 Our last panel discussion here is along the lines of the theme of this week, which is that all of us working together can help ourselves and others at the same time. Our next panel has companies who compete fiercely in the market, Alibaba, Tencent, Huawei, JD, Microsoft, IBM. They're very fierce competitors, but they also agree that working in open source together makes each of them stronger. And so I'd like to welcome to the stage the executive director of the cloud-native computing foundation, Dan Kahn, who is going to moderate the panel on cloud-native technology. Please welcome Dan Kahn. We're going to dive in here and I'll introduce the folks as I answer the first question. I do want to share with the crowd that this is a very intimidating panel for me because almost everybody here is on the governing board of my organization, the cloud-native computing foundation. So these are all of my bosses. But I've still been trying to ask a few hard questions here and I'm actually going to go a little bit out of order in order to shake this up. Let me start with Michelle. Love your keynote yesterday. So you are on the Kubernetes Steering Committee, 13 people who run the Kubernetes project, and you're also the developer representative on the governing board. Do you hold those positions because Microsoft is a flat member of CMCS? That's a good question. You know, the Kubernetes Steering Committee, this was created based on an election that happened last year. And a fun story when I was deciding whether I wanted, I got nominated and I said whether I wanted to take that nomination or not. My co-workers were worried or like, you know, do you really want to take it? You don't have to, don't feel pressured. So it was really refreshing that even though, you know, this has been good for Microsoft, I wasn't pressured. No, it doesn't really have anything to do with that. I just really wanted to give back to the community that gave me so much and I want to help make that a positive place to be for a long time. Thank you. And Michelle, I just want to take a picture from, remember from yesterday, senior software engineer and actor, and COVID-stig apps and a helping cleaner, as well as those other positions. Todd, you're the chairperson of the CMCS governing board, and vice president of open technology at IBM. I'd like to hear IBM's perspective on hybrid cloud, where I think you're one of the few companies that actually has two different certified communities offers. Yes, and, you know, from hybrid perspective, having so customers as, you know, our clients and developers, often times in large enterprises, find themselves with the need to deploy on-preface and not share necessarily our input in the public cloud. But then they want the capability to be able to move workload into the cloud, use the same tools, get the same skills, build up using Kubernetes, right? And so from our perspective, making that available is really important. And so, you know, we have a program of having a hybrid cloud where we have, supposedly, IBM cloud private offering as well as public cloud offering. Again, it gives people the best of both worlds and the skills that they need to be developed and used across the whole cloud infrastructure. So it's very important, especially to the enterprise, to have those kind of capabilities, especially when you look at the financial services industry or insurance, et cetera, where data on breadth of evidence is used. Great. Jack, hi, our new board member from Alibaba, everyone I think knows Alibaba as the dominant cloud company in China. I'd love your perspective on Kubernetes here and what you've seen in terms of cloud native adoption here in China. Okay. So I think, you know, our containers and microservice and Kubernetes, they have been gaining a lot of momentum in China. And that's why Alibaba cloud also provides a full set of cloud native services to support this movement, like we have the Kubernetes container service in our cloud. And along with register service that will support their distribution and measurement of cloud images. We also have, you know, our function compute service in our cloud, which is probably a layer on top of Kubernetes, but they are also providing our cloud image support for our customers. But, you know, even though these momentum are, you know, going on among the development resources, if we look at the revenue generated by the business services in our cloud portfolio, you will still see that, you know, elastic computer service, our cloud storage, as well as, like, you know, the relational database service, they are still among the top ones in terms of the revenue. But, you know, we also find that some users, some companies, they buy virtual machines from us. But inside of the virtual machines, they are still deploying containers and a microservice, you know, those cloud native stuff. So I think it will still need some time for the end of the life to switch to the, you know, ready to use cloud native services provided by the cloud vendors instead of building their own stack. I think it's probably, you know, some concerns about vendor locking. And that's why I think standard in this space will be very, very important. And I was chatting with Jim. I think, you know, this Linux foundation and the sense that we get to play an important role here so that people will feel more comfortable to move to the cloud native service, you know, instead of using the lower level service. That's great. Thanks. Hi, thank you. From JD.com, you are the vice president of the technical infrastructure department there. And JD is the newest member of Platinum member of CNCF and our first Platinum end user. JD.com wrote a great blog post last year about your transition from open stack to Kubernetes. I'd love to hear how that is progressing and your thoughts on where you, the size of Kubernetes in JD.com today. In January of 2015, we saw that even though Kubernetes wasn't successful, in its design and in the future, it would be very promising. So we chose that technical direction. In January, we started to build the Kubernetes group. We cut off open stack and signed Kubernetes. After more than two years, we still have about 45% of open stack. And about 50% of Kubernetes. But the Kubernetes group is very convenient for us. So our second technical strategy is to provide the first tier service, which is the most. And then we have all the midwires, including cats, and a lot of new LAN storage. It's also very difficult to manage. The most important thing is the data layer in the back. Our database is very complex. It's G-Visible. It's also completely cloud-based. It's based on the Kubernetes solution. Even what we've been doing for the past 18 years is to do a lot of online and offline ads. We have a lot of big data computing. We're doing a lot of online and online computing. All of these are all based on the Kubernetes solution. If we use all of the Kubernetes solution, there's probably more than a dozen of them. And then there's about a dozen of them. This whole Kubernetes management machine is more than half of all the machines in our company. Half of it is like this. Thank you. And any line? You mean Huawei's cloud resource development team, and also the CMCF7 board? I'd love to take your perspective on two of these questions. One was the hybrid computing that Huawei also has two different certified Kubernetes offerings. And then I'll see your thoughts on uptake of cloud native in China and how that's progressing. Okay. So to answer the first question, obviously, you know, hybrid cloud is very important. I mean, it happens to our customers, especially the enterprise customers. A lot of them already have private clouds, and then they want to explore public clouds. And so I'll follow you to believe having one architecture, one API, and that kind of architecture would be best fit for our customers. So hybrid cloud is definitely very important. And this is why we invest a lot in kind of a hybrid kind of R&D and contribute to the community. For example, Federation cluster, that's very key. And there are other, and also multi-net working, multi-storage. All these technologies can help our customers move into a hybrid cloud environment so they can be more efficient and they can save costs. And to answer the second question about the uptake in China, you know, the thing is containers, you know, so pretty young compared to Linux. So there are a lot of different surveys. And I talk to developers, talk to users, and also I do some research on the survey. I think one of the surveys produced by China Information and Communication Institute. They just did a China survey of container usage in 2018 or caught native. And so it's very interesting, the kind of deployments, container deployments are most common in China are either very large or very small, either less than 500 nodes or more than a thousand nodes. So it's understandable that a lot of enterprises are supported by governments and these deployments tend to be large. So they have a lot of resources to learn new things and then adopt container technology. And you get a lot of small startups and they are more courageous and innovative and they want to check out container and caught native. And it's a medium one that's slower, a little bit slower. And also I still think there's a lot of problems. Well according to this survey, the three biggest challenges are number one lack of understanding and number two lack of use case of caught native and number three security. So I really think that NINF Foundation is doing a great job producing the websites and then translating the material in Chinese and having events here is going to help build up more education awareness and I think that's really good. And moving forward, we really want to focus on building out use cases that people can really leverage and so I think that's going to help increase adoption. Great. And finally, Liu Qin, we just enjoyed your keynote with the general manager of the mobile internet group of Tencent and also Tencent is the newest Platinum member of our parent, the Linux Foundation. I would love your impression of cloud native services in China today and how Tencent cloud is helping to address the demand. I would still like to return to the question. Because my leadership and management team is mainly Tencent's internal service and not Tencent's team. So it's hard for me to answer this question for Tencent. I think I can think through the technical value of Tencent's internal to think about the question. Since two and a half years ago, we have been using the culture to feel our values. At that time, we actually tested a lot of, tested a few, three, Accompanies, Docs, and at the end of the day, we chose to do the management arrangement by ourselves to write the management arrangement. Later, we also successfully signed all of our history and our team and our history services to Tencent. And we also achieved the contribution of the offline service and the offline service. For example, our offline service can be used in the daytime and then it can be used at night for the offline service. Basically, we also know that in Tencent, there are a lot of other business services and they will also provide the software technology. So they will choose to use Kubernetes as a base for their software arrangement. Last year, we achieved a unified structure within Tencent. We will provide a lot of free software within Tencent. After doing this, we will achieve a variety of technical teams and a unified structure with the same technology. We will complete this kind of work. In fact, we will probably have Tencent in Tencent and more than 100,000 and more than 100,000 and more than 100,000 and more than 100,000 within Tencent. These services are currently providing a lot of offline tasks such as smart training and video transfer processing to provide a very big and comprehensive solution. Through this solution, we can see that even in a company like Compute it is actually going to solve technical problems, the problem of the cloud life. The problem of the cloud life is that we have to solve the new technology and the old technology, how to complete and solve it. At the same time, we have to solve the external open-source projects and our own technical development and how to perfect the structure. I want to have a technical experience like this. If we use the existing workers and the existing corporate IT workers, I think this is a very good solution It really would be. That's fascinating to hear. I'm not actually familiar with how much Tencent has already adopted those cloud-related architectures. It is interesting, though, that across a range of almost every large company they've had to reinvent this same container technology and try and deal with the same approaches toward scaling. And now with Kubernetes and CMC effort, we share a lot of that expertise. This is a question for any one or multiple of you on the panel. But, you know, there's a term for China where people talk about the second mover advantage. Often to say, people say, oh, well, it's better to come first. But in a lot of ways, the U.S. and Europe wasted huge amounts of money installing, for example, baseline telephones all throughout the country where in China, because it's developed later, it's been able to move immediately to mobile phones and have a much more efficient investment. And I'm curious to hear seeing some of those same second mover advantages from some of the younger companies here in China that maybe have even skipped the whole generation of virtual machines and are able to move directly to containers and cloud meters. Yeah, I can go first. I think to me, it's less about the technology adoption centers because, you know, in recent years China has been catching up in technology so fast. If you look at the two types of major users about computing, one being the internet companies and the other being the big enterprise, they have invested so much in the recent years in China, too. So their legacy status is pretty much the same as their counterparts in the western countries. So to me, I think the difference really lies in the people. You know, people's mindset, people's skill and people's self-image and I think that's more important, you know. Do you think you are a leader in this industry and you want to find new technologies or you want to be a follower and see others going first and I'll follow that, right? In that regard, I think most Chinese internet companies as well as some pioneer big enterprise users like China Mobile, they have been so aggressively using cloud and internet technologies while, you know, in general, the majority of their enterprise users I think they are still lagging behind a little bit, but, you know, not so much, maybe six months. Yeah. So in the enterprises, they are more conservative. They still have legacy stuff. So they're still going to have a need of VN and bare metal for, you know, H2C kind of computing. For continual, cloud-nated, they understand they have to innovate. So generally, the customers, Huawei, we deal with large enterprise customers. They always have, you know, investments in all of them and they're definitely cloud-nated. It's very important for their 5G, you know, their AI stuff, but existing data centers, transformation, digitization, I think, VN and bare metal are still very important. So, you know, we don't use VN, we just use cloud-nated and bare metal. But from another point of view, for any big company, for example, there are a lot of companies in China that are very good at technology development. Many companies, whether it's BAT, JD, many of them, they have a lot of legacy systems. We have a lot of products. We have a lot of products that have some technical features. Then, in the company, we promote some changes. I think, no matter it's JD or Alibaba, many companies are the same. Especially in Instagram, there are a lot of changes. There are a lot of challenges, a lot of challenges. There are a lot of obstacles between the companies. Including Instagram, Samsung, there are a lot of challenges for the developers. But I think, in this company, Colony is the best. This company is the best. It has a very active role. Because it has a lot of products that are very active. It has hands, a lot of features. Later, many of them will be sold out. But in this way, it will form a set of products. For example, Instagram is a platform. It will not let us form a set of products as a standard. This is for both China, Europe and the United States. For many companies, the whole industry and the entire company will have a big role and become a standard in the world. This is why many of our companies, including Samsung, have a big role. I really appreciate that endorsement of CNCS. And I do want to emphasize this. It's why I'm here in China, why so many of you are here. And I will mention it more than one time. It's why we're so excited to have our first ever QCon, CloudNativeCon in Shanghai this November 14th and 15th. We will be there. Thank you. Hi. One of the reasons that I was so excited to come to China in particular was because when we first started the Special Interest Group SIG apps in 2018, I had reached out to some Chinese people and some people from Australia and I asked should we change our meeting schedule to better accommodate your schedule? How can we get you involved? Because I know that you're looking at the technology too and that was very early on. And the response like that was just record your meetings and we'll watch them and please take good notes and we'll read those and we'll follow along. And in the recent past I've gotten even more interest around engaging with the community. Some Chinese folks reached out and said we'd like to demo SIG apps. We'd like to be more involved. And so I was so excited and I am so excited to be here from a Kubernetes community perspective just to see what you all are into and what you're doing and how we can actually get you more engaged in the community and what your needs are because we want this part of the world to feel as engaged as possible. So C&P app is best known for hosting Kubernetes and we describe it as anchor tenants. But we now have 23 projects in the foundation and more coming in. I'd love to get several of your impressions on particular projects near to your heart for your companies and I might start with with my friend if you could mention the text because I think you guys have been some of the early adopters of that project. For C&P app when it comes to this project I think that Kubernetes is definitely an extreme project it's just a flat sheet and in our company it's also a lot of infrastructure and also we also have Helm Helm is currently a package management and it's a package management management and now in our community it's about 30% to 40% of the sales it's very good. And then we also use a very big Vita Vita is a before they developed YouTube they developed a 8-year Mexico group and Mexico scaling solution it's very mature from my personal judgment I think Vita is a a Mexican group or even New Circle currently the most stable the most mature the most reliable a a plan because it can buy a lot of a Mexican system it's it uses Mexico to do pure and complex this is for many big companies mature DBA group this is a very good a very good a very good a very good a very good a very good a very good a very good a very good and I think it's a very good a very good people while This is the most difficult work load, which is database. This is a hardware company. This database is one of its most popular data assets. We can use this database as a source of life. We really use cloud native database. This allows us to use TES, E2Y, database, including the back stage and offline tools, to be able to use Kubernetes to edit it. We really do our best to create our own database. Of course, we have been working closely on the foundation and evaluation of this new project. For example, ANWO. ANWO is the next generation of service providers. This is MAP. The service network is definitely in the next direction. There is no problem in the next direction. We are working closely on ANWO. We are still using ANWO. We are also doing a lot of optimization for ANWO. The main thing is to improve the performance of the database. We will also use some internal technology. Of course, ANWO is closely connected to the database. We need to use some fast-forwarding to improve its performance. In fact, in the end, as long as we can improve the performance of ANWO, I am very happy to see that ANWO's design experience and its application progress. Now, the key issue is how to improve performance. Of course, I am also very interested in other projects. A lot of new projects will be actively updated. Then I have a small idea. The direction of our efforts is... I think that ANWO has a lot of... in the computer management, the projects are very successful in this respect. The projects that can be improved later should be related to the standard of ANWO. But there is a way to... buy or use ANWO's projects, but this can also be improved. For example, there are some ANWO's document management projects or some non-circuit database projects that will allow the whole... the whole technology to follow up or the whole project will be more full. Thank you. That's great. I look forward to working with you on looking at some of the source solutions. I know Vitesh is one of our newest projects. It just came in this year. But to have a company in JD.com scale using it in production is a very strong endorsement for it. Now, Michelle, who mentioned Helen, so I wonder if you could comment on that and maybe just follow up on your keynote on this. Yeah, so I've been working on Helen for about two and a half years and it has been an incredible journey and we're just so excited to be a part of the CNCF because we think that the CNCF has a lot to offer and we've also been already taking advantage of the CNCF, so it just makes sense. Yeah, we started off as a Kubernetes sub-project. So if you go to Kubernetes GitHub or organization that wants to see how... We're going to move that to its own organization on GitHub and start making some transformations and some progress there. HoloLens has grown up so quickly and there's a huge community behind it and we're excited to serve that community and ecosystem even more. And also learn from Kubernetes. Kubernetes rubberally bounced and has had all these sailing challenges and has gotten a lot from CNCF too and so we've been seeing some similar patterns and we're kind of learning and trying to mirror that as well. Help provides a lot of value to the end user and because of that it really has come around really quickly and that's a thing for us as well too. It's a quite useful tool and anybody who's getting involved with Kubernetes really needs to take a look at it. Thank you. So if you look at the number of sessions at QCon for a non-CNCF-listed project I think ECO... ECO of this. And it builds on top of Android that I think mentioned. I wonder if you can tell us about IBM. Well yeah, and service methods become so important to have to manage all of the different services themselves for security and logging and all of the things you want to do to run that large microservices-based thing that you're putting out there and depending on is very difficult and being able to centralize that with the type of capabilities that between Envoy and ECO and other projects related to that provide you now take that into a central fashion and can much more easily control what you're doing. There's just tremendous value in that and that's why we believe that the whole service mesh world and that we will all come together around very quickly as well too because like Helm, ECO, and Envoy and other technologies just make it so much easier to operate and run at scale and it's scale that really counts the largest of us. I am going to add in one more advertisement that if you do want to hear more about the tests and Helm and ECO and of course Kubernetes and our other 19 projects that HubeCon and CloudNativeCon in Shanghai this November 14th and 15th is going to be by far the best place to learn about it. We're going to have simultaneous interpretation not just for keynotes but for all the different sessions both ones that come through the call for proposals and also the project sessions run by maintainers. And I do want to give one more quick ad to say that the call for proposals for that a conference is still open run as a community conference any of your organizations that are playing with these technologies that have run them in production even negative experiences that you've tried when you ran into certain bottlenecks or hurdles we would love to hear about it and so it's you can just search for HubeCon but the deadline is July 6th and we would love to get many, many submissions from Chinese companies talking about their experience. HubeCon Europe was a huge hit. We never had a company out of Epic in Europe to go and spend any time on it. The technology has really taken Europe by storm as well too. We've grown from 500 people at HubeCon CloudNativeCon two years ago to 4200 and 4300 and I do want to be clear that China's only the third conference that we're at and that we're so focused on opportunities here and working with the Chinese community the Chinese companies and especially and users that we really are excited to make a multi-year investment to do with that. I'll just add real quick as well I served as co-chair for the HubeCon last year and have a lot of experience reviewing CSEs which stands for if you want to ask questions about how to write a CSE or just want to chat a little bit about that and I'll be outside hanging out. I'd love to chat about that. And I think what's the spot of the submission is just that out of stress, right? So it's very short. It's pretty easy. Well unfortunately I need to end the panel here but I hope folks will come to Shanghai and I know many of you or hopefully all of you will be there. If I could just add it's hot to stay for a second but thank you all to the rest of our panel that was great to do.