 Thank you very much for this opportunity to share what we're doing the Open SES project with you guys. So let me explain a little about what the Open SES project is about. It's actually a Venus Foundation project that a lot of people don't know about. So that's why I'm here to talk to you guys. And the Soda is actually a rebranding of the project or a change of the transformation of the project from Open SES to Soda. And I'll explain later why we are doing that. And part of the reason is because as the title says, we are going on to the path to data planning. A lot of people may not have heard of FutureWave. So FutureWave is actually in the news every day and CNN and so on. So FutureWave is the US R&D arm of PowerWave. So I'm basing on Silicon Valley, the US R&D arm of PowerWave. So if you look at this picture, this shows all the kinds of popular applications that everybody runs nowadays. Big data, IoT, machine learning, and AI. And the data that you have usually is either stored on premise or in the cloud. And on the right hand side, your applications may be running on containers, virtualization, cloud or even edge. And in between, the data that you have, it needs to be stored, it needs to be provisioned, it needs to be protected. And for things like big data, it needs to be transformed, it needs to be cleansed and so on. So the problem with this picture is that actually a lot of these solutions that you have in there are spot solutions. So I was giving this same talk in a group called in San Diego. And after this talk, a person from South West Airlines came up to me as Data Solutions Architect. He was saying, this is the exact problem that they're having in South West Airlines. So the problem here is that there are a lot of spot solutions. And the difficulty actually in this picture is not about how you transform the data or how you provide the replication for the data. The problem here is how do you connect all these different pieces of solutions together? Okay, that's the key problem. Okay, so you have all these final solutions. So the division that we are trying to do is sort of is actually to create a seamless integrative framework that connects all these different pieces together. So the data that's on your left-hand side, whether it's stored in a cloud or on a premise, can be protected, can be replicated, can be transformed seamlessly from the applications that run on the right side. So besides that, in order to do that actually, you need things like telemetry, you need things, a standardized way of collecting data for telemetry. Automation and orchestration to move the data through the different flows. Intelligence and AI to make your operations more efficient and multi-cloud to connect to your different AWS cloud, Google cloud and so on. So the vision actually is to allow you to store, run and any data anywhere. And that's what data autonomy is about. So thanks for instance Kubernetes, the CSI, I mean container storage interface. What it tries to do there is actually try to standardize the ways Kubernetes connects to storage. So the problem with this solution or with this standard is that through this standard interface, each storage vendor creates their own kind of a solution. And the solution itself is kind of like a micro ecosystem. You have like your data in vendor A that sits within vendor A and you may be able to replicate data to the cloud in AWS, but the data in the cloud can only be recognized by the same vendor solution. So similarly, I mean vendor B solution, vendor C solution, I mean each is actually still a silo. So it does not solve the problem of trying to actually remove this silo kind of a solution. I mean silo kind of a problem. What Soda tries to do is that it actually sits in between CSI and storage. So it kind of abstracts the storage from the Kubernetes, all these parts and applications. And in a way, besides unifying the orchestration which allows you to actually move data across different kinds of vendor storage, it also unifies the data that's in the cloud. So it's not about each vendor putting its data in the cloud, it's about all the data that you have of premise, no matter who's been, I mean where the data comes from, it goes into this cloud and it can be accessed by anybody. Okay, so that's the unified orchestration part that we try to do there. And besides that, the telemetry, intelligence and multi-cloud that we have in there. So Soda tries to make a mobile for Kubernetes. So this is the framework for Soda. So a lot of projects since here, I mean right now is the biggest project. In the Linux foundation, it's probably the world right now. So everybody's talking about cloud native, Kubernetes, cloud native and so on. But our problem is not just the cloud native. I mean there are a lot of enterprises out there that are still using OpenStack. They're using VMware, using bare metal kind of stuff. So Soda's goal is to be able again to enable data mobility for any kind of applications. So it's a data mobility framework. And this picture itself, what you see is that the blue parts here are actually the projects that have been developed by the OpenSS project. So the OpenSS project, the main mission initially was just to connect the different kinds of storage together to a GSA control plane so that it can connect to different kinds of storage and allows OpenStack Kubernetes and so on to use the different kinds of storage. But again in Soda itself, the goal is actually to enable data mobility. So I mean we are going to be working with creating a project for like the edge to enable data mobility between the H&D premise of the HDD Cloud. The Soda Flake is actually an in-cloud management module. And the data store project, this is something that is in preliminary kind of our discussion. So I'll talk more about it in the next slide. So the data store part, what we are trying to do is actually to create a data kind of solution for data streaming, for data analytics, for machine learning and AI. So the possibility of this came together because in the OpenSS project what we did was that we created a multi-cloud kind of data controller. And China Unicom, which is probably the second biggest telecom company in China. They have an internal project, an internal solution that they use for object storage. It leverages the self-storage backend to allow them to, through a single interface, be able to scale the backend using multiple kind of self-clusters. So China Unicom has donated the project to Soda. And what we are doing right now is to integrate this interface together with the OpenSS multi-cloud data controller project, which is the Gelato project. And with this, we are able to do something that's very useful to our end users. So what it does is that it allows through a single kind of interface be able to let you store your data either on premise or in the cloud. And you can control it through any kind of data policies. And so one of the other things that we're going to do is there's going to be a search engine. There's a company that we are talking to that has a search engine that they are open sourcing. So we're also very happy for them to be joining Soda. But I can't mention that they are the company yet. So with that, I mean it's going to make this whole Soda as a data mobility for any kind of framework. It's going to really make it come true. So the other use cases that we have are like for Yahoo Japan, they have like 52 clusters of OpenStack that they use like the Cinder, the monitor project for the file provisioning. And also they are adding more, I mean for cloud native, they are starting to deploy Kubernetes cluster. And in fact migrating some of the OpenStack clusters to Kubernetes also. So the Soda what it does is that it actually consolidates all these different kinds of different silos in 2.1. And then with this, the different OpenStack or Kubernetes clusters can actually use the kind of storage as a unified pool. And resources can be allocated and reallocated very easily. This multi-cloud use case is by entity communications. The use case is like they use AWS, I mean AWS is a partner of entity communications. And to support that, they develop a cloud controller for AWS to actually move data between their data center to the AWS cloud and AWS cloud to some remote sites. So the problem with that is that then they have the silent partner, Google, and then another partner, Microsoft. So with that, everything that every single cloud partner that came in, they have to do the same thing over and over again. And besides that, what they have to do is also like, if we add up their service, they have to redevelop things for it. So that's why I mean entity communications is actually live review Soda to try to solve this problem. Okay, so all these use cases that I talked about, they are not in production yet. Okay, there are things that we are doing for this end users, but they are being tested but not in production yet. So another use case is Toyota, the data life cycle kind of a need. So what they do is that the autonomous vehicle that they have, I mean the sensor data that they have, they want to move it to the edge. And they need a way of like moving the data between the edge to the data center for primary storage. So at the edge, they do things like high performance kind of computing. Then in the primary storage, they do things like machine learning and AI kind of stuff. And then as the data, and the data that's stored there, they store it for like half a year. And after half a year, they want to move it to like secondary storage. And the secondary storage, I mean, they're going to use it for like long term kind of machine learning and stuff. And this, they're going to store it for like, I can't remember the period but it's about two years if I remember correctly. And then after two years, the data is like too old, so all they need to do is to archive it either to the local storage or to the cloud archive. So and then the data that they have to archive and store, retain over there is actually like a period of like 15 years, which is how long they expected power to last. So this picture, I mean OpenSDS is supported by all these different companies. And as we move from OpenSDS to Soda, because the original goal of OpenSDS again was just to define a control plane for storage. But Soda has grown a lot more than what OpenSDS is doing, and that's why we are doing that, doing the transformation. And most of these companies that you see here will continue on to Soda, but some will not. So the mission of Soda is to foster an ecosystem of open source data management and storage software, again for data autonomy, create a digital platform for cross-products collaboration. Because we see a lot of, part of the problem out there is that there are actually a lot of great open source projects out there, but which is actually playing within themselves and they, for instance, like set this up, if you want to replicate some of the stuff like OpenUMD, it doesn't work at all. So we want to act as an intra-platform to actually foster that kind of collaboration. And ultimately what we want to do is actually to surf out and use this to provide ability and to have solutions as well. The goals of the source project, so it's going to be open in any way. The difference between our project and a lot of other projects out there is that our roadmap is actually driven by our end-users. I mean the requirements come from them and we prioritize the roadmap according to the requirements from the end-users. And really, this part is actually about Soda already. This is a certification program that a lot of our end-users are asking for. So in the existing world, when they want to purchase any kind of storage or data kind of solutions, they need to specify like pages of specs and requirements for the product or the solutions. So what they hope for is like in future when they want to order something, I mean when we get a purchase order, all they need to specify is that it's going to be Soda already. So Soda already is about conformance, it's about compliance, interoperability and so on. And then also we have come up with a specs for like how to certify the service providers and also to trade administrators and developers. This is no different from like any of the other end-users foundation kind of certification program. So this is something that we are hoping to do. And then the governance model, this again is similar to a lot of the end-users foundation projects. The main steering committee is the advisory board, the technical steering committee, the support of our end-users advisory committee and the outreach committee. And then the architecture workgroup that will oversees the collaboration or the design across all the different projects, the quality workgroup that will actually define that Soda-ready program and the different six across the different technologies and industries. And also the regional committees, like in Japan actually we have a really healthy community here. So we do have things for that. So the timeline for the project, I mean we kind of started Open SDS back in 2016. Our growth was not as spectacular as we hoped in like the high-region project and so on. But we do hope that this is a new start for us. So we are having a formation meeting actually tomorrow at the Sony HQ across the street. And then we're going to have Soda Forum which is being held here in room 406. So if you are free you can join us. And then the first release of the Soda, which has been developed since the Open SDS project days is titled. And actually we will have the integration between the S3 multi-cluster self-project by China Income and what Open SDS is doing. I just verified with the CTO of the China Income. This is actually a project that's in production and is currently storing like four petabytes of data. So this is a really reliable kind of opportunity. And in Q1 we'll be releasing our second release of Soda. So we have every quarter there's going to be a release. So Soda Lake is not really a fixed project name yet, but colleagues Soda Lake for now. It will start supporting streaming. And we're going to do a Soda Forum in Amsterdam, QCOM. So and then in Q2 itself the Soda already we expect a specifications draft to come together. And then Q3, the main software release will be, we'll do another forum in Shanghai, QCOM. And then the Soda Edge will come in Q3 itself. And then together with like the search engine, how to support and then the live cycle. So all of it comes together. And then finally at the end of the year we're going to do another forum, Soda Forum in India. And then also in Austin. And the release that we have, I mean the Soda Lake we expect it to be able to support AI and machine learning kind of our applications. So this is the program for the Soda Forum. You'll start tomorrow afternoon at 2.30 here. So if you're interested, you can register there. It may be a bit hard to read but basically slash Soda Forum 19. Just go there. Or just come look for me. You'll be free to get by me. So these are the different speakers. Most of these speakers actually come from companies that we participate in the project. And in the evening, I mean you can join us for the forum do. Come and join us for the evening reception. This starts from 6.5 until 9 o'clock. We have a lot of great food. Like what I was saying, we're going to have a formation meeting tomorrow morning. So this will be held in Sony headquarters. If anybody is interested in like your representing a company to join. Actually, we already have a full house of representatives from over 20 companies. If you're interested, let me know. We'll try to squeeze you in. So this is kind of like the high level kind of agenda. And the high level of this is actually at the end we get a tour, get a tour of the Sony square in the building. So the OpenSS has done a lot of activities around the world, from Barcelona to India to Tokyo. Tokyo we come and do a lot of things here, China and so on. We welcome you guys to join us. I mean to help us with this data autonomy kind of movement. So thank you very much.