 Good morning, everybody. Welcome to your last keynote session here at ONS. I hope that everyone has had a great week. So, I'm back on the stage after our demo on Tuesday. I had the courtesy of VP of Community and Ecosystem for LF Networking. And I have an immense joy and honor this morning to be with a panel of ONAP leadership. And for I have to say one of the few times in my career, I have the joy of sharing this stage with a bunch of women. I have been the token woman on so many panels in my career. And it is so awesome and fun to be up here with all of you. So, thank you all. So, round of applause, actually. All right. So, I'd like to go through and introduce my panelists. So, first up here, we have Ola Goldner from Amdox. She's the director of technical strategy and standardization. She is a TSC member of ONAP. And we just sort of did our first TSC meritocracy elections within ONAP. So, I think that's something you can be quite proud of. And she is one of the chairs of the use case subcommittee. So, Ola, if you just want to say a few words about yourself as we start. Yeah. So, just probably something interesting that, you know, I, two years ago, then I decided that I want to change jobs. I decided that I want to try something new and go for an open source. And this is how I actually joined the Amdox. Haven't regretted so far. But for this journey, extremely interesting. And we are going to touch it and discuss it a bit. And also, you know, speaking of ONAP status and of our position and what we are doing in ONAP. As company representative and as women. Right. And next up, we have Catherine Lefevre, AVP of Network and Shared Services at AT&T, and also our brand new elected TSC chair of ONAP. And sort of our first TSC chair after our migration into meritocracy as a TSC. So, you know, I think that that is an awesome achievement. So, welcome, Catherine. Thank you so much, Heather. So, yes, it has been a long year since we have launched ONAP. And I've been since in the section. So, first of all, I was working on the open sourcing of the ACOM platform, which was the so far defined network platform that we were building in house. And after one year and I, I think we are currently working on our third release. And I have the honor, thanks to the vote of confidence of the community to be the ONAP chair. Thank you. And then next up, we've got Helen Chen, who is a principal architect for Global Technology Solutions at Huawei. And she is the PTL of the ONAP integration project. And has spent 18 years, I believe you said the other day, in the open source community. Eight years. Okay. Oh, eight years. All right. Thank you. Okay. Maybe there's a dot in my notes. Thank you, Heather. And also, good morning for everyone. Yes, 80 years. It's already a long time, not 18 years. Or, you know, maybe it feels like 88 years, but yeah. All right. 80 years in the open source community actually actively are playing with that. And our recent project is the ONAP. And also my other passion is in the innovation. And also I worked with a very broad range of technology. You cannot imagine, my first job was build a nuclear reactor. And right now in the city world also has been working with IT. You're right. Actually, I have more than 18 years working experience in the high tech. My current big title and passion is the PTL of integration project of ONAP. Also, thank you. Thank you. All right. Cool. And next up, we have Lingli Deng from China Mobile. She is a senior technical manager of open network industry. And sort of with the industry promotion lab. You help oversee that for China Mobile. And you are also on the TSC and we're the founding TSC vice chair, I believe, for ONAP. And used to serve as a board member for OPNFE as well. Yes. I think I've been very lucky when China Mobile and along with several other service providers designed you to actually invest more and even leading some of the open source communities to accelerate our network transformation journey. I was part of that since day one from our company. I started at OPNFE and then open all and now ONAP. So it has been a quite incredible journey and we now have a dedicated team working specifically on ONAP. And we see their great value of leveraging that and working with our partners across the industry. And that's it. I'm very glad to be here and excited that we have so many female leadership inside this community. Great. All right. And last but not least, we have Ji Hui Li who is a senior architect at VMware in the network intelligence architecture for the cloud group and you have been leading and being the PTL for the multi-vim project which is obviously I think very important as we look at architectural transformation ahead for NFV, SDN, and ONAP in general. Thank you, Lair. Actually, I'm very happy to hear to join all the excellent women's and actually I have been leading the development efforts and architecture of the NFV product in VMware and how to make them better integrated with the ecosystem. I have been very tightly involved into different kind of open source communities such as Big Data and OpenStack. And I'm very happy also the committers of OpenStack and ONAP also. And I'm very happy VMware is more active today to contribute more open source words. I'm very happy to be here. Great. All right. So now that we've all been introduced, let's continue on with the panel. So my first question actually is to you, Catherine, as a newly elected TSC chair with the exciting but perhaps daunting task of looking ahead over the next year, what are the sort of the biggest sort of either challenges or your hopes or expectations for success for ONAP in this coming year? Where are some of maybe the focus areas you plan to look at as a leader? Yes. So up to now I was focusing really on the execution plan of any release of ONAP working closely with Ellen, not only at the time of the integration, but also ensuring that what we are delivering is aligned with the expectation of the community. So what I will do as a chair is support it not only by myself but also by the order TSC members and don't neglect the fact that we have a high expectation of PTLs and they need our support as well as our subcommittee chairs. So with all this group of people, what we would like to achieve is continue to become a seamless ONAP in the sense that we would like to remove any identity or print flute from AT&T or China Mobile because we were working together to create the ONAP but we don't want that it becomes an AT&T ONAP thing or it can become a China Mobile ONAP thing, I believe maybe Ling Ling you agree. It should be a ONAP community things at the end. So it would be important to even if we have an end user authority group which is composed by the operators, it will also be important to consider what the third party vendors will be able to deliver to meet the expectation of the operators and also consider what the cloud infrastructure because it's a big ecosystem, there are multiple parties can support from a virtual network of possibilities because they are helping us a lot by giving us some free labs as well. So try to demystify, try to go to more meritocratic while we are still considering the operator needs because at the end we are delivering this platform for the operator and try to move in a more agile way. So far we are reaching milestone after milestone, it's a long cycle. I think there is an eagerness when you move to production to have more frequent build so you can have a kind of retrospective earlier, you can understand earlier what the future feature will be delivered. It will also help LN to be well prepared. So try to reduce maybe not the life cycle but try to deliver more often. So a lot of improvement have been discussed over the past days. So we will have our backlog as a TSC based on what type of improvement and lesson learned. We have collected since we are already set up since more than 1.5 years. Thank you. Were you about to say something Helen? No? I'm fine. Okay. But anyway, thank you, Catherine. Definitely, I really appreciate your help in the past and the OLAP could successfully deliver to you and right now in the third one. To yesterday's status, I was running the integration on weekly. We are still in very good shape and at least they won't give me a good night. So I appreciate and also definitely looking for next week's doubling will be more stable and lots of sense could be improved even more with your leadership. So next question, Lingley, coming from a service provider perspective, and we certainly hear from Catherine sort of as a TSC chair but also as a service provider, but specifically as a service provider, what is your sort of biggest desire out of seeing how OLAP develops over the next year? Right. So because I'm moving from the standards background and standards are very important from our telecom industry. So I think we see a service provider perhaps three layers of values that a community like OLAP could provide. The first layer is actually the open reference architecture, which I think OLAP has been doing very, very good job in actually unifying all the sorts about, you know, across the industry how the next generation operating supporting system would look like. The ideas or concept about separating design time and the run time and also the model driven approach and also the data collection analytics and policy driven automation cluster. I think those are the ideas that has been, you know, putting across the industry going anywhere, any organizations, any company, any service provider, those ideas are, you know, planned. And taking to China Mobile's case, we are actually doing our next generation OSS assistant design and also taking reference to that architecture. So I think that is a huge value, you know, out of this community. And the second level I think would be, you know, decoupling those modules and provide reference implementation of the workflow, the APIs and also the models between them. So it's not saying that the community has to deliver production level source code, but by having, you know, demonstration use cases and also, you know, testing intervability between different commercial solutions against this platform, we would know that based on this architecture, based on those specifications, those different modules could interoperable together. And it would project the future that the vision that, you know, all our partners could take a role in this big picture, but not having to dictating one service provider or any vendor to provide a net platform as a whole. So I think that is the second level of value that we could gain from this community. And we have been working hard, you know, with AT&T and also our partners to try to define a common like a suite of APIs or even, you know, providing flexible options to different providers to specific use cases about this so that we can all benefit from those reference specifications and enable intervability both internally and externally. And the last one I think it is a common value of the open source community is that if we have a mature enough source code, you know, service providers like us could actually leverage the source code and perhaps, you know, making our own solutions. That is something that we are starting to do, you know, this year. And like Dr. Feng mentioned in her keynote yesterday, we have been doing trials with some of the components that we picked from our net Beijing release. And we are doing trials in our use cases which is dedicated to core network virtualization. All right. So from a vendor perspective, you know, what are, you know, what are you hoping to sort of, you know, experience or kind of see the impact of Onap? Yeah, actually, that's a very good question because I think Onap provides a very good channel as an open community for different roles to work together and cooperate together and to understand each other better and to, you know, such as to contribute to the, you know, bigger goal together. So that's a very good channel. And we have so much, you know, new trends today such as 5G, edge computing, AI, all these trends, we need to work together to achieve them. And actually, different companies have different advantages and accumulations in different fields. And we cannot just, you know, work alone today. It's such, you know, a huge word. So we need to combine the advantages from different companies together to achieve a goal. And Onap could be a future defective, you know, industry standard. So maybe towards that goal, all of us can, you know, based on the same understanding to contribute that. For example, you know, VMware already initiated the multi-way project with several other partners together with the AT&T China Mobile, Intel, Microsoft, AMDocs together to, you know, provide a cloud agnostic support to the cloud, to the Onap. Actually, that's a start point. We try to contribute more to achieve, you know, a better future together. So. Do you have anything to add? I know you spend a lot of time thinking about use cases. Excuse me. Last day of the conference, my voice always goes. You know, sort of being down, you know, kind of in the use case discussion, you know, has that impacted, you know, sort of any, you know, your company's direction? Or, you know, has that, you know, had, you know, sort of positive or kind of unexpected impact, you know, for y'all sort of being in those use case conversations, you know, to your point of understanding, working together actually in the open source. Yeah. So basically, as I said, I'm chairing the use case subcommittee of Onap, actually since its inception. And, you know, at the beginning, we started from actually merging the code base of e-comp and OpenOR and showcasing it on some use cases which were required in the very first release. Again, the major goal was actually to showcase that the code base works. Those use cases were residential, VCP, Vault, and virtual DNS, virtual firewall. But then moving forward in Beijing release, the major focus was on non-functional requirements. And now that we are in Casablanca, and especially moving forward to Dublin, you know, which really gets us closer to some real deployments, really, you know, get to implement many more use cases which are demanded by industry right now and will be demanded in the future. And as a fact, yesterday, for instance, we had a panel on 5G and H automation, which are becoming very relevant. And, you know, we had, like, I think, seven sessions in parallel. So I was a bit concerned how many people will come to the room. The room was full, actually. So that only shows how relevant, you know, this thing is to industry. Because, again, we are moving to the real deployments, hopefully. And then, you know, the real feedback from service providers is needed. And actually, you know, what I was also trying to achieve, being a use case subcommittee she is to get as many feedback from service provider as possible, both in the proactive way, then they bring their requirements, and then they answer on questions that we might have. Now, as a company position, I think it goes in the same way. So, you know, of course, you know, each one of us works in the community, but also represents its own company. And leading on activities internally, you know, I come back and say, okay, so we have that feedback from service provider X, we have this feedback from service provider Y, those use cases are needed, and that somehow, you know, influences also our road map. And where we want to influence on, in which direction, clearly 5G and H automation are those areas that we are becoming more and more involved. Is an, I don't know, I would like to add something. So today, the on-app committee represent, I mean, a wireless provider that serves 70% of the global wireless subscriber. So it means that, yes, we are speaking about deployment in production. Of course, AT&T has a platform deployed since more than three years. I know, Ling Lee, you have a friendly user as well since a long time with the Open O component. And now you are integrating additional components. But Canada has already started to move in production. But we are only three. So we will need, and I know Orange just released an RFP to invite their vendors also to be on-app compliance. So one of the challenges will be through the use case community to the architecture is to understand what are we missing so additional carriers can consume it in their premises. And now it will really become a worldwide platform and already recognized as a standard platform. But in reality, it had to be in production. If I may add, I think the work and use case subcommittee led by Ella has been doing a fantastic job. Because on-app is a huge platform. And thanks for the contribution from also Katherine Steenback. Sometimes, you know, other service providers would struggle in understanding this power and how it can apply to their real, realistic use case. And use case subcommittee tried to collect all the ideas from different service providers and share their ideas and combine. Combine them and map them into functional or non-functional requirements with the help of architecture subcommittee, which kind of like feeling the gap between the service provider, they're talking, you know, in their language, in their requirements. But the PTLs behind them who are supporting, who will be supporting and delivering the features in the release cadence, they're speaking in their language. But use case subcommittee and architectures of the committee together, they're just bridging them and did a very good job. Yeah, I totally agree with Lingli because we have noticed that the PTL were working sometime in silo, but really focusing on their project and only start to interact with the other components when they need to have a flow that they need to validate together. So another opportunity of improvement will be to one juror, thanks to the architecture team and the use case committee to start to demonstrate additional end to end flow or more demo. So it can not only, we don't evangelize it for the platform, but people start now to understand how to use it or to tweak it from a VNF perspective. I mean, in order to, for the VNF vendors, they can already test their VNF before they commercialize them and sell them back to the carrier. So it's a kind of academy research environment. But again, we need to break this type of silos. It does not mean that people are not working together. That's not what I meant. But due to the fact that we have, I think, Lingli, we have 30 projects. As an open source, it's quite big. We know there are open source that are even bigger. But that's a challenge at the beginning, I think. But we are working well. I mean, it's a challenge, but it's nothing, it's impossible. One of the things I love about being an open source is that there always seems to be this sort of focus on continuous improvement. I think people, we have a certain, I think, pragmatism, I think often in open source communities where we're aware of what could be optimized. And sometimes when people here talk about this could be better, they assume we're talking about a problem when we really just always want everything to be better. I would actually echo for that one. Your use case subcommittee played a very good and important role to balance. Just like yesterday, we had a kind of long hour until this building kicked us out of here. I'll talk about the balance between new use case and also the S3P. Because on one side, we want, the OLAP has a huge codebase. On one side, we want it stable. And on the other side, the community and industry will not get excited if we don't have new use case added to support the new use case for each release. So the balance you guys need to keep. Of course, because OLAP is huge, just like Katherine said, we currently have 30 projects. And also, Summer project also has a sub project. And each project is kind of huge. Use case subcommittee, architect subcommittee, and also modeling subcommittee. This subcommittee, I think, is very, very important for the OLAP community in a way to define some kind of unified architecture to balance between those solos of each project, provide the use cases as a requirement, and also architecture, how to implement that, and to end the flow, how to interact from each other, and also the standard API of data model in between those kind of projects. I think those subcommittees are very important. And also you guys all did a fantastic job on that one. Of course, we could do better in the future. But the more use case I will have, but the more load we'll get to you for integration. Yeah. I was about to say, if we're talking about the number of projects, Helen, you do God's work. And, you know, the difficulty of the job being the integration, PTL, I think, is one of those things that cannot be overstated for any open source project. You have to help pull all that kind of stuff together. So, you know, how do you approach that, and how does that work? Actually, that's the interesting part, and also exciting part. Maybe I'm the person really like to take a challenge. I was earlier scanning for the audience to see if it feels over here. Last on this to the North American, we discussed OLAP is a fat baby. They're teasing about fat baby of OLAP. On one side, OLAP has a huge code base. That's very big. Probably that is the biggest open source I have been working with so far. But it's a baby in a way. It's not really mature or stable at this kind of point, as we expected, as people hope. I mean, the community are doing a very, very good job on this one. So, integrate them together is not an easy job. For the integration team, the team did a fantastic job. They need to be a cloud, I mean. They needed to be a network system engineer. They also need to be Java developer, Python developer, a lot of skills. So, you could put those things together. Thanks for the OLAP community, and since we have main developer more. So, I especially appreciate them for the whole support and collaboration, and also trust to put two successful release. Just like earlier, we said we're matching the third one. We already passed M4 milestone, and we're doing a good job. Of course, we could do better. Lots of ways. For example, the architect could design better to make all the projects could be more seamlessly working together. Right now, I definitely could see a lot of code is not really married together night, great. They are basically this kind of piece, and to support other use cases, sorry, we have some hard-coded code. And also, we have some kind of branch workaround on those areas. Definitely, we could do better in the future release. And I think we have confidence to make that better and also provide more automation. Earlier, I was responsible for Catherine talking about being more agile. She said it's helping us. Actually, it also has a higher requirement for the integration team and the whole OLAP as well. If we need to continue the release, even though that is our dream, that is great. But a continuous release requires very, very high automation and the whole tool chance over there with such a big code base. And also, so many things over there, it's very challenging. Like, I put the four layers of the test modeling in the OLAP community, unit test, CSIT testing, and also on the component there are like a pair line testing, and then we have end-to-end testing. I expect all to be automated. But right now, each of the layer, we don't get 100% which will make next layer even higher. Like yesterday, when we passed M4 last week, as yesterday when I checked, the progress of pair line testing, I don't see any results at this moment. So we have quite some work to do, could do better. Cross my finger. I think I've put in place already great tools. I mean, with the automation that your team and the community did, I mean, I'm thinking about the fact that now at any times, we know if all the labs are up, we have great, great graph that anybody can access from the community to see if the application is up in the different labs that we have. You also check if the container is starting, so focusing on the stability. So we are building up. So don't give up. I'm very optimistic on this one. Like yesterday, Orange Erica and his team demonstrated some very good CSID from the OpenFV community. We can leverage some of them just like earlier said, use their tools, could test if the cloud is ready to do some kind of basic check and layer by layer to check on sense and then could get some kind of sense move forward. We're definitely looking forward to that, but I just kind of say OLAP is very, very unique open source project. So we need to handle it in a relatively different way. We maybe needed to create something pretty unique, specific for OLAP. Even the industry standard IT is agile or continuous release, that kind of model. If that is a good one, at least I don't see that could happen in the next one or two release because of lots and lots of realities. OLAP as a platform is different than other open source. They could do a service like cloud native or those things, even though people talk about that one as a buzzword. But I think if this is a correct model for OLAP, because OLAP is a platform, we want it to relatively stable and also the dependency among that 138 Docker images currently, that's what we have and it's not easy. No, I had to say the operational OLAP manager project did a fantastic job to work on the containerization. Zingui, maybe you would like to share with us what you did with the multi-vim because if we are able to deploy it on different environments, I think there have been some work that you are leading. I totally agree because actually I come from IT world. I just decided this is a very good channel for everyone to understand each other better. So I think the use case-driven integration is a very good window to breathe everyone together. We already have a very good CICD infrastructure from different communities, Big Data Hadoop, OpenStike, OPNIFV. Actually, we can tightly work together to see if they know what's the best practice for this community and we can work together to identify our best way to move forward. And I am certainly a believer in CICD. I have gotten that religion completely, so it's great to hear that. We've had a lot of good discussion and one of the things that I really appreciate about this panel is a lot of times when you're a woman in the tech industry and you're invited on stage, you're invited to talk about being a woman in tech while you're talking about being a woman in tech on stage. You don't often get the opportunity to actually talk about your work or to be a woman in tech. You always are sort of in this meta space and we've just had, you know, I think 30 minutes of conversation with six women who are leaders in a very important open source community here, talking about our vision and our challenges and the technology and the use cases. And I think that's amazing, but I also think that we would, you know, sort of be remiss if we didn't take just a little bit of opportunity to kind of say, you know, it's a positive exception. You know, when Alla was proposing this panel, she was like, it's great because, you know, ONEP is a positive exception in that there are so many women leaders. There are a number in OPNFV as well, you know, and I'd just like to, you know, maybe, you know, ask some of the panelists. I'll start with you, Alla, since, you know, this was, this was your grand vision of this panel. Do you think there's anything in particular about ONEP or open source that you think has given rise to, you know, having this many women in influential positions? Yeah, so we've been discussing it really because I think it is quite unique in the industry that we have so many female leaders, but actually thinking about it, you know, open source environment is not only about technical side of the story, but mostly, I would say, about, you know, emotional and, you know, psychology, understanding people, working with people because, you know, you're not a leader because you were assigned by someone. Yes, your company can send you to work on open source. Eventually, your skills are determined, your skills to work in community are determined by how people, if people listen to you, if people are ready and willing to follow you and how they follow you, and whether you indeed, you know, have that force, you know, to convince people, not to force people because, again, this is not a company that someone appointed to you. This is a much more complicated society than you have to prove your skills, your social skills, also on the constant basis, and this is the place where actually women can play a key role because, you know, we do have those skills, I would say, so, and we face it really in Onup. In Onup, we do have all these women that you can see on the stage, you know, we lead the activities, you know, and in general what also I would also like to touch what can be done more because the fact is that we do have quite a few, you know, leading women. We do not have that many developers or I would say I would like to see more developers or I would like to see more, you know, women becoming PTL and even, you know, even more coming to the leading position and probably the way to do that would be, you know, having some joint activities, you know, putting some emphasis for that. As a fact, we had, you know, the day before yesterday we had a women lunch sponsored by Linux Foundation, I think it was great, you know, an opportunity to meet with each other and probably we have to do the same inside of Onup, you know, when we have developer forums, you know, encourage all those newcomers or developers, you know, to come to get to know each other, to get them excited about the opportunity which Open Source actually provides here, which I think is quite unique. Thank you. So, any other comments and, you know, and I also just any kind of comments on sort of culture? I mean, I live, you know, sort of very near Silicon Valley, there's a lot of conversation going on about women in tech in Silicon Valley. I think it's kind of fascinating, I'm the only American up here, you know, are there things that you think that you find for yourselves where you're coming from either company-wise or geographically culturally that, you know, you think are lessons that you'd like sort of the, you know, broader folks to take that you've maybe experienced from a positive point of view because, you know, how can we, you know, that sort of ethos, how can we do better, how can we be better? Perhaps not from, you know, female, from the gender side, but I see, you know, people from Asia, they turn to be more reserved and especially, you know, when they're from a technical, you know, background. Sometimes it could be, you know, hard for them to speak up and express themselves, especially if it is like opposing, like an opinion, and sometimes it's because, you know, they're afraid they would be making a technical mistake. But a personal experience for myself is that, you know, it should always pick up if you think differently. Otherwise, there will be no meaning for you to be within this community. You are here because you're representing yourself and also the company behind you and also the partners behind you. So you are here, you are welcome here because you will be adding value by speaking up. So I've always, you know, encouraging, you know, our Chinese participants also from my partners or inside my team to speak up. And sometimes we are very appreciated that, you know, people from Western world will be very patient in, you know, accommodating our, you know, not-so-good English and hearing our opinions. And I think there are a lot of foreigners in our lab and other communities that more and more Asian participants and also some of them are, you know, in the leadership position. So their voice needs to be heard and it has been more and more, you know, appreciated. I also want to add a little bit, yeah. I feel as a woman working in the high-tech, very lucky of this sense. And also, I guess, most of the women in here as well is like the support we get from our families and all other things. Like when we are here, our families take care of our kids and our cats. Everything, that kind of work. Definitely appreciate only with their support. We could do whatever we need to do. A lot of time on the society, the reason why from, actually, from the elementary to from the undergraduate school, you could say in the STEM field, there's much, much less women is because the society, the parents, a lot of the environment, the influence who caused that. But I feel I'm very lucky to get the courage to pursue my dream. And also, secondly, I really appreciate my company to support us to work with that. You know, that's a lot of investment for the integration team to put full-time job to basically purely work for the community to get all those kinds of things to happen. That's a lot of commitment. So I feel very appreciated. Certainly, I also wanted to encourage the women to do things. A lot of time, I feel the open source. It's probably, it's a very good career for women to choose because we are good at whatever social. And also, when we are even already in the high-tech world, technically, we're strong. And we could do a very, very good job in this kind of field. And also, as a result, we could also influence other women. They could participate in the high-tech world more. That's probably some kind of contribution we could do for the society. All right. Well, we are out of time. So I would really like to thank all of you for being up here. Let's give our panelists a round of applause, please. Yeah. And I just hope that all of you, as you go out, just always be aware of how you are making the community more welcoming to people who don't speak English as a first language, who are women, who are from different racial backgrounds, who are LGBT. We are greatest when we are, we are greatest and strongest when we are diverse, inclusive and hear from everyone. I have been very impressed with the open source community that I have been in for the past four years. I think it's amazing. And let's go out and make it even better. That's what we do as a community. So great. And we are now, if y'all can go on and leave the stage. If you want to, y'all can go on and go. That panel is over. Thank you again.