 Live from Boston, Massachusetts, it's theCUBE covering OpenStack Summit 2017, brought to you by the OpenStack Foundation, Red Hat, and additional ecosystem support. Welcome back to theCUBE's coverage of OpenStack 2017. I'm Stu Miniman with John Troyer. And John, you know what they say. If you're somebody without data, you're just another person with an opinion, and that's why I'm really excited to have Heidi Jo Trethewey on, who is the senior marketing manager of the OpenStack Foundation. We are swimming in data. How many pages, how many users? So much information in here, Heidi. I've been swimming in it, digesting it as we do every six months when you guys put this out. So, first of all, thanks so much for joining us. Thanks for bringing us some pretty tables and charts and lots of data to swim in. It's good to dive in and to have some really good visualizations too, which is, we have some new visualizations here too. Okay, so let's start at the top. Biggest survey ever, the number of users, number of deployments. Greatest number of deployments ever catalogued, and also 22% more organizations than our high-wide mark, which was the survey of a year ago. So, we've been doing this nine times now. It's our ninth semiannual survey. And I think that what's most impressive is, you know, we were proving different things early in the study. You know, we were showing OpenStack is production ready. It's enterprise ready. And we really emphasized the high degree of deployments in production, about two thirds of OpenStack deployments are in full production, full operational use. And then in this survey, we saw that inflection point where it's really taking off the number of deployments growing, that 44% more deployments than ever before. But then the corollary is you see the age of deployments is less than 1.7 years. And so I think that's really an exciting number because it says that OpenStack is growing and flourishing, people are launching new clouds. And recently, and so that speaks to the health of our organization long-term. Yeah, and how many are the existing people that you've seen them move along in their maturity and their deployment versus new ones that are coming into the survey? Well, a little more than half of the deployments were launched in 2016 or newer. And remember that the survey cycle time was only the first two months of 2017. So we're really looking at clouds that are between two and 14 months old, or maybe zero and 14 months old. And of those, more than half of them are already in full production, full operational use. And then if you look at the older clouds, those launched between 2010 and 2015, well north of 80% of them are in full operational use. So it's just kind of interesting to see clouds as they move through those phases of proof of concept, test, and then full production. What always impresses me is, this wasn't just five or 10 questions. I mean, you get in through everything from what kind of use cases they're doing, what vendors they're using, what are the cool new use cases there, what is new this year, any new questions, or any new data points that we definitely want to highlight? Yeah, well, the cloud age is one of my favorites. But my absolute favorite chart, which is brand new, is the question about what percentage of your overall cloud infrastructure is currently running on OpenStack. And so we asked everybody with a deployment about 600 people and they told us that between 61 and 80% of their overall cloud infrastructure is running on OpenStack. And it's a scary time to ask a new question. And so for this to be a new question and to have such a really high number coming back is really exciting, because it tells me that OpenStack is not just being used for some new applications or some fringe use cases, but it's actually permeating the entire organization, including legacy applications. And then when we also sliced and diced those numbers, went from looking at the full group of people who had deployments and then looked specifically at those with clouds that were more than a thousand cores, the amount of users who have OpenStack in deployments is actually even higher. They said 81% to 100% of their total cloud infrastructure is running on OpenStack among those large users. You know, I want to take it actually the other direction. I think everyone knows at this point, let's hope everybody knows that OpenStack scales and OpenStack is good for large deployments in the Fortune 100 and at service providers around the world. An interesting element to the survey I thought was that 25% of the organizations were actually very small and they were also in production, so it's not just for a team of rocket scientists. These are small organizations that are deploying OpenStack. Can you say anything more to that range? Yes, you mentioned the 25%, those are companies with fewer than 100 employees. And then we also looked at the largest users, more than 10,000 employees. That's about 32% of the group. And so I think actually the thing that I hear most frequently from media or analysts, they're asking us, what about the mid-sized users? How well are they using OpenStack? And I think it's interesting to see that actually the largest portion of OpenStack deployments are being used and operated by mid-sized users who are showing us that OpenStack has these capabilities even among smaller development teams. So while I can't point to a specific use case, of course, one thing I'd like to do is to use our analytics dashboard and you can apply these different global filters. So for example, you can apply an organizational size filter to learn more about any one of the questions on the user survey. One of the, I guess, pushbacks we've had a little bit, especially in the last year, is the core versus the big tent. We've had a number of interviews today that said big tent was a little bit too much, need to focus on the core. I'm curious if there's any data from the survey that says what customers really care about is the core or are there some of the outlying pieces that they find pretty important? Yeah, that's a good question. And when you look at our core services, they are being adopted by at least 89% of class. So we actually looked into project adoption and we also looked at projects of interest, like to see certain projects that are increasingly of interest to folks like bare metal, for example, or database services. Then we also looked at the number of projects per cloud. And if you are thinking, okay, we have six core services in open stocks, so maybe the average is six or seven, actually the average is nine. And for many clouds, they might be running 12 or more different projects within their cloud. So I think what that says to me is that, yes, the core services are useful and established, but there are always these use cases that want more. And the users are continuing to push for more capabilities and different capabilities and combining open stack in different ways. And so I think it's kind of exciting to see how the user survey proves that out. Okay, how about things like containers, some of the emerging technology, some of these questions you've asked in a few surveys now, seems like rapid adoption from the little numbers I've looked at, what can you, what do you say? Yeah, we're not asking people so much about their pace of adoption, although we did see in a 451 research group analysis that people who are running open stack are actually adopting containers at a rate three times faster than those who are not running open stack. So I think that's a very interesting data point. What we asked on containers is, first of all, we were asking about emerging technologies and how often, or excuse me, what were the emerging technologies of most interest to our users and containers top the list again? So that's no surprise. New on the survey this year is asking about Docker, for example, or excuse me, asking about using tools with containers to containerize open stack and Docker top the list there. Kubernetes continues to top the list of containers or pause tools that are used. So you can dig into a lot of different ways that container technology is being used with open stack. I think the important takeaway there is that it's all still evolving and so we're trying to express that data in a relevant and meaningful way, including showing a 12 month range and not just a six month range of user survey answers. Of course NFB is another area of intense activity and interest in the open stack community and amongst open stack deployers. Can you speak at all either to activity in that area or also to the split between more enterprise-focused clouds and service provider-focused clouds in terms of deployments? Well, in terms of NFB, actually one of the big highlights I saw is out of our NPS scores, where we asked what was the basis of your reason to recommend open stack and a lot of users mentioned that they specifically recommend open stack for NFB because it is proving out as such a consistent technology for NFB. In terms of differentiating between service providers and private clouds, for example, you heard about our new remotely managed private cloud offering in the marketplace on openstack.org and then we also have additional data about private clouds, community clouds, public clouds in the user survey, not seeing significant differences there. Does open stack global community, is any changes in that global distribution this year over last? Well, it's the most international survey to date. Actually, I think it's about 61% of our users and 74% of clouds are physically located outside of the United States. So we just see increasing internationalization and I give you a little preview of what I'm hoping for the next user survey. We're actually working with open stack translations team that normally translates about a million words worth of documentation for every release cycle and we're partnering with them to actually translate the user survey to make it more available. First of all in China and Japan and Korea and then hopefully worldwide, I'd really like to see as much input as possible across the globe and not just concentrated in the US. Curious, there's a lot of talk about where open stack fits in a hybrid cloud deployment, anything North America versus overseas that you saw so many high deployments of open stack overseas that you've got the, 70% is on-premises, does that differ across the globe? I know that we have, there are quite a lot of fabulous public cloud offerings in Europe that are available, but in terms of the user survey data, I don't think I can correlate a direct answer for you. Okay, and last question I have is, how do people get the survey? Do you take input from community people as to how this gets, the next iteration gets done and can you get the source data for this? That's the magic question. And so I'll answer you a couple of different ways. At openstack.org forward slash user hyphen survey, you can get not only this book, but you can also get the access to a short video overview with like the top 10, 12 charts that we have available. You can also go to openstack.org forward slash analytics and use your own kind of data scientist bent to apply different global filters to the data. And then to tell you a little bit more about how it's created, both openstack's user committee, which is now an elected body, as well as a group of community volunteers, about a dozen of them from 10 different organizations all contributed to analyzing this report, vetting all of the charts, and going through about 1800 different verbatim comments to actually define what was going to be showing up in this survey. And so I really appreciate that user survey workgroup. In terms of getting the individual specific data, you're going to need to volunteer for the user survey workgroup. We require everyone to sign a confidentiality agreement because you'd be surprised how often users mention their own company name, and because we've promised confidentiality to the people answering the survey, we can't expose the verbatim data, but that's why we're trying to find that happy medium with the analytics tool online. Hi, DeJoy Trethaway. Really appreciate you coming in, and there's beautiful stuff in here. I mean, if you love data, if you're part of the openstack community, highly recommend that you check this out. John and I are going to be here to wrap up day one of Openstack Summit 2017 here in Boston. Thanks for watching theCUBE.