 From 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, Brian Grace Lee and John Walls. And welcome back here on theCUBE, our continuing coverage at the OpenStack Summit 2016, live from Austin, Texas. Deep in the heart of Texas, we are here at the Austin Convention Center. 7,500 strong, so quite a beginning from just six years ago, 75 people up the street in a small hotel. The place has grown, you might say. We're here to talk about the Product Working Group now within the OpenStack Initiative. And with us to do that is Carol Barrett from Intel and then Shamal Tahir as well from IBM. Thanks folks for being with us. We appreciate the time here on theCUBE. Oh thanks, appreciate the opportunity. You bet, let's talk first to Shamal if you will about the mission of the Product Working Group. Obviously it's to listen, but you've got a lot of feedback than to put out to the various working groups within OpenStack. Absolutely, so from an origin perspective, the Product Working Group membership is formed of organizations that either run OpenStack or build OpenStack-related products. And so the mission was to have a forum where we could define requirements that the markets have identified, enterprise, telco, and the operator community that's using OpenStack, and be able to document them in the community where all of the contributors can collaborate from a product management and feedback perspective to help make sure that those requirements are related to their project teams as well as evolve from concept to reality. And so then within that context, the members of the working group would be, how many and from what parts of that ecosystem? Carol, where do you draw from in terms of who you want on that committee and then what's again the range of verticals that you're talking about there? Well, you know OpenStack's sort of an interesting open source community and project. It's not about like a volunteer army. It really is a dynamic where 80% of the resources come from 20% of the companies. And so we wanted to harness that dynamic to enable us to be able to bring together the people who have close connections with folks who are in the market, either current operators or prospective operators or users of OpenStack and be able to get their input, but at the same time harness the resources that they have inside of their companies to actually put resources against those requirements to make sure that we're building and focusing on the things that are most important to those users. So how do you meet on a formal basis? Is it an informal basis? I mean, how does this interaction take place? Because I would imagine you've got a lot of balancing to do, you know, amongst the members of the working group and you've got different priorities within members of that group. So I mean, how do you all get together? And then how do you decide? So this is where we're going to focus. Yeah, so we use all of the standard OpenStack tooling. So we use IRC meeting. So we have a weekly IRC meeting. We do user stories that we contribute to GitHub and we do reviews on them through Garrett. So we use community tool chain, if you will, for collaborating on the output. But from a partization and meeting perspective, we also meet face to face, at least twice every release cycle to discuss, here's a top 10 user stories. Now each person stack ranked them one to 10. And then we kind of aggregate the scoring from that. And then we also discuss updates and optimizations we can make to the process, which is our workflow of prioritization as well as implementation of the concepts. So for people that are new to OpenStack, a lot of us who've been here for a while, you know everything, but there's still a lot of new people coming. Like for example, AT&T was talking today and they sort of threw up a slide and they said, well, here's the nine or 10 projects that are part of what we're doing, but then we've got three or four down here. There's always a couple of new projects coming along. Give us a sense for even for the beginners, how has OpenStack shifted from, we're just doing new projects because there's stuff going on to this idea now of like the big tent and sort of core versus give folks a sense of how the community is thinking about sort of what's core and everyday and then where's experimental areas and areas to grow. You know, I think that the user survey gives us some really good information about what people are looking at. You can start to see how the adoption is increasing across different projects and that guides what we go ahead and include in our roadmaps. So we'll target the projects that are thought of as core. So NOVA, Neutron, Cinder, Horizon, those and projects that have 10% or higher adoption by operators as well. Then we'll also include projects that seem like they're on the rise from an interest point of view around operators. Things around containers these days is all around things that are on the rise. And so we look to those things to guide us as to what's most interesting and important to the operators and the end users out there. And I would like to add on, you know, as you mentioned, there's a lot of newcomers still OpenStack. In fact, every summit, usually about 50% or more of the attendees are first-time summit attendees. So to help newcomers into OpenStack be able to figure out what's experimental versus what's mature, there's actually a tool that the Foundation created last summit called the Project Navigator. And so in the Project Navigator, they have a set of criteria that defines the maturity, whether it's documentation, SDK, whether the project has stable releases. So you can use seven or eight different criterias to evaluate the maturity level of the project. So that's, I think, a really useful resource for people that are getting into OpenStack and learning about OpenStack to better understand what's new versus what's fairly established. Yeah, I know Stu and I did a crowd chat we were talking on Twitter a couple weeks ago and that was one of the big things people were asking is, you know, is it a little bit higher level? Is it stable? What should I, what version should I run? That's a great tool, but it also helps you figure out things like, you know, there's a containers project for integrating and making containers a deployment resource, but there's also a containers project that helps run OpenStack and you can kind of figure out the difference between the two of them. I love that you talked about, you know, it's very data-driven, it's user-driven. How do you sort of balance between what comes from the survey and what, you know, some of the contributors can say, look, I can help move this faster, right? You're going to have expertise from a vendor or a large customer. How do you balance all that? Well, I think it really is the discussion amongst the team. That's what really sets our priorities and directions. What are the things that we're hearing from customers combined with what we get from other data sources and how does that go ahead and set the top five or six, what we call user stories, which are collections of requirements, that will go ahead and be a synergistic target for resources from the different companies in the work group. So it's really a combination of a lot of things. Ultimately what we're trying to identify are the things that are going to be most impactful in driving further development and deployment of OpenStack Clouds. I think within the group, you have almost your own user survey, right? Because you've got some very strong opinions and very experienced folks, obviously. But when you get this information and then you look at the user survey and now you've also got your own opinions, you ever surprised? They, wow, you know, didn't see that coming or didn't realize like containers, obviously, you talked about 70% in user surveys saying they have a strong interest in containers. Anything ever kind of catch you a little bit off guard? Do you think we need to gear up a little bit or maybe we need to shift our focus away from area that we expected or we didn't expect? So I think from a user storage perspective, I don't think we've seen too many surprising things. I think what we have seen though is a timing thing. So I'll give you a great example where capacity management, quota management is something that's been on our list of, quota management needs to be more standardized and more robust across all projects in open site because in the enterprise world, quotas are heavily used potentially. And so we had that planned and it was kind of on our backlog of user stories, but just this summit there's a team forming around centralizing a quota library for nested quotas around all projects. So now that there's development work already happening in the community, it helped, it made us kind of escalate our prioritization of that story so we can at least make sure we can share our data with them in a timely manner. And then when you get information at the end of the day, I'm putting it to use, where is it going? Like within OpenStack itself, within the initiatives here. I mean, which groups get, how do you determine what's going to go where? What is that relationship that you have with the other groups within OpenStack? How does that work? Well, we have a multi-prong attack to go ahead and try and get the right information to the right folks with the right level of detail at the right time. We are looking to use the cross-project team and their team meetings and repos as a key central point for disseminating information as well as bringing together experts from the different projects to be able to talk about how do we want to design and implement this capability across all the OpenStack services? So that's a key place where we start the conversation on a user story to be able to get the right people plugged in. Then inside of the work group, we also go ahead and have cross-project liaisons. So that's a community standard approach for helping projects that have some adjacent relationship to each other to stay current on what's being implemented and how they're moving forward and what their plans are. So we'll have cross-project liaisons in not quite all of the projects, but as many as we can, which is a pretty big number at this point, and they work to bring the user stories into those teams and to socialize them, answer questions, help bring out the nuances of what's going to matter most about that user story to the different project teams. So those are two key tools that we use to get engagement and to help move it forward in implementation in the projects. Yeah, well, one of the things, if you've been paying attention to OpenStack for a while, every six months, I mean, there's a drumbeat, every six months, that cadence has been, you know, it's never slipped, it's never moved. You know, the OpenStack Foundation is a really good case study in how to get software out, how to work collaboratively across a lot of teams. Do you ever get sort of end customers, if you will, that come and go, can you teach us what you're doing to make all this stuff? Because that's something that every one of them is trying to, how do I do continuous integration? How do I run multiple paths of software? Do you get that question ever or often and people want to learn not only just about OpenStack, but the people in process stuff as well? Yeah, so I do. You do. From inside of, actually I was having a conversation with folks from OpenHPC recently who were talking about, you know, what are some of the tips and tricks that we can learn about from OpenStack and how do we apply those to the community that we're building and the structures that we're putting forward in our organization and our community, you know, to help accelerate them. Yeah, so it is a common question that I find people ask from a lot of different perspectives, but it is pretty common. Yeah, yeah, I mean, it's becoming sort of a differentiator for business. People have to learn how to do software. Everyone talks about software eating the world. Yeah, it's a fantastic thing. What are you excited about? I mean, you get, I mean, this show is all about innovation, you get a chance to see all of it, but personally, either a project or something, you're doing, what's getting you excited, you know, at this, you know, 10 year, you know, 10, 10 anniversary of the event? You know, for me, I really like the fact that the maturity of the community as a whole, right? Like I love the OpenStack community. I love the collaboration, as you mentioned, and just seeing some of the foundational aspects of cross-project initiatives now going to a next level of sophistication, maturity, and ramping up, I think that's what I'm process wise really excited about is just the potential of, you know, capabilities of projects collaborating more frequently with each other to deliver even greater functionality for the cloud in the future. Yeah, for me, it's seeing the conversation go higher levels in the stack. So we've talked about the different infrastructure components and what they, you know, what the projects need to have and the capabilities, and now we're starting to hear more and more about the application developers, right? So not only are we creating the software and having clouds stood up with OpenStack, but now people are saying, how do we ensure that those cycles get consumed in a way that delivers value to the business? And I think that that is a sign of, you know, getting stickiness in the market and really broad base adoption. Well, let me just turn Brian's question a little bit on its side. I hate to end on a down note, but so let's look at it as an opportunity, perhaps, instead of a threat. What's that, uh-oh, or the, hmm, that you see that you do think needs to be addressed or that you think you can more constructively address in order to maybe grease the skids a little bit better and get things moving a little bit better in a different direction? What do you think, Shumal? I mean, I think my answer will probably be the same because I think that's partially why I'm excited about cross-projects is because it has been a need and at the same time it was a risk, it's an identified risk and now there's progress we've made against it. So I would say that's probably the focal point that I'm both happy about and worried about is collaboration at the cross-project bubble. Carol, you're not in your head. This sounds familiar. You guys compare scripts before you came out. I think for me it's building upon that. We're in the midst of looking at how are we going to evolve our workflow as a community? We have the two releases, we have two summits a year. When we started out as 75 people in the design summit that worked one way, now that we have, you know, hundreds to thousands in the design summit, how do we re-engineer all of that so we can be as productive and as efficient and innovative as we need to be to continue to drive the software forward? So I think that that provides us tremendous opportunity to figure out how we redefine it as we go forward but change is always, you know, also a risk. And so making sure that we think through it comprehensively and we put it into effect rapidly I think will be really important for us. And you just touched on that too, change, risk. How do you get maybe those who are a bit, they had there's some trepidation, how do you push them off the diving board? I mean, it was like even within your working group because I'm sure there has to be some convincing that has to be done with a few folks who might want to move a little slower than maybe the group might. How do you get them to that point, Shamal? You know, I think a lot of it is again, collaboration. So in the working group, for example, when we have that we've understood that we all have, you know, organizations that are consuming or building OpenStack products and we all have our own differentiated needs. But at the same time, OpenStack as a platform is a foundational layer for all of us. So there might be things I might want today that someone else might need tomorrow. But if there's enough of us saying that we need this today the person who wants it tomorrow will eventually say, you know what, I'll chip in today so you chip into something else tomorrow. So I think there's this give and take of, yeah, this was lower on my party, but if we can all make sure that we all invest equally in prioritized requirements, then we make a path for all of them to eventually be implemented. So I think it's that give and take of, I'll give today so I can get tomorrow. All right, yeah. And I think the marketplace has evolved. As a foundation, you're going as fast as you can and then the marketplace offers, you know, if I don't want to run on Mataka I can run on the previous release or two previously. That's okay, they'll support you. So the market has sort of worked itself out in terms of where do I want risk? Where do I want speed? And that's a nice thing to see it. You know, it's the way markets should work and communities should work, I think. Yeah, I think so too. And especially when you look at the markets that we're looking to serve, right? They're large and they're diverse. So being able to have a robust ecosystem that OpenStack can interoperate with so people can create the solution stacks that they need is really what's going to make the difference, I think for us. Very good. Well thank you both for joining us here on theCUBE. We certainly appreciate that and wish you continued success, both at Intel and IBM, but obviously with a working group as well. Thank you so much. Thanks for being with us. And we will continue with the coverage here of the OpenStack Summit 2016 from Austin here on theCUBE in just a bit. It's always fun to come back to the...