play with the technology. Is the tooling sufficient right now in your opinion? Absolutely. I always like to say that the iPhone shoots 1080p. And a couple years ago on NCIS Los Angeles, they used iPhone footage. So you've got in your pocket the same cameras that are being used to shoot multimillion-dollar television shows, even more so with DSLRs that are on the market. And you can craft a video at the same level that we're doing on large-scale projects. So Guillaume, you guys went down the path to build your own software. I'm curious if that's difficult to maintain. Because I've talked to friends of mine in the media business, and they say, we had a need. We had to do it. There was no other option. And then you go three, four years down the road. And there's stuff that's off the shelf that might be able to be cheaper, faster, and easier. But this is the way that we do our processes. Can you speak to that? Sure. I mean, there was two major motivators for us. I think we had grown weary of being battered around by companies that were changing their software or going out of business and being subservient to their software and their workflows. And I think it is difficult to maintain. But we really came to the conclusion that to grow the business and to be financially successful, even in post-production, we had to iterate and create our own software to deliver a unique experience that only we could deliver. So on the OpenStack side, most people think of, oh, Big Bank is using it. You guys aren't like a huge bank, right? It's like a huge customer, I mean, huge footprint. But you're an active customer. Does OpenStack doesn't matter how big you are to have OpenStack? I mean, you guys are a good example of size. Yeah, we're a 40-person company in total. And there's about four of us who deploy and maintain OpenStack. And OpenStack is so agile, and it's so, you can evolve it and mold it to your needs. So no matter what your size, it can make you more efficient if you utilize it correctly. Talk about some of the technical things going on with your business. I'm in the cloud, so you talk about the tooling, creative process, which I could talk about that for an hour. It's so interesting, and I think it's very relevant, especially for the young kids coming in. I think what you're doing is fantastic and I love it. But there's also back-end support. You need to have infrastructure to run it. So you need state-of-the-art on the delivery side. You need its workflow, and I can imagine if you're doing end-to-end delivery, it's probably very complex. What are you using for storage? Just take us through the Sausage Factory that is your business, you know? I mean, as you make the product. Sure, you know, in a very basic level, we started out by needing cheap, reliable storage. We had shows that were shooting 50, 60 terabytes of data a day. You know, it's going to even grow larger into the future. So, you know, we started by implementing Swift Storage as a way to not only build out large amounts of storage at digital film tree, but also build out nodes of that storage on site. So we have two shows right now, Mistressism Perception. Each of those shows has a Swift node in editorial that's backing up that raw camera data. They have access to their own raw camera files, raw video files, and that's being sunk back to digital film tree. So that was like our first most basic need, and it sort of grew into this idea of how could we push content, how could we do shows that were, you know, across the country, across the globe, and sync that material and have instantaneous access to that material. I think that the key thing we were being asked is, you know, especially in terms of video, is people were wondering creative individuals on these shows. You know, I have instantaneous access to the content I produce on my phone. Why don't I have instantaneous access to the video that's being shot on my television show? So we really want to deliver that instant access, that same feel that they can do with a Vine or with a Twitter, but in their professional creative environment. I noticed I was watching some of the hockey games with big Boston Bruins fan, playing the Canadians lost game six, but Hockey Night in Canada does really instant replay clips on Twitter, which I find fantastic because you know, Stu and I were out last night, we couldn't watch the game, but on Twitter I'm getting really epic footage almost instantly cut and deliver to the Twitter feed. I mean, that's kind of where this was going, right? That kind of instance. So do you have the tooling on search? How do you find the clips? Is it indexed? Is it, do you guys write that software? Is it off the shelf software? Well, you know, we've built a lot of software ourselves, but you know, we're using a lot of no sequel databaseing and OpenStack tools that allow us to search large amounts of metadata. And really the key is that, you know, we have an infrastructure with OpenStack where we're aggregating all that metadata too. And so, you know, we're using all that metadata to build associations and give our end creative a way of saying, you know, where is this actor in a shot? And they don't need to know the shot number, the shot name, or the date it was shot. It just goes and finds that material for them. So, you know, really through metadata aggregation, we're making that process much more efficient. What are the biggest challenges you see going forward for you guys and opportunities as well? I mean, obviously the tooling's being baked out, OpenStack's growing, it's evolutionary. It's a continuum process with the cloud. What are the, what's on your wish list? Maybe the better questions. Tell us your list for the top wishes right now. I would say bandwidth, bandwidth, and more bandwidth. We're at a point where cloud and OpenStack and storage technologies are maturing. And, you know, we have great partners like Rackspace in the public cloud that we work with. But, you know, transporting, you know, 60, 70 terabytes of material every day is doable but cost prohibitive. So, you know, really transport for us is the big bottleneck right now. Yeah, I'm great to have you on theCUBE. Love the keynote presence and I love what you're doing and I just want to get some personal comments obviously in the Bay Area where you grew up. We take a lot of interns in and I give talks to kids all the time about the future of journalism and film. What advice would you have them? You know, for the kids who really have a desire, there's something, doesn't seem to be a track anymore. The old tracks kind of lead you down kind of the old way, you know, the old school high presale. This is the way you do it. You're in the new school. You're doing some stuff that's cutting edge. What would you tell those kids who say, hey, I want to do, I want to play, I want to learn and I just want to get involved in the direction of filmmaking? What do you share with them? What do they do? You know, I think it's a very old adage but if you learn how to tell a great story, then, you know, no matter what you do, if you get into the technology side of it or the creative side of it, you know, that will dictate your entire career. You know, if you know how to tell a great story then creating a workflow around technology to tell that story is just the same as writing it down in a script. Yeah, so one of the things we hear at this show a lot is, you know, everybody needs to code. Everybody needs to contribute. You guys obviously believe in contributing. You know, can you speak to that following up on what you were just saying kind of career path wise? Sure, you know, getting involved in a community and contributing to that community and having valuable feedback come back from that community is I think key in anything you do. And certainly, you know, writing code and understanding the very building blocks of the technology you're working with is key to a great career. Yeah, so, you know, one of the concerns we've had at the show is that there's not enough of the contributors and users of OpenStack that are talking. So what would you say to your peers about that? You know, what has it done for you to not only contribute but, you know, come speak at the show and, you know, how would you recruit others? Well, you know, I would definitely say for me personally, I mean, it's been a whole breath of fresh air into my career as a whole. I brought up at the keynote this morning that I read a Yahoo article about four or five years ago that the post-production industry was going to be one of the businesses that was gone in 10 years. And really redefining that around software and finding a new path around OpenStack and the community was really a reinvigoration for me. And, you know, we're living the same path ourselves with journalism. People said journalism is dead and, you know, we decided to invest more in the video operation, which is essentially live. And there's no post-production goes right to YouTube. When we're done, essentially it's stop and upload. You can probably do some more work with your mentoring, of course. But journalism is supposed to be dead, too. I mean, it doesn't really die, it really changes. I mean, that's what you're saying, right? Right, it's great storytelling. Storytelling lives on. Good content. It's been going on since the beginning of time. Good content and good creative. Always wins the audience, as I always say. You know what I mean? Thanks for coming on theCUBE. Really appreciate you sharing your story. Great success. Love what you do. Creative and art and science coming together. It's social science. It's computer science. That's our motto. This is theCUBE. We'll be right back after this short break.