 I'm really excited to welcome out from Walmart. I'm on deep seeing. Thank you, Jonathan. Hello, guys. Good morning. Welcome to the summit. Before I dig into the journey, the cloud journey, I'm going to tell a few things about Walmart, how we use technology at Walmart, and why it is so critical for us. Walmart, everybody knows, is the biggest retailer, if not the biggest company in the world. I think day in and out, all we do is bring everyday low prices, which are called EDLP, to our customers. And technology plays a very critical role in that. To give you some examples, back in 1980s, when internet was not kind of there yet, it was just a network of networks. We built the world's largest network of, private network of satellites, obviously, private. US government had a lot more satellites. So we connected all our stores, which were a lot of times in remote areas, to our head office, to get the latest and greatest inventory information, pricing, and whatnot, all with the sole goal of providing value for the customers. 1990s, we introduced barcodes all across the chain. Every SKU, everything had barcode. 2000, we had RFID. We built something called Retail Link, which was one of the first vendor management inventory system. It connected hundreds of thousands of suppliers to get, again, the lowest cost, lowest prices to our lowest cost, which in turn leads to lowest prices for our customers. So we've used technology. And we used cutting-edge technology all the time. We use it at scale. So we are kind of a retailer and also a technology company. So that's how I lead into our story. I'm going to break down the story into three big components. I'm going to talk about the Walmart scale, because scale is very critical here. Cloud story cannot be done without a scale. Story, the next thing is what we are doing at Walmart Labs, which is to solve the complexity of e-commerce 3.0 and then, eventually, why OpenStack. Physical scale. We have 11,000 stores going up to about 12,000 stores. We keep adding stores on a regular basis. We are in 27 countries. We get about 250 million customers per week. We have about 100 million square kilometers of retail space. That's huge. Think about it. We are probably the largest private property owner in the United States. With this much retail space, it's this much inventory lying on our stores and the customers going in and out. That's huge. And there's another stat, which is pretty interesting. Our trucks travel about a billion miles per week. That's how much inventory we move around on a regular basis. We probably are the second largest trucking company in the United States. So that's scale. And we are growing globally. We have 27 countries. And these are some of the new countries which we added in the last few years. So there's no question about physical scale. Next, on to the digital scale. We've been in e-commerce for about 10 years, and whether you know it or not. I mean, we've kind of under the wraps grown. We have about 11 websites. I mean, some of them international, some local, some club warehouses and whatnot. Last Thanksgiving, we did about 1.5 billion page views over the holiday period. That's huge. So our digital scale is also not to be ignored. And we're growing leaps and bounds. We grew 21% last fiscal quarter, and we keep growing. These are some of the properties we have. We have Asta in the UK. We have Brussels. We have Sam's Club. We have Chinese, Yehudian, and whatnot. So these are all the properties which we manage, digital properties we manage. Some of the products we have. We have mobile apps. We have a one-hour guarantee, which is a digital physical connection. We have a lot of different savings. Scatter is a new product we introduced, I think, last year. Where you go, and we give you, if you end up buying something and you end up not getting the lowest price, we refund you back. We ourselves go and match the prices with the competitors and refund you money. So these are some of the digital properties we have. And we are at a scale. And we're growing. As I said, we last quarter, we grow 20% plus percent. If you combine this complexity, if you multiply, the complexity grows by leaps and bounds. Again, when you combine the physical inventory space of 100 million square kilometers and the digital properties which we are growing, and you combine them together, that's the complexity of e-commerce 3.0. Let me dig a little bit deeper into what e-commerce is. It's any product. It's any, any, any. Any product, anywhere, any time. e-commerce 1.0 was mostly P2P sellers listing their items somewhere and buyers purchasing those items. A lot of times, sending money orders or personal checks and wait for items to eventually arrive. Sometimes they arrive. Sometimes they did not. All you had to do was leave a feedback. I didn't get my item, but that guy cashed my check or not. So that was 1.0, eight years ago. Then came 2.0, which is the e-commerce we think of today's e-commerce, where retailers built fulfillment centers. Some of them built mega-fulfillment centers. But then again, you had a bunch of fulfillment centers delivering items locally, mostly, sometimes globally. Now, 3.0 is a scale. It's many scales bigger, because now you're combining the scale, the distribution channels of physical stores which you have. For example, we have 11,000 stores. We have many more distribution centers. And we have fulfillment centers. All that inventory is now available to all the customers all the time. And the customers would want to pick those items sometimes from the store. They could use pick up today, buy an item online, go to the store, pay cash or credit card or whatever, and pick the item. Or they could have it delivered to a pickup point. Or they could have it delivered to a locker. Or they could have it delivered to your home. So all these combinations keep multiplying the complexity. Next slide is just the digital and physical. So digital is electrons moving from your desktop on to some cloud system somewhere. And eventually, at the end of the day, physical goods arrives at your doorstep, or you go pick it up. Now, this is the problem which the business people have. They want to solve this problem. They want to make sure that the customers which come to Walmart for the promise of everyday low prices can actually receive all these benefits of e-commerce 3.0. Also, and to do that, obviously, they wanted to, and we want to use technology. And I'm a technologist at Walmart, and this is the problem which was given to us. With the problem comes some needs and wants. Obviously, there are certain needs which you cannot get away from, like, for example, security. You don't want to shortchange on security. You don't want to shortchange on high availability. But then there are wants. I want on-demand capacity for infrastructure. I want DRs, so many data centers, or whatnot. So there are some needs and wants. And that's the challenge which, as a technologist at Walmart, we have to solve. And then, obviously, there are constraints, economic constraints, like Apex. You can only spend so much in Apex. You can use Apex or all those financial things which most of the technologies don't even understand. On top of that, as I said, we have been using technology for a year. Walmart is about a 50-year-old company, more than 50-years-old company. And we've built stacks and layers and layers of technologies. We've used different vendors, different stacks, different infrastructure pieces. And we've kind of grown over the years. And as a result, we've kind of entangled a lot of these services, applications, into a ball of rubber band, I guess. And I think that one of the need was to detangle that and kind of create a modular stack, something which you know where exactly what is. And if you don't know where what is, you don't know what to fix. So first thing is to detangle it, and then you start figuring it out. And obviously, what better to do than not use a cloud stack? We had to use a cloud stack. Everybody's using cloud. I mean, there are two things which are happening in the world today, cloud and mobile. So we already have mobile, so we want to do cloud. Now, what is cloud computing? So that's the next question, right? You ask anybody what is cloud computing. Everybody has a different answer. Everybody has a different vision what a cloud should be. If you talk to an application developer, they think it completely differently than if you talk to an operations guy or an infrastructure guy. And if you talk to a business guy, it's even more different, right? These are some of the definitions I picked up from internet. Gartner has its own definition. Forester has Wikipedia, which everybody realized more than anybody else, right? That's the source of truth. This is the definition which Wikipedia has. But the common theme here in all of these is at the end of the day, you want a cloud has this promise of that infrastructure will be available very rapidly. You don't have to wait weeks, if not months, for your infrastructure to arrive to you. And you want it to be flexible. A lot of times you order infrastructure, and three months later it arrives, and it's kind of outdated or it doesn't meet your needs, and you can't adapt it to your changed. So you want it to be flexible. You want the infrastructure to compute to be flexible. You want it elastic, right? A lot of times you don't know what your capacity needs are going to be upfront when you're building an application. You don't know how successful your app is going to be. And suddenly your app becomes very successful, and God knows what happens, right? It crashes, and then it's from a very successful to a very unpopular, right? It, lastly, helps you manage that. And then at the end of the day, application developers don't want to go to operations people anymore to get infrastructure needs fulfilled. They did this concept of DevOps and whatnot, so they want everything to be on demand, self-service. So there were three or four things we wanted. We wanted it to be rapid, instantaneous, if not instantaneous, flexible, elastic, and on demand, right? And to do that, we had to, you know, we did a bunch of POCs, we tried different technologies. This was like a couple of years ago while when we were venturing into solving this e-commerce 3.0 problem. And lo and behold, OpenStack. We ended up looking at OpenStack. And OpenStack, you know, it was there. It was, it met most of our requirements. It had a good promise. They were, you know, it was, I mean, we had some pains in the early VVs from Falls and Grisly to Havana, but it met our needs, which most of our vendors, you know, most of the other products which we tried did not. Some of them met certain sections. We wanted something which was extensible throughout. We didn't want which was a standard, which was not a very proprietary set of APIs which you can only use with the certain kind of vendors, with a certain kind of hardware. So OpenStack kind of met all those performance metrics and functional metrics which we set ourselves. And I mean, over the years, what we've realized, I mean, this is when we started the journey, but in the meantime, what happened, OpenStack matured a lot. In the last year, two years, it has matured significantly. It has become, you know, talk to any vendor today, it's a de facto standard. You know, all those native APIs which these other vendors had for their applications and softwares have now an API kind of a connector to it, right, OpenStack API connector to it. So that's working in our favor again. So that's helping. OpenStack is contributed by all the contributors, hundreds of thousands, thousands of contributors, 100, not 100, thousands of contributors and individual contributors and also the enterprises. I mean, we talked about all the big contributions by one company before us, 30,000 lines of code. I mean, that's helpful because there are consumers who are using it and fixing the product. Unlike certain vendors who build the product thinking this is gonna work and then shove it, you know, give it to the customers and then they, you know, and maybe adapt to customer needs, right? In this case, it's the consumer. You, yourself, we ourselves will go and adapt it to our needs. And obviously it is continued, you know, like we talked about federated keystone, that's a big problem we've had and, you know, and this is one of those things, you know, there's a feedback, there's this, you know, feedback loop from consumer to the producer and then, you know, there's a continuous improvement happening and then, you know, you get what you want, you get in, you know, if not in six months in a year, right? So it's happening. And this is another slide which tells you, you know, what our adoption was last holidays. And I know for sure we've grown a lot bigger but this is the number we have. Anybody who's ever worked in retail knows that retail is a very seasonal kind of an industry. I mean, we've worked from holiday to holiday. We prime up our workloads for holiday. November, December are the two months where we have to satisfy all the needs of all the customers. And that's what we built. We built about 140, 140,000 cores last year and this number is gonna be much bigger, not simply because we have more applications coming on, more properties coming onto our cloud and also we'll be ramping up for the holidays next year. In the end, I think it's important for us to, I mean, it's very important to, you know, figure out what our cloud strategy is. OpenStack is a big piece of our strategy but our strategy starts from, you know, supply chain optimization to having a preconfigured cabinets of hardware, you know, limited SKUs, then, you know, building our own, you know, connecting it to our own platform and having a kind of, you know, a kind of a policy structure, enforcement structure for our application developers to build, you know, applications which are cloudy by nature. But in addition to that, you know, there are other components which make ECOMOS 3.0 work. You know, there's CDN, there's computes, there's databases, how do you manage bandwidth, how do you manage, you know, all these things show back and everything combined together. So we at Walmart use OpenStack and we use a bunch of other open source technologies to make open source and proprietary technologies to make it work and, you know, it's a journey and I think we've just covered the first leg and we were on Havana last holidays and this holidays we'll be on Juneau, we'll be a little bit lagging, you know, but that's always helped us to be a little bit on the lagging side figuring out what our needs are then to, you know, change it. When you change it, you change overall. And I guess that's it. Thank you. Thank you. Thank you, Amandeep. So it's, I've said this before, but it doesn't get any more production than Walmart on Black Friday. And that's pretty cool to see that OpenStack is powering that.