 So my name is Shijjaz Abdullah. I'm a data center solution specialist, a cloud solution specialist at Microsoft based here in Qatar. My job at Microsoft is to help customers and partners build solutions on the cloud to make the cloud more useful for them and basically take advantage of the cloud and the power that it brings to achieve their business objectives. So today, I'm going to talk to you about Microsoft's cloud offerings. Basically, the Microsoft Azure platform. Azure is a cloud services platform from Microsoft, which helps customers do a variety of things, and some of these things we're going to discuss today. Basically, if you're here in this room, you probably have an idea. You have an idea to build something. You have an idea to start up something great. Most organizations interestingly today that are big in the cloud like Microsoft, Google, Amazon, they all started up as an idea. Somebody had an idea that they put to work. They began in the garage of somebody, and they ended up building these huge empires that are now cloud service providers. It's interesting to see how these providers can actually enable and fuel the growth of other startups. It's like a continuous life cycle, and it's interesting when you think of it that way. So basically, my agenda today or what I wanted to discuss with you today is to make you familiar with what Microsoft has to offer in the cloud space in general. And again, also look at why Microsoft, I mean why think of Microsoft when it comes to cloud, give you an overview of some of the offerings we have, and also some quick wins or some opportunity areas that you can apply further thought on and see how your ideas align with what we have to offer. Again, it has to remain generic when we, or general when we do these kind of presentations because everybody has a different idea, but we're gonna actually try to give you a fair amount of, fair understanding of what we have in Azure. So again, when we think of cloud computing and organizations, the traditional way of doing things, yes, we all agree that cloud is taking over and that's a new way of doing things, but again, the traditional way of doing things was to run everything in data centers. Organizations, government organizations, private organizations, startups, small organizations, big organizations, they all used to build data centers and they used to have a computer room or a closet where they used to have their servers in there, their applications in there, and they used to have a bunch of people who just manage the stuff. So everything that you see in blue over there, right from the hardware, the nuts and bolts, all the way to the application, we are actually managed and in some cases, even homegrown by the organization. Now, enter cloud computing, we have multiple different options that come up. So we have infrastructure as a service, platform as a service, and software as a service. Now, before we go ahead, I just want to take a quick survey. How many of us know something about the cloud? I mean, how many of us have heard about the cloud? How many of you are actually new to the cloud? You're just trying to understand what the cloud is and what it can bring. Okay, great. So basically, there's infrastructure as a service, platform as a service, and software as a service. When we say infrastructure as a service, we're talking about basically having a cloud service provider like Microsoft in this case, managing some amount of that infrastructure. Basically, the stuff that is hard to maintain and the nuts and bolts of the environment, the networking, the storage, the servers and stuff, that's managed by the vendor, Microsoft, and you focus on the things that actually enable your business, which is your applications and things, your customers, basically. We also have models like platform as a service where that line goes a bit higher, where even the runtime and the operating system and all that stuff is actually managed by the vendor, you just build the application and focus on that application that you need. For example, it might be a database, it might be a website, it might be something that's what you care about. The other stuff is totally taken care of by a vendor like Microsoft, and that is an SLA that this is backed with. So in the case of Microsoft, it can range between 99.9 to 99.95 or 99.9 percentage, depending upon the kind of service that you're offering. In terms of uptime SLA. And then you have the last model, which is the most convenient model to end customers, which is software as a service. Software as a service, basically, the whole stack is managed by the vendor. You just manage your users and control that access so that they can use the system. You focus on running the business. So examples of this could be Office 365, Google Mail for Enterprises, things like that. Examples of this could be something like a website hosting provider. They provide that platform for you. You manage your website, but don't worry about is my servers up and running? Is the middleware okay? Is the web server okay? I mean, the services are actually managed by us. An example of this could be something like Rackspace or even Azure, where you don't really worry about the hardware and the storage and all that stuff. Let me focus on the operating system and above. I've got a team of administrators who will actually manage that stuff, but I think below that, I don't want to be bothered about hardware and things like this. Microsoft provides solutions in all three areas. With Azure, we focus on the first two areas, and this is more like specific. If it's a collaboration and email and that kind of stuff, we have Office 365, we have CRM solutions, Customer Relationship Management solutions, we've got ERP solutions that are completely hosted in a data center and even completely maintained by us. You are just a tenant on that system and you just use it for your business. So we have all three models as Microsoft. Again, as you move towards the right, you have lesser control. Yes, it comes at an advantage that you don't have to manage things, you can focus on the business. But again, you have lesser control because you're not really controlling the entire stack. It's not in your data center, it's not at your doorstep. Again, as you move to the right, the ability to customize reduces because you're having a package solution here. It's built once for many tenants. So the ability to customize actually reduces, even though it brings a lot of benefits on efficiency, cost savings and all that stuff. Again, the cost also decreases, which is a good thing. One of the key drivers to move to the cloud is about cost. And the second thing is about efficiency. So they don't want to be in the business of managing IT. Your customers want to do their business. They're depending upon others to manage that piece for them. And that's where the benefit comes in. And plus the cost is actually much lesser. How does the cost go less? If you wanted to build your own data center, you're gonna get a quotation from hardware vendors and all that stuff, they give you something and then you build it. If some other customer comes and they need lots of servers, of course the unit pricing goes down. So think of a company like Microsoft where we're running thousands of servers, tens of thousands of servers in each data center, we actually get a work up a better deal with those vendors. And we're passing that to you so the cost actually goes down. So where does Microsoft Azure live? Where does it come from? So Microsoft Azure is powered, as I said, it's a cloud service completely managed by Microsoft. It's powered by our data centers across the world. As of today, we have 19 different regions. A region can have multiple data centers. So there are 19 different regions from which we offer services to our customers around the world. And they're shown on the map here. India is coming up very soon, but the other regions we already have established data centers. Now each of these data centers are massive. I mean, a typical data center is the size of 10 football stadiums, right? They're huge structures and having tens of thousands of servers compute. We are like the ISP in that country. We have our own internet, we're not dependent on, and we actually build those data centers. Microsoft builds and runs those data centers. It's not outsourced, it's not sent to a vendor or something. Typically, the number of people managing those massive data centers are like five in number, including the electrician, right? So they're extremely less staff, it's very efficient ways in which we manage those data centers, and they're huge. And we also do data center tours and that kind of stuff and typically takes three hours to just walk around the data centers, just to have a look at the stuff. So that's the kind of structures that's the kind of data centers that we build and we offer a variety of services from these data centers to the world. So just to put this into figures, 57% of the Fortune 500 companies in the world are using Azure. There's more than 300,000 active websites that are hosted on Azure. Just to give you some idea of what Azure is. And then you have like more than a million SQL databases running on Azure, 30 trillion things stored in the storage in Azure, 3 million requests per second on the storage that's coming in, massive numbers if you look up from the IT world. More than 300 million people who actually have accounts or Azure Active Directory accounts, 21 billion authentications happening per week. This is massive stuff. And again, this is actually fueled as there's a continuous journey here and that's where startups and organizations come here. There's a steady stream of developers, startup organizations, big organizations who are actually developing solutions on top of Azure. Azure is a platform. Azure is what you make out of it. You can actually build a solution that you can take to a customer, it's running on Azure. But what you sell to the customer or what you have for the customer is your idea and your solution. And we are just providing you that platform that helps you realize that idea. So there's more than 1.65 million developers registered and they're building solutions on top of Azure. This could be like individual developers. Somebody who has a blog and a website, he's developing stuff on Azure. It could be somebody as big as companies like Riverbed and Barocoda developing firewalls to use on Azure. So it's very interesting to see where this is headed to. Now in terms of partners, in addition to the developers and all these people, there's a huge list of partners and growing who are actually building solutions on top of Azure. Everybody wants a piece of this cloud cake. They all want to, organizations, IT organizations know that they won't be relevant if they don't have solutions that run on the cloud. If I'm completely doing on-premises stuff, building stuff solutions that actually just run on-premises, my days are numbered because everything's moving to the cloud. The cost benefit, efficiency benefit, all that is too big and too massive to lose. It's an opportunity to lose if people are not doing things on the cloud. Be it customers, be it even solution providers or partners. So almost everybody you can think of is actually developing solutions for the cloud. And the other good thing about cloud is that the way in which innovations happen. Today, if you are buying a package software solution, be it from Microsoft, be it from anybody else, it comes with a set of features. Let's say, let's use the example of Windows. You had Windows XP, it had some features and then came Windows Vista with more features so you upgraded. But what was the gap? It was a couple of years before the new features could roll out and it keeps happening. Take that with any software solution, that's the case. But when you think of a cloud solution or cloud service that's available, feature rollouts are extremely fast. We roll out, we ship features every month. So the service that you see today is not the service that you will see in a month from now because there's new feature enhancements being rolled out. It becomes much easier because it is centrally hosted. It's managed by us. We support it, we build it, we manage it. We actually are able to bring features to our customers at a much faster rate. We're trickling features down into customers rather than waiting for the next big release. So you're having a continuous improvement of things. Just in 2014, we had all those services which we just released. These are major ones. I'm not talking about small features and stuff here and there. These are new services that we launched in Azure just in the past one year. So anybody who's got an Azure subscription here knows that they keep receiving those newsletters every two weeks and there's new features that's coming in ready for you to capitalize on and use and build. So when it comes to the economics of the cloud, why do your customers care about the cloud? Why should you care about the cloud? If you look at it, I mean, of course, cloud, there is private cloud and there's public cloud. Microsoft does both as far as Microsoft is concerned. Private cloud is cloud that you build in your own data center. You have your own hardware, you put it together, you become a mini Azure sort of, you become a mini Google or a mini Azure and you build your own sort of cloud. And that's a private cloud. You completely control that cloud but it doesn't give you the economics of benefit just like the public cloud because you're building it on your own. That is a capital expenditure. But again, it is a more efficient and more automated way of doing things. On the public cloud, on the other hand, it gives you much more agility. Now if you look at it, the sweet spot for cloud, any kind of cloud is actually in the small and medium business, the startups, the small organizations. They have limited money to spend for CAPEX. You don't want to make a huge investment right from day one. You want to start small. You want to pay as you grow. You want to pay as you go. And the cloud brings that. I need X amount of compute, X amount of few services. I have just five users right now when I started. Who knows? Maybe I'll grow into a multi-billion dollar business or I might just end up, you know, that is a risk as well. So it helps you to manage that risk and it also helps you to invest as you grow. So you move from this CAPEX model where I'm going to build this data center. I'm releasing this new service to my end customers. I'm a startup and this is an IT-based solution that I have for my end customers and this thing requires massive compute power. So I'm going to build a data center in Qatar and I'm going to bring all this hardware in here and I'm going to actually, of course that comes at a capital expenditure. You pay a lot of money upfront for something that you will realize the return, the ROI you're going to realize in the next four or five years, right? And you have to invest upfront. After four or five years, the hardware reaches end of life and you reinvest, you buy hardware again. You can get into that model or you can move into an operating expenditure model where you pay a small amount of money per minute or per hour of compute that you need. And that's what the cloud brings you, right? So most benefit, the most of the benefit actually comes in the smaller you are as an organization, the better the benefit for you. That doesn't mean the enterprises don't take benefit from the cloud. There are enterprises in Qatar and around the world who are actually taking advantage of the cloud for various use cases. Now what kind of workloads does the cloud make a lot of sense to? I'm just going to open up everything so that I can explain it much easier. So when you have that idea in your mind, of something big that you want to build, be it something like Uber or be something, fantastic ideas come from startups like you. So your workloads most likely or your solution most likely comes into one of those areas. It might be an on and off thing, I mean where it works during some fixed period of time during the year. People usually tend to use this at this time of the year and that's the only time the solution works. Other times it's not needed. So or there might be components inside your solution which need to work only at certain specific times of the day or times of the month or the year for that matter. So that's an on and off kind of thing. Now why is the cloud good for this? Because if you're buying hardware you're gonna buy it and you're gonna use it for some time and you're not going to use it for the rest of the time but you still paid for it, right? It's lying idle there. And you're actually in the cloud model, you're paying per minute for the servers. The moment you shut down a server you don't pay anything for it. The other thing is growing fast. This sort of I gave you a hint of this in the previous slide where you're starting small but you're growing fast and you're expected to grow fast. You want to pay as you grow. And that model is actually a very attractive, you know, pattern for cloud. Unpredictable bursting. You don't, you can't predict. I mean, you're starting up this company, it might become really big, you know? And at times it might be dull but you don't know when this is going to happen. And you know, when you have a lot of users signing up for your service, it's brand new. People are excited. You did some marketing campaign. People are excited. They come in, they go to your website.com and they start registering and start doing transactions. That's something that happens all of a sudden. And you don't want to actually have your website break when that happens. So you want to scale up but at the same time, when that spike dies, you want to scale down so that you don't pay more for that hardware and that infrastructure, right? And the cloud helps you do that. We have automatic scaling. So basically, let's say your machine runs on, you know, your service runs on like five cores, you know, five processing units. And then, you know, you did some publicity and then suddenly like a lot of people are coming to your website. You need like 16 cores, you know, for that period of time because the people are signing up. Then we actually automatically scale up so that you have more resources, your website doesn't die. And then when we find that the demand is subsiding, we actually reduce, you know, automatically. So you pay more when you use more and you pay less when you use less instead of sticking to one, you know, one configuration. Then you have predictable bursting. How did Amazon start? Amazon is one of our competitors today but I have a lot of respect for Amazon. How did Amazon start? We all know Amazon as the company that, you know, sells books and things and stuff and we all love to shop in Amazon. They get the most, which, let's take a guess. What time of the year do you think they have the most traffic on their website? December. December, you know, the holiday season. That's when people go out shopping for gifts and stuff like that. And they realize that, you know, I have predictable bursting. Every November, December, October onwards till Jan, I have a lot of, you know, a lot of traffic. And I want, if my system breaks at that time, I'm going to lose business, right? But at the same time, I don't want to really pay for all that, you know, all that compute throughout the year, just to handle the December spike. So they get into this model where, you know, I have predictable bursting at this time. But what do I do with the remaining compute? And that's how Amazon Web Services was born. They started selling that excess compute that they had for the rest of the year to small companies, you know, because they had the infrastructure in place. So that's how the business actually evolved in a way. So coming back to your model, you actually pay when you actually burst during those times. So probably if Amazon were a customer of Microsoft, they would pay more in December than they would pay in June, right? So you get that model. And of course you have the 24 by seven steady applications. But believe me, this is the smallest, if you walk into any data center or any, you know, customer here or anywhere in the world, these kind of workloads are the least. You might think that, you know, this is the kind of workloads that are like, you know, the most common, but they're not. Most workloads fall in those categories because systems are designed that way. You don't have something that does the same thing all over the, you know, all the time. So basically when you move into a cloud model, you're actually paying for what you're using. But whereas in a CAPEX model where you buy hardware and stuff like that, you're actually paying for the whole thing and you're probably not using. So this blue area is what you're saving, you know? And again, if you do economies of scale and if you're actually calculating costs, the costs are also very different because you're doing a CAPEX and you're doing AMCs. There isn't CAPEX and an OPEX for hardware that you buy. Whereas in this case, it's flatly moving into an OPEX model. You pay for what you're using. That's the Amazon, I mean, sort of example where in December you have this huge spike coming in. And again, when you think about Microsoft, Microsoft has changed a lot from what it used to be. When people think of Microsoft, they think Windows and they think Office, you know? And everything else is nobody cares. That's not true. Satya Nadella, our new CEO, actually had recently had this, you know, very interesting news coverage where he actually had a slide that said Microsoft loves Linux, right? So in the cloud world, it's not about our platform. It's about what customers need. You might have this great idea that runs on Android or on Linux or on Apple or whatever it is. The good news with Microsoft is that it's not about Windows. More than half the workloads that run on Azure today are Linux, believe it or not, right? So we have, we actually run more Linux in our data centers than Windows. And this is what our customers choose to run, right? And we're okay with that. So for us, it's not about, you know, trying to have this, you know, conspiracy about, you know, having more Office and Windows and customers running that. No, it's about giving you something that actually reduces your time to market and gives you value. So for us, it's, we support open source just like we support our technology. They're all first class citizens on our platform. And these are some customers who are huge and open source on Azure. Now the next question that comes up into your mind and probably you're already thinking of this is now I'm putting all my data and my this new idea that I have is a system. It becomes a system and it's actually sitting in your data center. Or even my customer's data is sitting in your data center. So how can I be sure who's having access to my data? Who is looking at my stuff? You know, is the NSA having a look at it? Are they sitting in our data centers? All kinds of, you know, questions that come up in the mind, especially with the recent, you know, things that's happening. So when it comes to compliance, we comply with, we actually comply with a lot of compliance programs. In each of our data centers, it is consistent across the world, including PCI DSS, HIPAA, ISO 27001, SAS 70, there's a huge list. If you go to that website aka.ms slash azurecompliance, it takes you to the trust center where it details all the things that we comply with and our methodology and the certificates and the reports and stuff. And how can you be sure we're actually complying with this, we do regular audits of our data center by an independent third party auditor, which is Deloitte right now. And they actually make those audit reports available directly to customers from them, not from us. So if we ever fail an audit, our end customers get to know. So for us, it's a very, you know, it's transparency for us all the time. And in addition to that, we allow customers to do penetration testing of our data centers just to ensure security. There is a process for it, but yes, it is something that we allow. Not many cloud service providers want to do that. We also allow data center tours. If you're going to do this huge project on Azure, and let's say you're gonna host it in Dublin and you'd like to visit the Dublin data center, you're more than welcome to do that. And we have one of the very few cloud service providers who actually allow customers access to our data center. Yes, there is a procedure, there's background checks, there's a lot of things that we do, but it is to actually instill a peace of mind for you so that you see the things in place. And some of these policies are so stringent. For example, ISO and SAS, they require that our administrator, there are checks and controls in place so that none of our administrators can actually access data even if they wanted to. So there are things in place that prevent them from doing that. So these things are all documented, and again, they're verified in these audits that happen. So you can rest assured that the integrity of your data, the privacy of your data, the security of your intellectual property is in safe hands at Microsoft. So what are the services we provide in Azure? And now this is where your thinking cap comes on, and then this idea that you have in mind, you start thinking how Azure applies to those things. So again, there's a lot of things we offer in Azure. Some of the common things are being explained over here. So starting from the bottom, there's the storage that we have. And instead of just blob storage where you put data, you also have table storage, storage, if you think of MongoDB, you know SQL, most applications that you develop, and again open source applications, really take advantage of the stuff. It's very fast messaging and data sharing between systems. You can use that. We've got customers in Qatar also right now who are actually using it for high-performance computing. I mean, instead of investing millions to build an HPC environment, you can actually use stuff that's already there in the cloud, cues for messaging between applications. We have that stuff. Again, from a databases perspective or a data perspective, we support SQL, we support MySQL, Oracle as a first-class citizen. It's not about competitors anymore right now. So databases, machine learning is very interesting. In the slide that Ahmed showed in the beginning, I saw some very interesting trends there. I saw machine learning, which is predictive analysis. We call it machine learning. So if you're a bank and you want to actually decide, like should I give a credit card to Ahmed or not, right? There's a lot of things that go into it. That prediction, making those decisions easier is what predictive analysis and machine learning do. And also things like the internet of things. I mean, we've got customers using Azure for really interesting stuff. Like when you talk about the London, I'll give you an example of the London Underground. The London Underground, if you think of any railway or metro system, there's rails, there's trains, there's escalators, there's lifts, there's security, there's passengers, there's tickets, there's a lot of things. A lot of moving parts that go into a railway station or a metro station. So what London Underground uses Azure for is actually there are sensors connected to all these things. Temperature of the tracks, the escalators, the lifts, the elevators, all that stuff. So what they do is basically gather that information, big data, unstructured data, bring it together and make meaning out of that data. So for example, so they were actually giving me an example of the escalators, the escalators that people ride up and down. If the gap between the stairs changes beyond a certain threshold, it's a risk because a kid's finger could go inside and it can lead to injuries. So basically they monitor even that and if there is an alert, it actually notifies the nearest service technician about this issue and shuts down that system so that he can come in and he can look at it. So you can go to any extent with this. There are farmers in China who are actually having sensors attached to that cattle to know if the cattle is sick, if the cattle is pregnant and that kind of stuff. So there's fantastic ideas that come out when you think of internet of things. Then there's big data, HD Insight, that's Hadoop on Azure, caching, backup and recovery. Customers use this a lot, backing up your data. Okay, fine, I have my data center here, what do I do? I mean my backups are also in Qatar. I need to put it off site. You have that solution. We have solutions around storage. If you're building an application, which is a mobile application, and most likely, almost everybody in this room is actually thinking of a mobile app to fuel their idea. Azure offers a platform for you to develop applications for Android, iOS, Windows. We offer that platform on the cloud so that you get going quickly. You're probably thinking of a website or a web services platform for enabling this particular idea. Azure provides you that. You're looking for scale, infinite cloud scale. You're starting small and you have this unpredictable or predictable burst. Cloud services is there. You need a couple of servers. You've got VMs. You need to, if your idea is around media and media streaming and indexing and that kind of stuff, you've got media services. If you watch...