 Austin, Texas. It's theCUBE, covering OpenStack Summit 2016, brought to you by the OpenStack Foundation and headline sponsors Red Hat and Cisco. Now here are your hosts, Stu Miniman and John Walls. And welcome back to Austin, Texas here at the OpenStack Summit 2016 here on theCUBE, along with Stu Miniman, I'm John Walls. And if you make something, the best way to find out whether it works is way to go to the people who are using it because they'll tell you, believe me, very quickly whether it's working for them or whether some changes are in store. And that's what the user survey that OpenStack publishes every year, actually twice a year, is all about. And with me to talk about that is Heidi Jo Trethewey, who was with the OpenStack Foundation and Heidi Jo. Thank you for being with us, or Heidi Joi. And you said, I love the way, by the way, I said, your last name said Trethewey like it takes your breath away. So it's very good. Well, this is the survey. 60-some odd pages, semi-annual. And before we get into kind of the nuts and bolts of it, tell us about who answered the breadth and the depth of the pool of respondents that you had here. Yeah. Well, you know, when we put together the cover, it totally made sense for us to have just as many users as possible on here. This is the largest survey we've ever done. It's 25% larger just even than last survey. 1600 plus people answered, representing more than 1100 companies. So that's really exciting to have that kind of breadth. We're not seeing it being overly weighted toward one particular size of company. We actually saw really good distribution of the size of company, everything from small startups, one to nine people to companies that say they're over 100,000. So really, really good cross-section of the community. Yeah, I think the one tidbit that related to that point was, I think it was 43% of respondents here, 1,000 or fewer employees. There was something along those lines. Does that sound about right? That sounds about right, yeah. Which I was surprised, again, it tells you about small, medium, large, which we're getting here. In terms of deployments now, what do you find out here about how many are in production and whether that's up, down, how does that stay? Yeah, well, I mean, that's our favorite headline here. Of all the things that we learned in the user survey, 54 odd questions. Our favorite thing that we found was that deployment, 65% of OpenStack deployments are currently in production or full operational use. And that is up 33% over just a year ago. So when we were back in Vancouver about a year ago at the summit, we were saying, rah, rah, we're at 49% of deployments in production and now we're almost to two thirds. So that's thrilling. I'm wondering if you could help us understand when you're talking to users, what does production mean to them? Because we kind of said it's, the way questions are worded sometimes, the way you build the tool, you kind of get a reaction back. So what feedback do you get as to what that means? So we actually added to the definition of production in this round because we wanted to be really clear that production means full operational use. Your cloud is fully operational. And then the other two categories that we gave people were either QA or testing stage as kind of a secondary next step. And then another step down from that was proof of concept. So we let the 400 odd people who answered our survey with regard to specific deployments actually indicate for themselves at what stage their cloud is at. Okay. And then in terms of one of the sections I found it's emerging technologies. I mean that tells you a lot about what's on people's minds, right? What's maybe not keeping them up at night but what's keeping them excited and what's optimistic and containers. It's all about containers right now because that was far and away the number one emerging technology. I mean what does that tell you about what's going on in the community right now? Well 70% of the users in the user survey said that containers as an emerging technology were interesting to them. And that outpaced software-defined networking and network functions virtualization and bare metal as well as the other two kind of top three interesting emerging technologies. So what that tells us is really the future of OpenStack. When we see people indicating 44% of people indicated they were interested in using Magnum as a future project. Right now very, very few clouds are running Magnum in production. But when we see 44% of the community indicate that they are interested in Magnum you can expect some major growth out of Magnum and then also out of some of our other services that have well over 30% interest. I mean how do you join? When I looked at it the things that's real telling when you kind of dig into it is first of all the core pieces Nova and Cinder and Swift have been there for a while but Neutron finally getting up there seems stable, seems like it's getting used but as you said expanding out using more pieces so definitely the maturity of OpenStack seems to come through a lot in the survey. And I hope you saw the part about we have almost 50 clouds that are running 15 or more different projects. So I mean there are six core projects so once you start running 15 projects you're going way down the line on some really experimental interesting emerging technologies. Yeah it is one of the things that you kind of find out and listen to the keynote this morning and looking there just the classes of users that are in there because not all users are created equal you know sure. You're a telco provider yeah I'm a user but you know there's only so many AT&T, Verizon and Swiss comes the world out there. There's the technology providers great to see SAP stand up or you know all the logos that are behind you but they're the technology creators. What about the enterprise? How many of the users there would you consider just kind of you know they're not the creators of technology but they are just kind of the consumers of technology? Well although I won't speak specifically to brands because I don't want to get in trouble with them. I will say that some of the best known brands the largest brands in the world are using OpenStack. We learned today in the keynote that more than 50% of the Fortune 100 companies are running OpenStack. I mean that is a phenomenal number. So you can't simply say oh this OpenStack thing it's isolated in IT or it's the domain of telcos or even just large companies because as you pointed out small companies are using OpenStack as well. You're seeing it pervasive in all walks of life whether you are going to Walmart to buy bread or whether you're driving your VW you're going to see that throughout. Yeah what do you see in terms of the breadth of response then and what does that tell you about the industries that are coming around and that are looking to be more aggressive in terms of their OpenStack deployments. I mean can you break it down to that granularity here to say okay we're seeing you know maybe whether it's healthcare whether it's government whether it's energy whatever it might be. Yeah in terms of significance testing on this survey versus the last survey I didn't see major changes in industry. I think that the industry that's probably the most interesting for us to look at right now is telcos and our most recent report on network functions virtualization goes deep into 10 different use cases 10 different telcos not just North American telcos I mean all over the world to dig into what they're doing with NFV but I couldn't break down for you specific industries or consumer oriented industries with major growth because the significance of the numbers isn't there yet which is not to say they're not growing I just can't prove it statistically yet. So I'm wondering was there anything that really kind of surprised you know you and the foundation when it came on a lot of these it's like okay 49 to 60 something percent is kind of the maturity we expect you know growing to you know 15 it's good to see but it's kind of that linear kind of progress is there anything that kind of jumped at you you know market shifts greatly or anything like that? Yeah there was one in particular because you're exactly right while we saw that lovely up into the right curve on the deployments in production for example the thing that was just absolutely a stunner so much so that I actually had to go back to our last report and dig into the baseline numbers and the Excel charts to make sure we hadn't done it wrong in the first place the big stunner for me was around business drivers that 97% of the people who took this survey so 1600-odd users told us that one of their top five drivers for adopting OpenStack was standardizing on that platform OpenStack has become the enterprise standard for infrastructure as a service and that's a really big deal and the way that I mean surprised is when 97% of your users say that's a business driver and only 60% of them said that just six months ago and yet I see very homogenous answers between this survey and the last survey that big of a shift means that something happened in the industry that people now believe that this is critical as uniting around these standardized APIs. There's also, there's some new things that got added into the survey this year the other one that jumped out at me is the title of cloud architect was thrown in and seems a lot of people jump towards that bucket what are some of the big deltas this year? Well the cloud architect actually was informed to us by the last survey when a lot of people used the other box to fill in cloud architect we also asked about chief information officer or IT infrastructure manager we saw more users doing that. I guess some of the other deltas I noted containers and I think I'd also like to point to the maturity of some of our top projects our most mature projects last cycle six months ago we were looking at about 78 to 85% adoption rate around most of the projects and now the top five projects are at 90% adoption or greater that's just really substantial uptake. So I'm curious cause one of the things that surprises me and it's not a red flag but it's a little bit worrisome is just open stack releases every six months and when I look at the spread of customers have it's like there's still a bunch in there running Havana and of course Ice House and here we are at the M's and every six months comes out and yeah there's this beautiful picture that you've got in there that they'll show. So yeah, what does that tell you people deploy it they don't keep up to date on it? Anything that the foundation has commentary around that? Your response? Yeah that's a really good observation that people are kind of spread out across the releases and we've seen that in every single release so it is expected. But in the last release six months ago what we saw is that it was concentrated around the three most recent releases at the time of the survey. This survey what it's concentrated around is the two most recent releases. So we're actually seeing a compression and people moving faster towards greater adoption. Now it's not really troublesome to see folks on the older releases because we know they're getting there we're seeing a nice wave toward greater adoption and we know that not all companies can manage an adoption every six months. As I said it's not quite a red flag but when we talk general trends if I'm using a public cloud and you ask them hey what version are you running on some big company's public cloud? I'm running whatever version they have me on and it's a new version so either IT needs to get in gear and move there but we're usually a lot more secure if I'm running the latest version I'm going to take advantage of features but six months is always an aggressive pace for something that somebody's going to take and deploy. Yeah, you said red flag and I really would call it a green flag because we're seeing that increased adoption that people are adopting faster than they were before. So we talked about the six month cycle. You're on your own six month cycle with the user survey. So how is what you have learned here driving that next cycle? How do you start forming areas of interest questions to try to broaden that user response group? All those things, I mean I assume those plans are full speed right now. Yes, actually on Thursday we're going to have a work group where we're inviting people to come in and comment and anybody can do it and also anybody can join the user survey common analysis team. We only require a confidentiality agreement to ensure that if something specifically identifies a user that they would be able to prevent that. So they would not share it if it's specifically identifying a user. So that said, what are we going to do in the next cycle? Well the next cycle might be six months from now or it might be 12 months from now. We have done a user survey seven times in a row since April 2013 and what we know is we have data between the last survey and this one of 2,500 users and that is strongly representative of the community. We're also seeing consistency across the data points. So we're thinking that it might be more appropriate to survey every 12 months, less onerous on the users, less onerous on us. It definitely, this thing eats my brain about two months every cycle, right? And if we did it every 12 months maybe we would see larger deltas in the data and that would be entirely appropriate. Also hopefully have greater responses. So Heidi Joy, you've actually worked for the last two surveys and you've added some with kind of the publication. Can you explain to our audience kind of how you're hoping to permeate it out there not just electronically but in other forms? Yeah, I think it's really important to be able to share things in as many formats as possible. So for example, this user survey is something you could just buy it on Amazon. We sell it at cost with no royalty toward the foundation. But I mean you could sign up and get it on your doorstep in a couple of days. My background in publishing enables me to understand how some of the print-on-demand publishers work and how the Amazon Marketplace works. So that enables us to actually have it out. And the cost is about $8 per piece. Contrast that to me taking it down the street to FedEx Kingo's last cycle when it was about $37 to print one copy of the user survey and look how much nicer it looks. So I really enjoy bringing some of my publishing skills. Not only did this but to the Enterprise Path to Cloud workbook. We not only published that as a print-on-demand piece that's also available on Amazon, but we did it in ebook format and in Mobi format. So you can read it on your Kindle, you can read it on your phone, on your tablet, or on your computer, really making it a flexible way to consume new content. Well it is certainly chock-full of information. The OpenStack user survey coming to a doorstep near you. Heidi Joy, thank you for being with us. A first-time CUBE visitor, by the way, I must point out, and I hope not the last time. It's been a pleasure having you on. Thank you, it's my pleasure. You bet, thank you. And we'll be back with more from Austin right after this. It's always fun to come back to the...