SATYA NADELLA:  Good morning, everyone, and welcome.  Great to be in San Francisco this morning to talk about our cloud.

This morning, what we’re going to do is spend time talking about three things:  We will define the Microsoft Cloud.  We will talk about the uniqueness of our approach as well as the offering.  We will also go into some detail on one chapter of our cloud strategy, which is around infrastructure, and we’ll share some news and progress.  So that’s what we plan to do over the next 45 minutes.

I always like to start with our world view.  We are well and truly in this mobile-first, cloud-first world.  And when we say “mobile-first,” what we mean by that is the mobility of the individual experience.  It is not about the mobility of the device, but it’s the mobility of the experiences that you as individuals interact with in your digital life and work every day, from the sensors to the small screens to the large screens.

And that experience mobility comes about because of the cloud and its ability to orchestrate that mobility and power that mobility with intelligence, because it has the ability to reason over all of the data and bring the ubiquitous computing fabric and this power to help with those experiences light up across all of these devices.  That’s what we mean by a mobile-first, cloud-first world.

And Microsoft’s unique contribution in this world, which is the question that needs to be both asked and answered, is about productivity and platform.  That’s our core, that’s the soul of our company, and that’s what we want to bring to this mobile-first, cloud-first world.

Simply put, we want to empower every individual and every organization to be able to thrive in this mobile-first, cloud-first world.  When it comes to productivity, we think very broadly of productivity.  We think about individual productivity, we think about teams, we think about business process.

You see it in the offerings that we have.  Our intelligent agent in Cortana reasons about all of the data that you have, both at work and life, and helps you get more out of every moment of your life.

Applications like Delve and Office 365 or Power BI and Dynamics take it beyond that to your team productivity, information discovery and business process.  So that’s how we think about productivity.

In platform, the core sensibility that we have as a company is about how do we harmonize the interests across end users, developers and IT?  We believe by bringing together those three constituents is how you build platforms that create opportunities for everyone and for businesses to be able to get true advantage of technology.

Believe me, we make our share of mistakes sometimes in trying to bring together that harmony.  But that’s what we’ve done over and over again in terms of the platforms that we have delivered into the marketplace.  And that’s core to everything that we do.

That’s the context with which I want to talk about today the Microsoft Cloud.  The Microsoft Cloud is the most complete cloud offering that empowers every business across every industry in every geography.  And at its core, you have the most ubiquitous, horizontal business applications in Office 365 and Dynamics.

These are the applications that every user inside of an organization that uses these applications touches uses.  I mean, collaboration and communication in particular are central.  And that’s what Office 365 enables.

And to power a lot of that, we’ve built a hyper-scale cloud.  Azure is our hyper-scale cloud that’s also part of Microsoft Cloud.

Now, we don’t start there.  We also enable true hybrid and private cloud deployments.  So this coming together of hyper-scale cloud infrastructure that even extends to the edge by supporting private cloud and hybrid cloud deployments, and powers these ubiquitous SaaS applications in Office 365 and Dynamics is something that only Microsoft does.  That’s the uniqueness of our Microsoft Cloud offering.

Now, the best way to perhaps understand even what we have in Microsoft Cloud, what has inspired us to build this, how we’re evolving it, is to look at some of the customers using it.  The variety of use cases — in fact, between what I’m going to talk about and what Scott’s going to talk about, we’re going to talk a lot about Microsoft customers and their use of Microsoft Cloud because that’s what really drives us to evolve the cloud with the flexibility, with the openness, with the power that is needed for each one of these businesses.

When it comes to powering the variety of businesses, let’s start right here in this city.  In fact, San Francisco runs on Microsoft Cloud.  The City of San Francisco, Miguel is here, who is the CIO of the City of San Francisco.  They’ve chosen to run Office 365 to power their workforce.  In fact, one of the main considerations was what happens during any disaster recovery situation?  How can the cloud help?  And that’s been a fundamental choice for them, and the Microsoft Cloud helping in that context.

San Jose also.  And Vijay from San Jose is here.  They run on Microsoft Cloud.  Their choice was primarily driven by the mobile workforce and the need to be able to have a cloud infrastructure that can enable everyone in their workforce.

So it’s not just the largest of the enterprises where we have traditionally done well and we continue to do super well, but it’s even public sector and small businesses that are adopting the Microsoft Cloud.  So that gives us one flavor of how Microsoft Cloud is powering businesses.

A second example I wanted to talk about is NBC.  NBC has been using Microsoft Cloud in a variety of different ways for some time.  They use it for encoding their video.  They now not only do encoding, but they do live streaming.  In fact, starting with the Olympics, the Winter Olympics and then going on to the World Cup and now the Sunday football, they’re using the cloud for both encoding as well as live streaming.

But the interesting part is not just that.  It’s the fact that now they have over a thousand-plus live events that they’re going to stream.  This is not something that they did pre-cloud because now they have the ability to stand up cloud infrastructure, hydrate it, dehydrate it, with the speed and agility that it gives them, the ability to now go ahead and take these thousand events with niche audiences and really serve them with live events.  So that’s a business model innovation for them.

Another example of this is around business insights.  This is more about how one can transform an existing business.  ThyssenKrupp, which is an elevator company out of Germany, it’s actually a fascinating story.  I mean, one of the things that is going to happen is seven out of 10 people by 2050 are going to live in cities.  So if there is one thing secular about that, there are going to be more elevators needed.  And ThyssenKrupp is sort of planning ahead.

They today have over 1.1 million elevators that are directly connected back to a cloud infrastructure they’ve built on Azure.  They’re able to monitor all of these million elevators in real time, know which elevators are in stadiums, which elevators are in hospitals, which elevators are in any other place.  And their ability to reason over the data that’s coming back and offer differentiated service.  Because the business model transformation for them is to be able to do things like predictive maintenance in hospitals that are different, perhaps, than in stadiums.

So that business model change is enabled because of the insight that they’re able to get from this Internet of Things application.  In their case, it’s the Internet of “their” things as opposed to Internet of “our” things.

Another example of using that cloud infrastructure, as we today across the industry, academia and the government is all coming together to find solutions for Ebola, one of the things that tomorrow morning we’re going to do is make available Azure compute power to the research community.  In addition, we have some tools that Microsoft Research has built to be able to do vaccine discovery.

And so we want to take all of that and make it available for the research community as the industry, government and everyone else comes together to be able to tackle the challenge of Ebola.  So that’s just another example of how we believe that companies like ours can help in tackling some of the challenges that we as a society are faced with.

The last example I want to give you to just sort of bring out again the variety of businesses and how they use the cloud is Paul Smith.  Paul Smith is a British fashion designer.  They also run retail operations of their own.  They opened four or five stores in a year.  They run a very, very tight IT operation.  They have around 10 people in IT or 15 people in IT.  They have been working with us for a couple of years.  They, in fact, have a virtualized infrastructure that’s super-efficient and it’s got virtualization rates of around over 80 percent.

And you could say that’s a beautiful operation.  What do they do next?  One of the fundamental things that they have now figured out is how can they use Azure for disaster recovery?  So if you have multiple stores, you can use the control plane of Azure to be able to do disaster recovery.

Also, one of the things that every datacenter does today is deal with storage.  In other words, they’re buying storage every quarter because what is true in datacenters is just the growth of data.

But now through an appliance called StorSimple, you can plug that into your datacenter and cloud tier it.  So all of the core storage is now in the cloud and you can materialize that when needed in your datacenter.

So it’s a great example of someone who’s already running a very efficient infrastructure on premises to be able to take the cloud and use the power of the cloud to get even more efficient.

So that variety of example is what inspires us to build Azure out, build this hyper-scale public cloud infrastructure out with the flexibility that is needed.  In some sense, the customers that we’re working with across geography and across industry, across sizes is what’s helping us get grounded in the realities of what is needed in public cloud infrastructure.

Because it turns out, even though in our own case we have some very big services that run on Azure which are our own, things like Xbox Live, Bing, Office 365.  But the real goal for us is to understand how does one serve the variety of businesses?  And which it turns out you really do not want to get hijacked by just one or two application patterns, but you want to build a cloud that truly can meet the needs of every business in every vertical industry across every geography.

And so the way we’ve gone about that is on multiple layers.  So, for example, the Microsoft Cloud is ready for any device.  You can be on iOS or Android or Windows Phone and use the cloud for your application state, for your notification and as a back end for every application that you build across those devices.

We are also a cloud that supports any OS on any container technology.  So we support both Linux and Windows Server.

This is something that I want to just make sure I drill home.  Microsoft loves Linux.  Twenty percent of Azure is already Linux.  And we will have always first-class support for Linux distros.  In fact, Scott’s going to talk more about some additional container technology as well as new distros that we will get supported on Azure.  But this is something that I want to make sure that everyone recognizes, that this is not some new news, this is, in fact, today true.  And we only plan to take that to the next level going forward.

It also is the case that we support all data stacks.  We have fantastic support for SQL Server, which is a big business for us, but we also have support for Oracle on Azure, we have support for IBM on Azure, and every open source distribution of things like Cassandra, Hadoop or Mongo.  We have all of these data stacks available on top of Azure today for developers and businesses to be able to take advantage of.

When it comes to developers, so once you have support for Windows and Linux, you have support for all of the data stacks, it also means that any developer using any language on any framework can bring their code, use those frameworks on Azure.

Today, we have great support, obviously, for .NET, for Java, which is officially supported by Oracle.  We are the only public cloud that they support.  And PHP, Python and Node.js.  This is just a subset of the open source as well as other middle-tier frameworks and languages that are supported on Azure.

And the last point is perhaps the most important point, which is we are not building our hyper-scale cloud in Azure in isolation.  We are building it to compose well with other clouds.

We today, for example, in Azure AD have the ability to do single sign-on with all the other public clouds from Salesforce to Workday to even Amazon.  So you as an IT professional can use this one service on Azure to be able to manage the credentials of your employees across all of your public clouds.

That’s one of the core architectural tenets for us, which is how do we make sure that there are no limits to how Azure can be used when it comes to being able to support these other clouds.

The same is true when it comes to the true hybrid and private cloud.  And we will talk more about some of the specific improvements that we will announce today, but it’s very important for us that we support every business in every industry, many of them regulated in every geography where there will be data residency laws.  And given that complexity or the real-world complexity that one has to deal with, we want to enable these private and hybrid clouds as the edge of Azure.

So we don’t think of these as businesses that are legacy businesses.  They are, in fact, businesses that truly will take the power of what we do in Azure and bring about the full revolution of distributed computing going forward.  That’s our world view of how to build clouds.  It’s not about just building one mega-scale cloud, you absolutely build a mega-scale cloud, but it should have no limits in terms of how it works with other clouds.

That’s really the approach that has led to the success that we have.  We, today, have many customers, 80 percent of the Fortune 500 are on Microsoft Cloud.  We spend over $4.5 billion of capital in expanding our cloud each year.

We have our revenue run rate of $4.4 billion, and we’re very, very happy with the progress we’re making, the feedback cycle, because it’s still early days.  In spite of the $4.4 billion, it is still early days.  It’s the ability for us to be able to learn from the real-world complexity and use cases and evolve our cloud.  That’s the feedback cycle we’re on.

Given the momentum that we have with customers has also meant that startups and ISVs are betting on Azure.  Over 40 percent of Azure revenue today comes from third-party ISV and startup applications.  You’ll see that in the rest of our presentation as well.  We’re the best route to the enterprise customer for any startup here or anywhere else.

We want to create that opportunity.  That is something that we’ve always had at the core of how Microsoft has approached platforms, and you see that in that statistic.

And so, in closing, the Microsoft Cloud is the most-complete cloud, inclusive of the hyper-scale capabilities, hybrid and private cloud, as well as what we do in terms of delivering the most ubiquitous applications in Office 365 and Dynamics.

And what I wanted to have is Scott come up next and talk in more depth about our infrastructure and how we’re making rapid progress in taking that hyper-scale cloud and Azure and delivering the infrastructure that powers the world’s businesses and applications.  Scott?

SCOTT GUTHRIE:  Satya.  Great, well, thanks for coming here this morning.  Azure is Microsoft’s cloud computing platform and enables businesses to move faster and do more.

A little over two years ago here at an event like this in San Francisco, we actually outlined our new strategy for Azure, a strategy that enables customers to take infrastructure as a service and platform as a service and use them together.  A strategy that allows customers to use the best of the Microsoft ecosystem and the best of the Linux ecosystem together, and one where we deliver unparalleled productivity and enable you to create apps and solutions that work with any device.

Since then, we’ve been hard at work building out that vision.  Over the last 12 months, we’ve released more than 300 major new features and services.  These improvements include not just significant improvements to our infrastructure capabilities, but also higher-level services as well.  Services like machine learning, live media streaming, as well as our database as a service offerings with SQL and NoSQL.

And these higher-level services enable customers to create even richer experiences and enable them to do it with a much faster time to market than the other cloud providers.

As we build out this vision, we’re seeing some truly great momentum in the marketplace.  Satya talked about how 80 percent of the Fortune 500 companies are now on the Microsoft Cloud.

To put that concretely, we’re actually signing up more than 10,000 new Azure customers every week.  We have more than 1.2 million SQL databases now hosted on Azure.  We have more than 30 trillion objects stored in the Azure storage system.  We have more than 350 million users in our Azure Active Directory.  We’re processing more than 18 billion authentications per week.

We’ve registered more than 2 million developers with our Visual Studio Online service, which we launched less than 12 months ago.  And what’s great is these customers are not just using the core infrastructure capabilities, but we’re now seeing more than 60 percent of our Azure customers also now taking advantage of our higher-level services as well.

Obviously, the cloud market is red-hot right now.  What I want to do is just spend a few minutes walking through how do we differentiate ourselves in the market versus others?

When I talk to customers about Azure, I often talk about these three circles of capability:  hyper-scale, enterprise-grade and hybrid.

The Microsoft Cloud is the only cloud that delivers on all three, and together, they provide a truly unique cloud offering that enables differentiated solutions for customers.

What I’m going to do is just drill into each one of these and start by talking about hyper-scale.

One key element of hyper-scale is having a cloud that provides truly global reach.  And we’ve been hard at work building out Azure all over the world.  The circles here on this map here represent Azure regions, which are clusters of datacenters where customers can now deploy and run code.

Next week, we’ll officially open up our two new Australia regions to the public, and this will basically bring us up to 19 Azure regions that are open for business around the world this month.

To put this in perspective, that’s more than twice the number of regions that AWS offers today, and that’s more than six times the number of regions that the Google cloud offers today.  And this enables customers to use Azure to deploy and run their applications closer to their own customers as well as their own employees than ever before.

And what makes Azure regions impressive is not just the global coverage they provide, but also the huge-scale capacity that each one delivers.  Each of our regions is composed, typically, of multiple datacenter buildings, each one up to a football field in size.  Put that in perspective, that’s big enough that you can actually park two jumbo jets complete with all the passengers inside them.

And what’s impressive is our regions are built out so we can actually run up to 16 of these massive datacenters, co-located together in the same area.  That’s enough capacity to run up to 600,000 servers in every region.  And, again, we have 19 of those regions open now around the world.

This hyper-scale footprint delivers immense scale economics, which allows us to continually cut prices.  It also gives customers the ability to instantly scale up and down solutions with confidence anywhere around the world.

And if I put that in perspective, I want to walk you through three fairly wildly different customer use cases for how some of them are taking advantage of this hyper-scale capability.

The first one I want to talk about is a customer named Milliman.  Milliman is one of the world’s largest providers of actuarial and related products and services.  Milliman is using Azure to deliver an SaaS-based service called Integrate that provides actuarial modeling and financial reporting in the highly regulated insurance industry.

Milliman runs compute grids that are each up to 50,000 compute cores on Azure for each of their customers.  And they’re able to use this sort of immense-scale capability to help their customers stay competitive in the global life insurance market.

And one of the things I was talking with actually Milliman last week a little bit about their solution.  One of the cool things they said about it was they never, ever had to turn down a customer or turn down an immediate request from a customer in order to actually run a compute job.  And the hyper-scale cloud gives them that capability and that ability to grow their business in a massive way.

AccuWeather is another one of our customers running on Azure, and they’re one of the largest weather forecasting services in the world.  They were actually originally an AWS customer and then they switched to Azure to power all of their weather services.  They process more than 6 billion data requests every day for temperature and forecasts, and they use Azure for weather forecasting, as well as to power all of their iOS, Android and Windows device experiences.

And the last one on here is Heineken.  Heineken’s using Azure for all of its global marketing campaigns on Facebook.  And one example of a campaign, to put this in perspective, is a UEFA Champions League game that they built that was a viral game that they deployed through Facebook and supported up to millions of simultaneous users.

One of the reasons why they love the cloud is it allows them to easily scale up capacity when they need it when a campaign is popular.  And if a campaign is not popular, they don’t have any particular ones running at that point, they don’t actually have to pay anything.  And so it allows them to basically pay for everything in an on-demand fashion with no upfront costs or fixed infrastructure costs.  It allows them to grow their business even further.

Going forward, we think all customers are going to need this type of hyper-scale functionality to power their businesses.  And we think there are only going to be really three vendors in the market that are going to be able to provide this level of hyper-scale footprint:  Microsoft, Amazon and Google.

And we basically plan to differentiate from the other two by the level of enterprise-grade support we provide as well as the unique hybrid solutions that we deliver that enables enterprises to deliver integrated business.

Gartner this spring updated its Magic Quadrant surveys for cloud and credited Microsoft as being a leader in their infrastructure as a service, enterprise application platform as a service, public cloud storage services and X86 virtualization magic quadrants.

In fact, Microsoft is the only leader in all four of the cloud magic quadrants that Gartner delivers.  Amazon is, in fact, only in two of those quadrants, and I believe Google is in none of them.

Last month, we released our new D family of virtual machine sizes on Azure.  These have 60-percent-faster CPUs, support more memory configuration and provide local SSD storage.  And having this level of performance is critical for running any type — delivering an enterprise-grade level of experience.

One of the things we’re excited to announce today is our new G family of virtual machine sizes as well.  These are optimized for data workloads and use the very latest Intel Xeon processors.  They have up to 32 CPU cores of compute capability, 450 gigabytes of RAM and more than 6.5 terabytes of local SSD storage.

They provide the largest virtual machine sizes in the public cloud.  To put it in some perspective, this is our largest G series, has more than twice the amount of memory of the largest AWS virtual machine and more than four times the amount of memory of the largest Google Compute virtual machine.  We, ultimately, think this is going to allow enterprises to run even greater workloads.

Today, we’re also excited to announce our new Azure Premium Storage offering.  And this allows customers to use remote durable storage with their virtual machines.  Our new Azure Premium Storage offering will support up to 32 terabytes of storage per virtual machine.  And customers will be able to drive more than 50,000 IOPS off of each virtual machine.  To put that in perspective, that’s an immense amount of I/O activity.  It’s actually, I think, believed greater than any other storage offering from any of the other cloud providers.

And we’re going to provide it with very fast latency.  In fact, sub-millisecond latency for all read operations.  The combination of these new virtual machine sizes together with a Premium Storage offering delivers really an unparalleled degree of performance in the cloud.  Customers have never before been able to stand up this type of power literally in seconds with just a button click, and it’s going to enable even more workloads, including workloads that have not previously been virtualized and need to run on bare metal hardware to not only run in a virtualized environment, but do so in the cloud with much greater economic savings and with much more agility.

So talk about some of our hyper-scale customers.  Let me walk through a couple of our enterprise-grade customers as well and talk about why they’re betting on Azure and our cloud solutions.

GE Healthcare is one of them.  And they provide the ability for customers to collect, analyze and report healthcare data in the cloud.  They have deployed multiple applications on Azure while meeting the most stringent of regulatory requirements.

First American Financial provides closing insurance services for title companies.  And they worked with a local startup that’s based here in the Valley, which is DataStax, to basically store and allow users to search property record information in Azure.  Ultimately, they plan to actually store and process more than 18 billion title documents across the U.S.

And Takanaka is one of the largest construction companies and general contractors in Japan.  And they’re starting to deliver IOT solutions using Azure.  And they’re able to aggregate data from sensors that they install all throughout their high-rise buildings throughout Japan, and then they’re using machine learning using our Azure Machine Learning service to analyze this data to do even better property management.

Those are just a couple of the examples, I think they kind of highlight some of the use cases that having enterprise-grade capability, both on the performance side, the security side, the regulatory side and the availability side really enables.

Microsoft’s also the only hyper-scale cloud provider with software that runs not only in the public cloud, but as Satya mentioned a little bit earlier, also inside on-premises datacenter environments.  This enables customers to maximize their deployment reach and deliver the types of integrated hybrid solutions that enterprises really value.

We now at Microsoft deliver a suite of capabilities and solutions that enable customers to easily connect and integrate literally every on-premises environment to the cloud.  Customers can add backup, tiered storage, disaster recovery services to any existing infrastructure, not just Hyper-V and Windows-based systems, but also Linux systems, VMware systems, and bare metal systems as well.

They can use a common identity security solution using Azure Active Directory that allows them to span a single security architecture across both their on-premises applications as well as the devices and applications they host in the cloud.

And with our new ExpressRoute networking service, customers can basically ensure that all the traffic that flows between their on-premises datacenters and Azure run over private, dedicated network fiber, which gives them total security as well as guaranteed network quality of service.

Microsoft is also the only hyper-scale cloud provider committed to delivering a consistent cloud experience across the public cloud, on-premises datacenters, as well as service-provider environments.

This enables customers to adopt a cloud-based architecture today and enable them to deploy these solutions at any point, anywhere around the world.

We previously enabled this type of experience using our Azure Pack software solution that customers can download and run on top of their existing hardware.

Today, we’re excited to announce that we’re going even further and also delivering a full, integrated hardware plus software solution.  We call this solution our Cloud Platform System, and it’s a fully integrated hardware and software solution that enables enterprises as well as service-provider environments to benefit from all of the learnings that we’ve had in terms of running our public cloud and be able to bring that scalability and efficiency into their own datacenters.

It delivers a consistent, Azure-based management portal, consistent management set of APIs, the same as what we have in our public cloud, the same hypervisor and many of the core same features that Azure delivers as well, including all of our IaaS as well as Web-based application services.

And it basically enables enterprises to step into the cloud with even greater control and build solutions that will work not only within their own datacenters, but within ours as well.  And we’re partnering with Dell to deliver the hardware for this new solution, and it will be available starting next month.

Let’s talk about a few of the customers that are taking of this type of hybrid capability.  First I want to talk about is one that’s actually taking advantage of our cloud platform system today, and that’s General Dynamics.

General Dynamics builds and manages IT systems for some of the most security-conscious and sensitive customers in the world.  In particular, U.S. government agencies as well as defense contractors.

And the company’s looking to use our cloud platform system as a way to quickly and securely move U.S. government customers to the cloud while giving them a range of cloud scenario options.  It allows them to reuse their existing datacenter facilities or take advantage of the services that we offer in Azure, in particular with our U.S.-government-based cloud that were also opened up this year.

Lufthansa and Mazda are two other great hybrid customers.  They’re taking advantage of the cloud and Azure specifically to enable disaster recovery solutions for their existing on-premises systems.

As an example, using Azure enabled Mazda to save more than 95 percent from their data protection costs, and allows them to reuse the same set of skills, the same IT professionals as they run inside their datacenters today with our public cloud environment as well.

So hyper-scale, enterprise-grade, hybrid.  Microsoft’s the only cloud that delivers on all three.  And combined, they provide a truly unique offering that enables a set of solutions that are differentiated for customers and provides enormous opportunities to help them as well as all the ISVs and startups that serve them to grow their business.

Now, the cloud really lights up when you run great applications on top of it.  And Satya mentioned earlier, we’re working hard to enable the development of these next-generation applications on top of our cloud.

Last week, we made several important announcements about how we’re embracing Docker as a core part of our application and infrastructure set of investments.  We think container-based approaches like Docker will be great both to increase server density and enable existing applications to run even more cost-effectively on fewer amounts of infrastructure.

But we also think this type of container-based approach helps dramatically with the development of next-generation applications and enable them to be deployed even more efficiently as well.

And we’re really excited to be able to partner with Docker and the Docker community to deliver these types of experiences for Linux.  We’re also integrating these types of experiences directly into Windows Server, and then we’re going to integrate the Docker hub community directly within the Azure experience and inside our management portal.

Today, we’re continuing to announce and expand our Docker support by announcing support for the CoreOS Linux distribution.  CoreOS is a container-optimized Linux that runs with a minimal memory footprint.  In fact, CoreOS is — Satya mentioned we love Linux.  CoreOS is, in fact, the fifth Linux distribution we now offer on Azure with enterprise support.  And we’re really excited to see all the great applications that we know will be built with it.

Now, our strategy as mentioned earlier is to have a cloud that really has no limits.  Customers can use any device, any OS, any data, any programming language, and this enables not only enterprises but also startups and ISVs that want to target them to use all of the tools and technologies that they’re already familiar with, and all of them will feel totally comfortable in the Microsoft Cloud.

Having both of these segments, all three of these segments in the Microsoft Cloud also, ultimately, enables a rich ecosystem where enterprises, ISVs and startups can meet and work together to drive businesses forward.

I talked earlier about the work that DataStax is doing with First American Financial where they’re taking their Cassandra-based services and enabling them to use it to power that enterprise forward.

I talked about some of the services that GE Healthcare and Milliman are using and providing to their customers in order to help drive their health and actuarial firm businesses forward as well.

These engagements are just a few examples of the types of virtuous cycles that we think can really develop using the Microsoft Cloud, where enterprises become more successful, and the startups and ISVs can really grow their business in dramatic fashion.  And we’re really excited to see these relationships starting to form in eagerness across Azure.  And as Satya mentioned earlier, it’s one testament to that success that over 40 percent of our overall Azure revenue now comes actually from startups and ISVs.

To help accelerate these relationships even more, one of the other things we’re excited to announce today is our new Azure Marketplace.  It provides a host of VMs, apps and services that any customer will be able to easily browse, purchase and for the ISV and for the startup building them, monetize.  We think it will help accelerate the connection of enterprises to ISVs and startups, and we’re really looking forward to seeing some of the great solutions that we know are going to be built with it.

What I’d like to do is invite one of our close partners on stage, which is Cloudera, specifically Mike Olson, who is the co-founder and chief strategy officer, to talk about the new Azure Marketplace and Cloudera running in it.  Thanks, Mike.

MIKE OLSON:  Thank you, Scott, and thank all of you.

Cloudera is a big data company.  Big data represents an enormous opportunity and an enormous challenge to the enterprises that we work with.

As the leader in the big data space, specifically the leader on Apache Hadoop, Cloudera has been in business for more than six years making that platform consumable, easy to operate, easy to deploy, easy to use.

To date, most of our customers have built large infrastructure on-premises to run those systems, but there’s increasing interest in public cloud deployment, and in hybrid cloud deployment, because infrastructure running in the datacenter needs to connect to infrastructure in the public cloud.

Today, with Microsoft we’re announcing the availability of Cloudera on Azure.  This we believe is for our customers a major step forward in making the platform more consumable still.

I want to show you briefly what it is that we’ve done, and I want to talk to you about why it matters.

You can see on the screen here that we’ve got the Microsoft Azure portal.  And up at the top I’ve got the Microsoft Azure marketplace.  I can select the marketplace in order to look for services that are interesting to me.

Now, I happen to care about data, storage cache, backup services.  You’ll notice that Cloudera Enterprise is one of the options now available for deployment on Microsoft Azure.

I’ve filled in the fields on my Azure instance creation dashboard with the right names and the right identity.  I’ve entered my credentials to allow me to spin up this Azure cluster.

You can see that I’m able, if I want, to enable high availability.  I can set up my Cloudera instance on Azure so that it takes advantage of the replication built into Azure to survive failures in a very robust way.

I go down here and I click on the Create button, and I’m spinning up now an instance of Cloudera running on Azure.

If I go back to the Azure home screen, I can take a look at the various dashboards that I’ve got available.  I can look at the Azure Cloudera demo instance here and see the cluster that I’ve spun up with the different machines that have been allocated and created for me, initialized by Azure, configured and operated by the Cloudera infrastructure.

And I can look at specific machines.  You can see here I’ve got a number of machines spun up.  They’re not doing much right now, no surprise.  I’ve got plenty of capacity to go work on data.

This is a 90-core instance of Cloudera spun up in the public cloud with the single click of a button, a remarkably easy, remarkably fast way to start Cloudera in the cloud.

This is great for those IT folks who know the Azure dashboards and instances, but, you know, at Cloudera we deliver management, monitoring operations as well.  You need to be able to set and enforce security policies.  You care about data lineage, who touched it, what did they do with it, how did they transform it, for operational and for compliance purposes.

Using Cloudera’s native tooling, Cloudera Manager, Cloudera Navigator, you’re able to observe your platform, take care of it, be sure that it’s operating well, set and hit the Service Level Agreements that your business users require.

Now, everything you’ve seen so far makes the IT staff absolutely happy.  They love being able to spin up these clusters so easily.

But big data isn’t just storing, processing, transforming, analyzing.  It’s interacting with users as well.

One of the most exciting opportunities that we see in working with Microsoft is Microsoft’s suite of analytic and data processing and exploration tooling.  The most common BI tool on the planet is Microsoft Excel.  We’ve connected Excel to Cloudera Impala via ODBC so that I can extract, explore, work with all of the powerful tooling in the Excel product that I know well, in order to understand my big data.  I can run those queries and work with that data live.

We’ve set up dashboards in the Power BI suite that Microsoft offers, from data that’s managed in the Cloudera cluster, not a month’s worth, not a quarter’s worth, but 10 years’ worth of retail data available to me in detail, every single point-of-sale transaction that my retail institution has done over that time.

And I can use natural language queries to explore this data.  So if I want to know what my in-stock percentage is, I can simply ask, and you’ll see 97.8 percent.  That seems pretty good.  But, you know, I’m interested in what happened in my Little Rock store by day, and I’d really like to see — you know, I’ve got a curious falloff in in-stock availability at the tail end of the year.  That’s kind of Thanksgiving to Christmas.  Maybe there was a lot of shopping going on.  But I’ll probably want to talk to the store manager to understand what’s happening there.

Now, I’m not really a retail analyst.  I don’t know if you folks realize that.  And I don’t expect you all care very much about retail analytics beyond the fact that the powerful query tools are available here.

Let me tell you about why we’re so excited about this relationship and why we think it matters.  I mentioned before we’ve got lots of customers with big data infrastructure in their private clouds back at home.  We believe data is going to live where it’s born.  You heard Scott talk about GE Healthcare, with enormous datasets in the cloud.

If we’re going to realize our vision of unlocking the value of big data, we’ve got to run in the public cloud.  We’ve got to be where that data is.

By embracing Azure, by making sure that Cloudera runs in a fantastic, easy-to-use way on Azure, we unlock that value.  We’re able to combine our back-end tooling with the fantastic Microsoft user interfaces, exploratory tools, to deliver way more value on that big data than we ever could before.

This is a substantial strategic partnership for us.  It’s been fantastic working with the Microsoft team for the last six or eight months to make sure that we’re able to do this.

I want to thank Scott and Satya for the opportunity, and I want to thank all of you for your attention today.  Thanks very much.

(Applause.)

SATYA NADELLA:  So that gives you a great example in Cloudera’s use of Microsoft cloud and the completeness of it.  They were able to take advantage of the hybrid support, the hyper-scale support, to be able to spin up a Cloudera instance on Azure.  And perhaps most importantly they were able to not only do the big data processing but they were able to take that output from big data and connect it to users inside of an organization through integration with Office 365, Excel and Power BI.

That is what makes the Microsoft cloud the most complete cloud.  That is what gives Microsoft cloud the uniqueness.

And so we’re very, very happy to work with all of our partners to not only give them the opportunities to serve the customers that we jointly are working with but ultimately enable businesses to take advantage in every vertical, in every geography, with every size, to be able to take the advantage of power of the cloud to drive their business forward.

Thank you so much for joining today on the webcast.  So we’ll end the webcast at this point.  Thanks a lot.

END