SQL Database provides predictable performance, scalability with no downtime, business continuity, and data protection. When to use : When your application requires data storage with referential integrity, transactional support, and support for TSQL queries. You can use Azure Data Factory to move existing on-premises data to Azure.

If you aren’t ready to move data to the cloud, Hybrid Connections in Azure App Service lets you connect your App Service hosted app to on-premises resources. You can also connect to Azure data and storage services from your on-premises applications. Docker containers, a form of OS virtualization, let you deploy applications in a more efficient and predictable way. A containerized application works in production the same way as on your development and test systems.

You can manage containers by using standard Docker tools. You can use your existing skills and popular open-source tools to deploy and manage container-based applications on Azure. Azure Kubernetes Service : Lets you create, configure, and manage a cluster of virtual machines that are preconfigured to run containerized applications. When to use : When you need to build production-ready, scalable environments that provide additional scheduling and management tools, or when you’re deploying a Docker Swarm cluster.

Get started : Deploy a Kubernetes Service cluster. Docker Machine : Lets you install and manage a Docker Engine on virtual hosts by using docker-machine commands. When to use : When you need to quickly prototype an app by creating a single Docker host. Custom Docker image for App Service : Lets you use Docker containers from a container registry or a customer container when you deploy a web app on Linux. It’s crucial to not only know who is using your applications, but also to prevent unauthorized access to your resources.

Azure provides several ways to authenticate your app clients. When to use : When you want to provide an SSO experience, work with Graph-based data, or authenticate domain-based users. Get started : To learn more, see the Azure Active Directory developer’s guide. App Service Authentication : When you choose App Service to host your app, you also get built-in authentication support for Azure AD, along with social identity providers—including Facebook, Google, Microsoft, and Twitter.

When to use : When you want to enable authentication in an App Service app by using Azure AD, social identity providers, or both.

To learn more about security best practices in Azure, see Azure security best practices and patterns. With your application up and running in Azure, you need to monitor performance, watch for issues, and see how customers are using your app. Azure provides several monitoring options.

Application Insights : An Azure-hosted extensible analytics service that integrates with Visual Studio to monitor your live web applications. It gives you the data that you need to improve the performance and usability of your apps continuously. This improvement occurs whether you host your applications on Azure or not.

Get started : Follow the Application Insights tutorial. Azure Monitor : A service that helps you to visualize, query, route, archive, and act on the metrics and logs that you generate with your Azure infrastructure and resources. Monitor is a single source for monitoring Azure resources and provides the data views that you see in the Azure portal.

Get started : Get started with Azure Monitor. Whether it’s provisioning VMs or publishing your web apps with continuous integration, Azure integrates with most of the popular DevOps tools.

You can work with the tools that you already have and maximize your existing experience with support for tools like:. Try it now: Try out several of the DevOps integrations. Azure is a global cloud platform that is generally available in many regions around the world. When you provision a service, application, or VM in Azure, you’re asked to select a region.

This region represents a specific datacenter where your application runs or where your data is stored. These regions correspond to specific locations, which are published on the Azure regions page.

One of the benefits of using Azure is that you can deploy your applications to various datacenters around the globe. The region that you choose can affect the performance of your application.

For example, it’s better to choose a region that’s closer to most of your customers to reduce latency in network requests. It’s always a best practice to store application data in the same datacenter or in a datacenter as near as possible to the datacenter that is hosting your application. Although unlikely, it’s not impossible for an entire datacenter to go offline because of an event such as a natural disaster or Internet failure. It’s a best practice to host vital business applications in more than one datacenter to provide maximum availability.

Using multiple regions can also reduce latency for global users and provide additional opportunities for flexibility when updating applications. Some services, such as Virtual Machine and App Services, use Azure Traffic Manager to enable multi-region support with failover between regions to support high-availability enterprise applications. For an example, see Azure reference architecture: Run a web application in multiple regions.

When to use : When you have enterprise and high-availability applications that benefit from failover and replication. Azure provides a rich set of experiences for you to create and manage your Azure resources, applications, and projects—both programmatically and in the Azure portal. Azure provides two ways to manage your applications and services from the command line.

You can use tools like Bash, Terminal, the command prompt, or your command-line tool of choice. Usually, you can do the same tasks from the command line as in the Azure portal—such as creating and configuring virtual machines, virtual networks, web apps, and other services. Azure CLI : Lets you connect to an Azure subscription and program various tasks against Azure resources from the command line.

The Azure portal is a web-based application. You can use the Azure portal to create, manage, and remove Azure resources and services. It includes:. For more information, see the Azure portal overview. Running your app on Azure likely involves working with multiple Azure services. These services follow the same life cycle and can be thought of as a logical unit.

Azure Resource Manager lets you work with the resources in your application as a group. You can deploy, update, or delete all the resources in a single, coordinated operation. Along with logically grouping and managing related resources, Azure Resource Manager includes deployment capabilities that let you customize the deployment and configuration of related resources.

For example, you can use Resource Manager deploy and configure an application. This application can consist of multiple virtual machines, a load balancer, and a database in Azure SQL Database as a single unit. Templates let you define a deployment and manage your applications by using declarative templates, rather than scripts.

Your templates can work for different environments, such as testing, staging, and production. For example, you can use templates to add a button to a GitHub repo that deploys the code in the repo to a set of Azure services with a single click. As developers, we like to dive right into the code and try to get started as fast as possible with making our applications run. We certainly want to encourage you to start working in Azure as easily as possible.

To help make it easy, Azure offers a free trial. Some services even have a “Try it for free” functionality, like Azure App Service , which doesn’t require you to even create an account. As fun as it is to dive into coding and deploying your application to Azure, it’s also important to take some time to understand how Azure works.

Specifically, you should understand how it works from a standpoint of user accounts, subscriptions, and billing. To create or work with an Azure subscription, you must have an Azure account.

If you don’t belong to such an organization, you can always create a subscription by using your Microsoft Account, which is trusted by Azure AD.

Every Azure subscription has a trust relationship with an Azure AD instance. This means that it trusts that directory to authenticate users, services, and devices. Multiple subscriptions can trust the same directory, but a subscription trusts only one directory.

As well as defining individual Azure account identities, also called users , you can define groups in Azure AD. Creating user groups is a good way to manage access to resources in a subscription by using role-based access control RBAC.

To learn how to create groups, see Create a group in Azure Active Directory preview. You can also create and manage groups by using PowerShell. A subscription is a logical grouping of Azure services that is linked to an Azure account.

A single Azure account can contain multiple subscriptions. Billing for Azure services is done on a per-subscription basis. For a list of the available subscription offers by type, see Microsoft Azure Offer Details. Azure subscriptions have an Account Administrator who has full control over the subscription. They also have a Service Administrator who has control over all services in the subscription. For information about classic subscription administrators, see Add or change Azure subscription administrators.

When you provision new Azure services, you do so in a given subscription. Individual Azure services, which are also called resources, are created in the context of a resource group.

Resource groups make it easier to deploy and manage your application’s resources. A resource group should contain all the resources for your application that you want to work with as a unit. You can move resources between resource groups and even to different subscriptions. To learn about moving resources, see Move resources to new resource group or subscription. The Azure Resource Explorer is a great tool for visualizing the resources that you’ve already created in your subscription.

To learn more, see Use Azure Resource Explorer to view and modify resources. When you allow access to Azure resources, it’s always a best practice to provide users with the least privilege that’s required to do a given task. Azure role-based access control Azure RBAC : In Azure, you can grant access to user accounts principals at a specified scope: subscription, resource group, or individual resources.

Azure RBAC lets you deploy resources into a resource group and grant permissions to a specific user or group. It also lets you limit access to only the resources that belong to the target resource group. You can also grant access to a single resource, such as a virtual machine or virtual network. To grant access, you assign a role to the user, group, or service principal. There are many predefined roles, and you can also define your own custom roles.

When to use : When you need fine-grained access management for users and groups or when you need to make a user an owner of a subscription. The next chapter contains a detailed explanation of how to use this portal to manage Azure services. The Individual components are explained with detailed pictures in subsequent chapters.

These models can be used either separately or in combination as per the requirement. SQL server database can be used for relational database. The storage module can store unrelated tables without foreign key or any relation and blobs. Blobs include binary data in the form of images, audio, video, and text files.

The process involves finding the nearest datacenter to the user who makes the request for web application, and if the nearest datacenter is not available due to various reasons, the traffic manager deviates the request to another datacenter.

However, rules are set by the owner of the application as to how a traffic manager should behave. The virtual network is another feature that is part of networking in services offered by Windows Azure. The virtual network allows a network between local machines at your premise and virtual machine in Azure Datacenter. IPs to virtual machines can be assigned in a way that makes them appear to be residing in your own premise. The following image shows how these two features actually look in Azure portal.

Azure offers HDInsight which is Hadoop-based service. Organizations often need to manage large amount of data which is necessarily not relational database management.

Hadoop is a prominent technology used these days. Thus, Azure offers Hadoop service on their platform for clients. This is achieved by executing code on many machines at the same time. Messaging Windows Azure offers two options for handling the interactions between two apps. The messages can be sent to initiate communication among different components of an application or among different applications using these two options.

CDN is used to cache the blob data that will be accessed faster by users around the world. Identity and Access This component is about management of users, authentication and authorization. It can synchronize with the related information on local machines residing on premises. Multifactor Access MFA service is built to address the security concerns such as only the right user can access the application. You can simply start using mobile development tools after logging into your account.

The push notifications can be sent, data can be stored and users can be authenticated in very less time. Similarly Azure backup can be used to backing up the on premise data in clouds. Data is stored in encrypted mode in both the cases. Media This service addresses multiple concerns related to uploading media and making it available to end users easily. Users can manage tasks related to the media like encoding, ad insertion, streaming, etc.

Commerce Windows Azure offers the opportunity to users to buy or sell applications and data through their platform. The applications are put in the marketplace or Azure store from where they can be accessed and bought by other users. Microsoft currently provides language-specific SDKs for Java,. In this chapter, you will find step by step explanation of each component: Step 1: First, login in to your Azure account.

Now you will see a list of models under Compute Model as shown in the following image. Step 2: Click Quick Create and enter the URL and choose a service plan from the dropdown list as shown in the following image.

And when you click the website URL, it will take you to the website. The following image shows how your website will look when you click the URL. This will let you choose the development framework in which you want to create your app. Windows Azure supports. There are several ways of publishing the code to Azure server. Various source control tools such as GitHub, Dropbox and Codeplex can also be used to publish the code.

It provides a very interactive interface to keep track of changes that have been published already and also unpublished changes.

Step 3: Choose the Operating System or Program you want to run. The Username and Password you set up here will be needed to access the virtual machine every time. On the next two screens you can leave the default values on for the first time. It might take a few minutes to show up. It will save a. Step 7: Open that. You can also use your own image by capturing the image of an existing virtual machine or virtual hard drive.

Virtual machines are beneficial in several ways. Creating a Mobile Service Mobile services compute hosting model is optimized to provide a cloud backend for applications that run on mobile devices.

For creating a mobile service: Step 1: Select Mobile services under Compute and click on create. A new window will be open as shown in the following image.

Step 2: Fill in the URL. Select the database, region and backend. Step 3: Tick the check box if you want to configure the advance push settings. This option allows us to configure our Mobile Service to use an existing notification hub or specify the name of a new one. If you leave this checkbox unmarked, a new hub will be created in a new namespace with a default name.

Creating Batch Service Batch service is needed when a large scale application is run and a parallel high performing computing is required. The developers can create batches to run a task parallel that eases the workload at no extra cost. They can schedule a task, put them in queues and manage the workload in cloud. Batch creation does not involve setting up a separate VM, cluster or job scheduling.

To creating a batch service follow the similar steps for creating other services under Compute model. The following image shows how a batch service can be created quickly. Once you have created a batch service, you can see the details by selecting it from the left panel. The following image pops up on the screen. When thinking of the components or services provided by Windows Azure, we wonder how all this works and what is happening in clouds. It seems very complex from our end.

Let us look into the physical architecture of these services to have a better understanding of Fabric Controller. Inside the datacenter, there are many machines or servers aggregated by a switch. We can say that fabric controller is a brain of the azure service that analyses the processes and makes decisions. The group of these machines is called cluster. Each cluster is managed and owned by a fabric controller. They are replicated along with these machines. It manages everything inside those machines, for e.

Each machine has a fabric agent running inside it and fabric controller can communicate with each fabric agent. It decides where the new application should run which is one of the most important functions of Fabric Controller. It also selects the physical server to optimize hardware utilization. When a new application is published in Azure, an application configuration file written in XML is also attached.

The fabric controller reads those files in Microsoft datacenter and makes the setting accordingly. In addition to managing the allocation of resources to a specific application, it also monitors the health of compute and storage services. It also makes the failure recoveries for a system. Imagine a situation where four instances of web role are running, and one of them dies. The fabric controller will initiate a new instance to replace the dead one immediately.

Similarly, in case any virtual machine fails, a new one is assigned by the fabric controller. It also resets the load balancers after assigning the new machine, so that it points to the new machine instantaneously. Thus, all the intelligent tasks are performed by the Fabric Controller in Windows Azure architecture.

Windows Azure allows developers to store tables, blobs, and message queues. The storage can be accessed through HTTP. In this chapter, we will learn how to create a Windows Azure Storage account and use it for storing data. A copy of the data is kept so that it is durable and available at high speed. It is retained even in case of hardware failure. There are 3 copies of each request made against the data that resides on separate domains.

The advantage is that even if there is failure on one facility, the data still can be retained. Three copies of data are created. One more advantage is that data can be read from a secondary location. The numbers of copies of data created are 6 in this case. The number of copies created is 6. The main advantage here is that availability of data can be maximized. So, choosing the replication of data depends on the cost and individual requirements. Here you can see four items under services.

You can create blobs, tables, queues and files in this storage account. There will a unique URL for each object. To access an object in the location is appended in the URL. Generating an Access Key Access key is used to authenticate the access to the storage account. Two access keys are provided in order to access the account without interrupting it, in case, one key has to be regenerated.

The following screen will come up. You will have to either create your own application or use an already built tool. Using the PowerShell commands is also an option to upload data. PowerShell is a command line application that facilitates administering and managing the Azure storage. Preset commands are used for different tasks to manage the storage.

You will find it under Command-Line tools. There are specific commands for each task. You can manage you storage account, create a new account, and create a container. Additionally, blobs, tables, queues messages can also be managed using PowerShell. Blobs Microsoft Azure Let us first understand what a Blob is. Blobs include images, text files, videos and audios.

There are three types of blobs in the service offered by Windows Azure namely block, append and page blobs. The block blobs allow the users to upload large amount of data. They allow random read and write operations. While creating a blob, if the type is not specified they are set to block type by default. All the blobs must be inside a container in your storage.

Here is how to create a container in Azure storage. Create a Container Step 1: Go to Azure portal and then in your storage account. There are three options in the Access dropdown which sets the permission of who can access the blobs. Step 2: Following command will let you access your account. You have to change the fields highlighted in all the commands. This will get you the details of you Azure account.

This will make sure that your subscription is all set. Get-AzureSubscription Step 4: Run the following command to upload your file. You can create a new container, upload blobs, see them in a listed format, and download them. Moreover, you can copy them to a secondary location in a very simple manner with this interface.

The following image makes the process clear. As can be seen, once an account is added, we can select it from the dropdown and get going. It makes operating Azure storage very easy. Queues Microsoft Azure In the common language used by developers, a queue is a data structure used to store data which follows First in-First out rule.

A data item can be inserted from back of the queue while it is retrieved from front. Azure queues are a very similar concept that is used to store the messages in a queue. A sender sends the message and a client receives and processes them. A message has few attributes attached to it, for example expiry time. A client usually processes and deletes the message. Windows Azure service lets the message to be stored for 7 days and later it gets deleted automatically, if it is not deleted by the client.

There can be one sender and one client or one sender and many clients or many sender and many clients. There are two services offered by Windows Azure for message queues. This chapter covers Windows Azure queue. Decoupling the components is one of the advantages of message queue services. It runs in an asynchronous environment where messages can be sent among the different components of an application. Thus, it provides an efficient solution for managing workflows and tasks.

For example, a message to complete a task is sent from the front- end of the application and is received by a backend worker, who then completes the task and deletes the message. Considerations The messages in the storage queue are not replicated anywhere, that means there is only one copy of your message. The maximum number of messages that can be processed are 20, The maximum size of a message can be 40 kb.

Step 2: Run the following command to access your account. Please replace the highlighted part for your account. Dequeue Next Message from Queue Step 1: First connect to your account and specify the storage account, by running the commands as shown in the above steps. Step 2: Retrieve the queue. Accounts will be displayed if you have added them during your previous use.

If not, you can add account and it will ask for your credentials. After signing in, you will be logged into your account in Azure Storage Explorer. Step 4: Add and delete the messages by selecting the queue in the left panel. Tables Microsoft Azure Storing a table does not mean relational database here. Azure Storage can store just a table without any foreign keys or any other kind of relation.

These tables are highly scalable and ideal for handling large amount of data. Tables can be stored and queried for large amount of data. Table will be created for a collection of entities. There can be custom properties and 3 system properties. Timestamp is system generated but you will have to specify the PartitionKey and RowKey while inserting data into the table. The example below will make it clearer. Table name and Property name is case sensitive which should always be considered while creating a table.

Creating a Table Step 1: Copy the following commands and paste into the screen. Replace the highlighted text with your account. Step 2: Login into your account. You can see that it has given the following end point as a result. Execute [Microsoft. TableQuery Define columns to select. Add “Author” Set query details. Properties[ ID ]. Properties[ Publisher ]. Properties[ Author ]. Primary Retrieve the table. After you are done with deleting the row, you can check the result by running the script for retrieving rows.

There you will see that the first row is deleted. While running these commands please ensure that you have replaced the accountname with your account name, accountkey with your account key. It is a free tool that you can download and install on your computer. The buttons are encircled in following image. You can see the rows by clicking on them.

Step 2: Enter Field Name. Step 4: To see the rows created click on the table name in the left panel. Azure Storage Explorer is very basic and easy interface to manage tables. You can easily create, delete, upload, and download tables using this interface. This makes the tasks very easy for developers as compared to writing lengthy scripts in Windows PowerShell. Windows Azure uses caching to increase the speed of cloud services. The process involves placing the data at strategically chosen locations and caching it.

As a result, it provides maximum bandwidth for its delivery to users. Additionally, it scales the instant high load in a very efficient manner. Step 2: Click on ‘New’ at bottom left corner. CDN endpoint is created as show in the following image. Earlier only Azure services could be linked to CDN, but now any website can be linked to it using this service. This is going to protect your data for better.

Step 6: Select the country in the next screen and you are done. Also you can specify the file type. By default, compression is disabled. Analytics: You can see very useful figures in this section. For example, number of overall hits or in a specific geographic region. The report will also show how many times requests are served from CDN endpoints and how many of them are going back to the original server. You will see a list of all the reports in the left panel as shown in the following image.

Step 2: Enter the custom URL in the text box and its done. Applications Microsoft Azure Windows Azure is usually misinterpreted as just a hosting solution, but there is a lot more that can be done using Windows Azure. It provides a platform to develop applications using a range of available technologies and programming languages. It offers to create and deploy applications using. In addition to. Windows Azure applications are scaled by creating multiple instances of the application. The number of instances needed by the application is specified by the developer while hosting the applications.

If traffic is increased or decreased on the website or web application it can be managed easily by logging in to Windows Azure management portal and specifying the instances. Load balancing can also be automated which would allow Azure to make the decision itself as when to assign more resources to application. Web applications support. Tasks such as scaling and backups can be easily automated.

Webjobs can also be scaled and scheduled. Azure platform is developed in such a way that developers need to concentrate on only the development part and need not worry about other technical stuff outside their domain. Thus most of the administrative work is done by Azure itself. A marketplace is also set by Azure where its customers can buy applications and services. It is a platform where customers can search applications and deploy them in an easier way.

Azure marketplace is available in 88 countries at present. An application purchased from the marketplace can be easily connected to the local development environment by the application developers.

The pricing is done using 5 different models, which includes usage-based and monthly fee. Some of the applications are even free of charge. Active directory is used to manage the database of users in a protected manner. The same kind of service is provided by Windows Azure to keep the users and their password safe.

Active directory is a feature that lets you create users, manage their roles, grant access and delete them. Enter a domain name which is a temporary DNS. Once its directory is created, you can map it to your own domain.

Step 1: Click on the directory name in the list of your directory. You can create a new user or link an existing Microsoft account.

You can even import a user from other directory in Azure. Step 3: Enter the user name in the following screen. Step 4: Enter other details and choose the role for the user. A pop up shown in the following image will be seen on the screen. You can enter the name of the application and follow the wizard. Basically, it will replicate your user database residing on your on-premise machine in cloud.

It will also automatically synchronize whenever changes are made on-premise. An on-premise directory can be connected using the three steps as shown in the following image. Datacenters Microsoft Azure When we think of cloud, we imagine a place with large number of machines in big rooms.

There must be a place where all the data is stored. Microsoft has datacenters all over the world from where Windows Azure services are managed. Datacenters are divided in regions. The exact location of these datacenters is not revealed by Microsoft for obvious security reasons.

Following are the 20 listed regions as can also be seen in the image. Region here specifies a regional datacenter. Performance: You should select the nearest datacenter to the users of your application. The performance can be affected by the relative location of the users who want to access the application. If a user is closer to the datacenter, the performance will be better. Cost: The price of hosting the application may also increase or decrease depending upon the datacenter you choose.

Price actually can vary according to the database hosting location or any other service being used by the application. You should choose the same location for all the services that are being used by your application. For example, database or any media service. Legal Aspect: Laws vary from country to country and restrictions could be enforced in some regions on what information can be shared and what cannot. Scenarios Microsoft Azure Understanding the basic scenarios of Windows Azure will help us understand its use.

Additionally, it will help us understand the services offered. Three basic scenarios are discussed here. In addition to the following scenarios, there can be many more ways of using Azure services based on the needs of clients, but all the basic uses are covered in this chapter.

Software Development Software development is the most popular scenario of Windows Azure. The software is developed and tested on local development fabric and then deployed in cloud of Windows Azure.

Azure hosts the web application and also the supporting processes, communicating with other web services. Testing of application in software development phase usually becomes too long for developers, if they need to change the configurations of environment being used to host the application. In Windows Azure, this is the not a problem as resources are absolutely in their control and can be modified as needed by the application.

Once a web application is hosted in cloud of Windows Azure, it is ready to be used by the end users and organizations. Moreover, deploying the application is very easy in Windows Azure using the tools provided by them.

The Visual Studio cloud project is also an easy option to deploy the application. An application is tested in the staging environment and then it is deployed in the production environment for end users to use it.

Enterprise Process Offloading There are situations for an organization where they need to reduce loads from their on- premise systems for a certain period of time or on a regular basis. This could be easily achieved by using Windows Azure services at a very low cost. Clients have to pay for only those transactions made on their application instead of paying for entire hardware and software.

This is an extremely cost-efficient way of using new resources for the organization. Azure in this context offers quite quick growth to businesses by extending resources on cloud when needed. Let us think of a scenario, when there is need for two different organizations to send and receive data between applications which is further processed by those applications. The cross-enterprise application integration can be done using Windows Azure. The service is called BizTalk service, which facilitates B2B messaging between on-premise or on-cloud applications of different organizations.

The process also includes validating and extracting the properties as required by the application at the receiving end. In a normal scenario, where communication is needed between applications of two organizations, the interaction will have to bypass the firewall by completing the due process.

Management Portal Microsoft Azure As the name suggests this is a portal to manage Azure services, which was released in This is a platform provided by Microsoft for its Azure clients where they can see, manage and buy the services offered by Azure. However, features are more or less same in both the portals. You will get a free trial and you can explore, learn and create your own applications using Windows Azure. Since here we have an application already running, you can see a list of them.

Your account will be empty for the first time. Left panel categorizes the application and the middle part lists all the application in the account. It will take you to the following screen.

This screen will show you all the details of your subscription, spending, and data usage. This way you can set a spending limit for you. Your services will be stopped once you reach the spending limit. If you scroll down on the page in the above image, you can see all that is available with your subscription and see the details on the right side. This is calculated by your average per day spending and it would tell you in how many days your credit is going to get over. Add a New Subscription Step 1: Click on your account e-mail id or on the picture at the top right corner.

Azure Preview Portal Step 1: Click on your account e-mail at the top right corner. Step 3: The following screen will appear. All the functionalities are same. Create Virtual Network Microsoft Azure You can create virtual network on cloud or you can also connect to the on-premise local network to the cloud network in Windows Azure. This tutorial will first explain how to create a cloud only network. Step 5: Enter the name and leave all other fields as they are except location. You will see that it will draw an image at the bottom.

The subsequent chapters will have a demo on configuration of these two options. Step 3: Click next and leave the default values on the following screen. Step 4: Click the next arrow and a virtual network is created.

You can add DNS servers and local network even after creating a virtual network. This chapter contains the detailed process including how to configure virtual machines.

Step 4: Enter DNS name. This has to be unique. The DNS name is used to connect to the virtual machine. Step 5: Select the image and size from the dropdown list. The size affects the cost of running virtual machine. Step 6: Enter username and password. You must remember to log in to the virtual machine later.

Step 7: Select the relevant region. It will take a few seconds for the machine to be created. In this screen, you find that choosing an image is easier based on their category shown on the left side. Let us create a virtual machine for SQL Server for which we have chosen SQL Server on the left side and all the software in this category are shown in the middle.

Step 3: Click on the Next arrow. The size dropdown would change items according to tier. In the basic version, you will get only first 5 options, while in the standard version you will get more options. It requires minimum A4 machine with 8 cores and 14GB memory. Step 6: Enter the username and password and click Next arrow. Step 7: Enter DNS name which should be unique as mentioned earlier and select the region.

Under the storage account, it will display the storage accounts that you have already created. As seen in the following screen, an account name is shown in the dropdown which is a storage account created earlier. You can choose an already created account or even use an automatically generated account. The last option is End Points. End points are used to communicate with virtual machines by other resources you can leave.

In a subsequent chapter, we will provide a detailed illustration to configure endpoints. Step 9: Click on Next and the virtual machine will be created in a few seconds for you. If you already have a virtual network created in Azure, it will be diplayed in the highlighted dropdown list as shown in the following screen.

You can choose the network as shown in following picture. The virtual machine will be displyed in the resources of that network as shown in the following picture. Please refer to it. It is best to choose the region nearest to the physical location of end users. Endpoint Configuration Microsoft Azure When creating a virtual machine, we come across a part where endpoints can be configured. The two default endpoints enabled while creating a virtual machine are Remote Desktop and PowerShell.

What actually is an endpoint? Virtual machine on same cloud can communicate to each other automatically. But in case we need them to communicate with our own computer, we will need an endpoint configured to make it happen. It is basically accessing the virtual machine through a port. An endpoint provides remote access to the services running on virtual machine. It has a public and private port that needs to be specified while creating an endpoint.

However, it can also be done in the same way as creating a new one on configuration part of wizard. Alternatively, you can enter a custom name. It will assign unused ports automatically. Or you can enter it manually. Step 7: Click on Next arrow. Access Control of Endpoint We can grant or deny the access of services to an individual host or network. If nothing is specified, the endpoint can be accessed from any host and network. Step 3: Enter Subnet Mask. Point-to-Site Connectivity Microsoft Azure In the last chapter, we saw how an endpoint can be created to access a virtual machine; this is quite a tedious task.

If a virtual machine in virtual network needs to be connected with on-premise machine, the point-to-site connectivity is needed. Point-to-site connectivity makes it very productive to work with remote virtual machines. Basically, a machine on-premise is connected to virtual network using point-to-site connectivity.

However, we can connect up to on-premise machines to virtual network in Azure. The access to the virtual network in cloud is granted through a certificate.

The certificate has to be installed on each local machine that needs to be connected to the virtual network. Enabling Point-to-Site Connectivity on Existing Virtual Network If you have already created a virtual network in Azure, you can access it in management portal.

Step 1: Log in to Azure management portal. Message shown in the following screen will pop up. Step 7: Click Yes and a point-to-site connectivity is done. You will need a certificate to access your virtual network. Step 6: Point-to-Site connectivity is done. You will see that the gateway is not created yet.

For it to happen, you will have to generate a certificate first. Then go to msdn link or the version of Windows for which you want the tool. Step 2: Download the encircled file as shown in the following image.

It will be saved as. While running the installation wizard, when you reach the following screen uncheck the encircled part. By default they are checked. In the above command change the highlighted part to the name of your network. Step 6: Next enter the following command for creating client certificate.

You can see the certificates here. Step Follow the wizard. You will have to name the certificate and select a location to save it. Upload the Certificate Step 1: Login to Azure management portal. Step 3: Click browse and select the location of the certificate you just created. Step 2: Scroll down and locate the following options at the right side of the screen.

Step 3: Select the suitable option and download it. You will see a similar file on your computer. Run and install it.

You will be connected to the network. Site-to-Site Connectivity Microsoft Azure Most organizations already have a network on their premises and would want to connect it to Windows Azure rather than putting everything on cloud.

It is also called hybrid network connectivity. It is connecting virtual net in Azure to on-premises network.

Setting up a site-to-site connectivity network is quite easy for someone who knows the basics of networking like IPs, subnetting and default gateways.

Step 3: Enter the DNS name for name resolution if you want, otherwise you can leave it empty if you want it to be automatically done by Azure.

Step 6: Enter the details of your virtual network in the address space. Step 7: After entering the subnets, enter the gateway subnet for your virtual network. You will have to create a gateway for it. You can configure the VPN device now using the information. Site-to-site connectivity is faster than the point-to-site connectivity.

It makes transferring of data easier. You just need a shared key to access the network. In fact, the same shared key works for each machine. Traffic Manager Microsoft Azure Let us first understand what is the service provided by Azure traffic manager. Basically, this service balances the traffic load of services hosted in Azure. The routing policy is defined by the client and traffic to the services hosted in Azure is redirected according to set policies.

Traffic manager is a DNS-based service. Thus, it will improve the availability and performance applications.

There are three options in this dropdown. When a DNS is requested, it is redirected to the region closest to the user. Traffic is distributed in round robin fashion by selecting a healthy endpoint. Step 4: Based on your needs you can choose a load balancing method. Step 5: Click create. You will see the traffic manager created and displayed in your management portal. Its status will be inactive until it is configured.

Choose the service type and items under that service will be listed. Step 4: Select the service endpoints and proceed. Step 5: The service endpoints will be provisioned. For example, if you enter 40 seconds the traffic manager will be queried after every 40 seconds for the changes in the traffic management system. You can choose the protocol; enter port number and relative path for a service to be monitored.

PowerShell Microsoft Azure PowerShell is a framework or you can say an interface built by Azure team that lets the user to automate and manage Windows Azure services.

It is a command line tool that uses the scripts or cmdlets to perform tasks such as creating and managing storage accounts or Virtual Machines that can easily be done using the preset commands.

You can run it as ISE by pinning it to the taskbar in Windows 8. ISE lets copy paste commands easily. You just have to perform this step once on your computer and every time you run Azure PowerShell, it will connect to the account automatically. Step 1: Enter the following cmdlet in PowerShell. Add-AzureAccount Step 2: The screen shown in the following image will pop up and ask for credentials of your account.

Enter the credentials and sign in. Using Certificate In this method, you can download a certificate on your machine and login to our account using that certificate. You will be prompted to save a file and the file will be downloaded on your computer with the extension. Highlighted part is the path of the file downloaded in previous step. Run the following cmdlet. It will display the details of your account and subscription. Remove Azure Account Run the following cmdlets.

Replace the highlighted part with your account ID. It will ask for your confirmation and it is done. Get-Help Azure There are lots of tasks that can be managed using PowerShell such as creating and managing web applications, storage accounts, virtual machines, etc. In fact, many users find it quicker and better as compared to Azure Management Portal. Monitoring Virtual Machines Microsoft Azure Monitoring virtual machines is important to keep a track of its performance and health.

Windows Azure provides an interactive interface to monitor the statistics related to the performance of virtual machine. Step 2: Go to Virtual Machine. Step 3: Select the virtual machine you want to monitor. Step 4: Select Monitor from the top menu as shown in following image. You can see 5 key terms on the above screen. It seems pretty complicated at first glance but when you look carefully, you can see that each line on the graph is in a different color which matches the color of the term.

For example, CPU Percentage is in purple color and the purple line on the graph represents it. The machine shown in the above image is quite new. The following sections will explain how figures are read. In the following image, you can see that in the last dropdown at the right top corner 1 hour is selected and, highest utilization is at which is 0. If the read operations are done more frequently on the disk, performance issues can be resolved using a faster disk.

If the application needs writing large amount of data on the disk, a bigger disk can be chosen. You can also see data for last 24 hours and 7 days. The network-in statistics can be in bytes or TCP segments received. You can also see relative or absolute statistics by selecting an option from dropdown encircled in the following image. The following screen will appear, in which you can check the desired metrics. Enable Diagnostics Enabling diagnostics allows you to collect logs.

Azure will collect logs and store in a storage account you specified. Diagnostic figures help in troubleshooting as the logs for errors can be tracked in the storage account. By default it will be set off.

Step 6: Since it uses a storage account to store the logs you will have to configure the setting for the storage account by clicking on the encircled part in the following image. This section displays the same metrics that we discussed in the section above. These figures help users to identify the causes of performance slide of an application. They can also generate alerts for these features; they go above the set limits.

Setting Up Alert Rules Microsoft Azure While monitoring a virtual machine we can see different metrics related to a virtual machine in Azure. Azure has also provisioned a way to alert the administrator of virtual machine when these metrics go above or below a specified limit through e-mail. Setting up an alert can be very useful in notifying the administrator about issues that require attention.

 
 

Programming windows azure pdf download

 

Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support. Feedback will be sent to Microsoft: By pressing the submit button, your feedback will be used to improve Microsoft products and services.

Privacy policy. Azure is a complete cloud platform that can host your existing applications and streamline new application development. Azure can even enhance on-premises applications. Azure integrates the cloud services that you need to develop, test, deploy, and manage your applications, all while taking advantage of the efficiencies of cloud computing. By hosting your applications in Azure, you can start small and programming windows azure pdf download scale your application as your customer demand grows.

Azure also offers the reliability that’s needed for high-availability applications, azjre including failover between different regions. The Azure portal lets you easily manage all your Azure services. You can also manage your services programmatically by using service-specific APIs and templates. This guide is an introduction to the Azure hp laserjet p2050 series pcl6 download for application developers. It provides guidance and programming windows azure pdf download that you need to start building new applications in Azure or migrating existing applications to Azure.

With all the progarmming that Azure offers, it can be an intimidating task to figure progarmming which services you need to support your solution architecture. This section highlights the Azure services that developers commonly use. For a list of all Azure services, see the Azure documentation. First, you must decide on источник to prigramming your application winvows Azure.

Do you need to manage your entire infrastructure as a virtual machine VM? Can you use the platform management facilities that Azure provides?

Maybe you need a serverless framework to host code execution only? Your application needs cloud storage, which Azure provides several options for. You can take advantage dowjload Programming windows azure pdf download enterprise authentication.

There are also tools for cloud-based development and monitoring, and most hosting services offer DevOps integration. Now, let’s look at some of the specific services that адрес recommend investigating for programming windows azure pdf download applications. Azure provides several cloud-based compute offerings to run your wundows so programming windows azure pdf download you don’t have to worry about the infrastructure details.

You can easily scale up or scale out your resources as your application usage grows. Azure offers services that support your application programming windows azure pdf download and hosting needs. Azure provides Infrastructure as a Service IaaS to give you full control over your application hosting. Azure’s Platform as a Service PaaS offerings provide the fully managed services needed to power your apps.

There’s even true serverless hosting in Azure where all you need to do is write your code. When you want the quickest path to publish your web-based projects, consider Azure App Service.

This platform provides authentication by using social providers, traffic-based autoscaling, testing in production, and continuous and container-based deployments. Because all three app types share the App Service runtime, you can host a website, programming windows azure pdf download mobile clients, and expose your APIs in Azure, all from the same project or solution. App Service has been designed with DevOps in mind. Prgramming supports various tools for publishing and continuous integration deployments.

You can migrate your existing applications to App Service by using the online migration tool. When to use : Use App Service when you’re migrating dowlnoad web applications to Azure, and when you need a fully-managed hosting platform for your web apps.

Get started : App Service makes it easy to create and deploy your first web appmobile appor API app. Try it now : App Windowe lets you programming windows azure pdf download a short-lived app to try the platform without having to sign up for an Azure account. Try the platform and create your Azure App Service app.

With VMs, you have total control over the configuration of the machine. When download test drive 1 pc VMs, you’re responsible for all server software installation, configuration, maintenance, and operating system patches. Because of the level of control that you have with VMs, you can run a wide range of server workloads on Azure that don’t fit into a PaaS model.

For more information, see the Virtual Machines documentation for either Linux or Windows. When to use : Use Virtual Machines when you want full control over your application infrastructure or to migrate on-premises application workloads to Azure without having to make changes. Rather than worrying about building out знакома cara download aplikasi dari play store melalui pc сообщение managing a whole application or the infrastructure to run your code, what if you manhunt 3 pc game download just write your code and have it run in response to events or on a schedule?

Programming windows azure pdf download Functions is a “serverless”-style offering that lets you write just the code you need.

With Functions, you can trigger code execution with HTTP requests, webhooks, cloud service events, or on a schedule. You can code in your development language of choice, such download action adventure for CFNode. With consumption-based billing, you programming windows azure pdf download only for the time that your code executes, and Azure scales as needed. When to use : Use Azure Functions programming windows azure pdf download you have code prpgramming is triggered by other Azure services, by web-based events, or on a schedule.

You can also use Functions when you don’t need the overhead of a complete hosted project or when you only want to pay for the time that your code runs. To learn more, see Azure Functions Overview. Get started : Follow the Functions quickstart tutorial to create your first function from the portal. Try it now : Azure Functions lets you run your code without having to sign up for an Azure account. Try it now at progra,ming create your first Azure Function. Azure Service Fabric is a distributed systems platform.

This platform makes it easy to build, package, deploy, window manage scalable and reliable microservices. It zaure provides comprehensive application management capabilities such as:.

Apps, which run on a shared pool of machines, can start small and prgramming to hundreds or thousands of machines as needed. NET Core. It provides SDKs for building services on Linux in both. NET Core and Java. To learn more about Service Fabric, see the Service Fabric documentation. When to use: Service Fabric is a good choice when you’re creating an application or rewriting an existing application to use a microservice architecture.

Use Service Fabric when you need more control over, or aure access to, the underlying infrastructure. Get started: Proframming your first Azure Service Fabric application. Azure Spring Cloud is a serverless microservices platform that enables you to build, deploy, scale and monitor your applications in the cloud. Use Spring Cloud to bring modern microservice patterns to Spring Boot apps, eliminating boilerplate code to quickly build robust Java apps.

Along with application hosting, Azure provides service offerings that prlgramming enhance the functionality. Azure can also improve the development downloaf maintenance of your applications, both in the downlozd and on-premises. Most applications must store data, so however you decide to host prograamming application in Azure, consider one or more of the following storage and data services.

Azure Cosmos DB : A globally distributed, multi-model database service. This database enables programming windows azure pdf download to elastically scale throughput and storage across any number перейти geographical regions with a comprehensive SLA. When to use: When your application needs document, table, or graph databases, including MongoDB databases, azre multiple well-defined dowwnload models. Azure Storage : Offers durable, highly available storage for blobs, queues, files, and other kinds of nonrelational data.

Downnload provides the storage foundation for VMs. When to use : When your app stores nonrelational data, such as key-value pairs tablesblobs, files shares, or messages queues. Get started : Choose sleeping crack for pc one of these types of storage: blobstablesqueuesor files. SQL Database provides predictable performance, scalability with no downtime, programming windows azure pdf download continuity, and data protection. When to use : When your iwndows requires data storage with referential programming windows azure pdf download, transactional programming windows azure pdf download, and support for TSQL queries.

You can use Azure Data Factory to move existing on-premises data to Azure. If you aren’t ready to programming windows azure pdf download data to the cloud, Szure Connections in Azure App Service lets you connect your App Azue hosted app to on-premises resources.

You can also connect to Azure data and storage dowload from your on-premises applications. Docker containers, a form of OS virtualization, let you deploy applications in a more efficient and predictable way. A programming windows azure pdf download application works in production the same way as on your development and test systems. You can manage containers by using standard Docker tools. You can use your existing skills and popular open-source tools to deploy and manage container-based applications on Azure.

Azure Kubernetes Service : Lets you create, configure, and manage a cluster of virtual machines that are preconfigured to run containerized applications. When to use : When you need to build production-ready, scalable environments that provide additional scheduling and management tools, or vownload you’re deploying a Docker Swarm cluster.

Windods started : Deploy a Kubernetes Service cluster. Docker Machine : Lets you install and manage a Docker Engine on virtual hosts by using docker-machine commands.

When to use : When you need to quickly prototype an app by downlkad a single Docker host. Custom Azute image for App Service : Lets you use Docker containers from a container registry or a customer container when you deploy a web programjing on Linux. It’s crucial to not only know who is using your applications, but also to prevent unauthorized access to your resources.

Azure provides several ways to authenticate your app clients. When to use : When you want to provide an SSO experience, work with Graph-based data, or authenticate domain-based users. Get started : To learn more, see the Azure Active Directory developer’s guide.

App Service Authentication : When you choose App Service to host your app, you also get built-in authentication support for Azure AD, along with social identity providers—including Facebook, Google, Microsoft, and Twitter.

When to use : When you want to enable authentication in an App Service app by programming windows azure pdf download Azure AD, social identity providers, or both.

 

[Programming Windows Azure Sriram Krishnan – PDF Free Download

 

By using our site, you agree to our collection of information through the use of cookies. To learn more, view our Privacy Policy.

To browse Academia. Log in with Facebook Log in with Google. Remember me on this computer. Enter the email address you signed up with and we’ll email you a reset link. Need an account? Click here to sign up. Download Free PDF. Microsoft azure tutorial. Alejandro Linares. A short summary of this paper. Download Download PDF. Translate PDF. Microsoft Azure About the Tutorial Windows Azure, which was later renamed as Microsoft Azure inis a cloud computing platform, designed by Microsoft to successfully build, deploy, and manage applications and services through a global network of datacenters.

This tutorial explains various features of нажмите чтобы узнать больше flexible platform and provides a step-by-step description of how to use the same. Audience This tutorial has been designed for software developers who are keen on developing best- in-class applications using this open and advanced platform of Windows Azure. Prerequisites To learn Windows Azure, you need to be familiar with the Windows environment and have a basic knowledge of cloud computing.

All the content and graphics published in this e-book are the property programming windows azure pdf download Tutorials Point I Pvt.

The user of this e-book is prohibited to reuse, retain, copy, distribute or republish any contents or a part of contents of this e-book in any manner without written consent of the publisher. We strive to update the contents of our website and tutorials as timely and as precisely as possible, however, the contents may contain inaccuracies or errors.

Tutorials Point Http://replace.me/19694.txt Pvt. If you discover programming windows azure pdf download errors on our website or in this tutorial, please notify us at contact tutorialspoint.

Cloud Computing — Overview Windows Azure Azure — Components Compute Module Fabric Controller Management Portal Create Virtual Network Deploying Virtual Machines Endpoint Configuration Point-to-Site Connectivity Site-to-Site Connectivity Traffic Manager Monitoring Virtual Machines Setting Up Alert Rules Application Deployment Self-Service Capabilities Multi-Factor Authentication Forefront Identity Manager Data Import and Export Job Disk Configuration Disk Caching Personalize Azure Access Personalize Company Branding Self-Service Password Programming windows azure pdf download Self-Service Group Management Create a Group Security Reports and Alerts Orchestrated Recovery Health Monitoring Cloud computing can be referred to as the storing and accessing of data over the internet rather than your computer’s hard drive.

This means you don’t access the data from either your programming windows azure pdf download hard новость! gta vice city cheat book free download for pc правы or over a dedicated computer network home or office network. Cloud computing means data is stored at a remote place and is synchronized with other web information. One prominent example of cloud computing is Office which allows users to store, access, edit their MS Office documents online in browser without installing the actual program on their device.

Back-end Platform: There are various computers, servers, virtual machines, etc. These clouds are accessed through internet programming windows azure pdf download users. These are open to public and their infrastructure is owned and operated by service страница as in case of Google and Microsoft.

Private Cloud: These clouds are dedicated to a particular organization. That particular organization can use the cloud for storing the propresenter 5 free download for windows 10 data, hosting business application, etc. The data stored on public cloud can’t be shared with other organizations. The cloud is managed either by the organization itself or by the third programming windows azure pdf download. Organizations can use private clouds for sensitive application, while public clouds for non-sensitive applications.

The hybrid clouds provide flexible, download updates windows 2003 and cost-effective solutions to the programming windows azure pdf download. Benefits of Cloud There are many benefits of clouds. Programming windows azure pdf download of them are listed below.

Allocation and de-allocation of resources по этой ссылке dynamically as per demand. SPI Next comes how cloud services are categorized. Windows Azure Microsoft Azure There are many cloud computing platforms offered by different organizations. Windows Azure is one of them, which is provided by Microsoft. Azure can be described as the managed data centers that are used to build, deploy, manage the applications and provide services through a global network.

Many programming languages and frameworks are supported by it. Azure as PaaS Platform as a Service As the name suggests, a platform is provided to clients to develop and deploy software. The clients can focus on the application development rather than having to worry about hardware and infrastructure. It also takes care of most of the operating programming windows azure pdf download, servers and networking issues. The whole process is not visible to developer and thus does not pose a risk of data breach.

There can be a different environment at Azure, thus the application might have to be adapted accordingly. Azure as IaaS Infrastructure as a Service It is a managed compute service that gives complete control of the operating systems and the application platform stack to the application developers. It lets the user to access, manage and monitor the data centers by themselves. The virtual machine can be completely adapted to the requirements of the organization or business.

This means application can be migrated to Windows Azure without rework. All the application dependencies such as database can also be migrated to Azure. This also helps the vendors to expand their business by selling the existing software or services in new markets. It might become difficult for them to migrate the application to future versions.

For example, higher server maintenance for patching and upgrading software. Some companies have well- defined processes for testing and updating on-premise servers for security vulnerabilities.

These processes need to be extended to the cloud-hosted IaaS VMs to mitigate hacking risks. Unlike PaaS, there is no provision of automatic server patching in IaaS. An unpatched server with sensitive information can be very vulnerable affecting the entire business of an organization. It can be stuck with the older version of the operating systems and application stacks. Thus, resulting in applications that are difficult to maintain and add new functionality over the period of time.

It becomes necessary to understand the pros and cons programming windows azure pdf download both services in order to choose the right one according your requirements. In conclusion it can be said that, PaaS has definite economic advantages for operations over IaaS for commodity applications. In PaaS, the cost of operations breaks the business model. Whereas, IaaS gives complete control of the OS and application platform stack.

Azure Management Portal Azure Management Portal is an interface to manage the services and infrastructure launched in All the services and по этому сообщению are displayed in it and it lets the user manage them. Getting started A programming windows azure pdf download trial account can be created on Azure management portal by visiting the following link – manage.

 
 

Programming windows azure pdf download

 
 

How customer reviews and ratings work Customer Reviews, including Product Star Ratings help customers to learn more about the product and decide whether it is the right product for them. Learn more how customers reviews work on Amazon. Top reviews Most recent Top reviews. Top reviews from the United States. There was a problem filtering reviews right now. Please try again later. The author did a great job of introducing the Windows Azure platform and offer enough useful code samples.

The first couple of chapters are introductions. Quickly, the chapters started with code samples that developers can use. What I liked the most is the code samples about interacting with Azure Tables and Blobs. For example, adding entities and download files chunks to MemoryStream. The author is evidently an experienced writer. He has done a great job in terms of topic selection, code samples, and presentation.

If you are new to Windows Azure and would like to start working on it, this book can be a useful reference. This book does not offer a comprehensive example of Azure application from end to end. If you are experienced programmer or if you have someone on your team is, then you should be able to create your Azure application with some information covered in this book and supplemental information with your own research.

Overall, I am pleased with this book’s depth and coverage. I wanted to have a basic knowledge of Azure’s architecture, development and cloud operating systems called “fabrics” , and programming techniques with Visual Studio so I could better understand the new features and enhancements planned for the November Community Technical Preview CTP. I wasn’t disappointed. The book’s Part I starts with cloud-computing definitions and answers the most common question, “What is cloud computing?

Roger explains how Windows Azure’s data center architecture groups computing and storage hardware into Upgrade and Failure domains so that operating system modifications can be applied and failed hardware substituted without hosted application downtime.

Chapter 3 covers Azure’s operating system and hypervisor that’s based on a modified version of Windows Server , as well as the lifecycle of an Azure application and how Azure implements multi-tenancy, a requirement for economical cloud computing.

I found this background helped me understand Chapter 4’s description of how and why Windows Azure’s basic data store uses highly scalable, RESTful entity-attribute-value EAV tables rather than relational tables, and containers of individually addressable binary large objects blobs instead of a conventional, hierarchical file system. NET Worker Roles to manipulate tables and blobs as. This requires a reference to the Windows Azure SDK’s StorageClient library, which enables using the on-premises framework provided by the Development Fabric and Storage features, as well as the production infrastructure.

Creating accounts with the Windows Azure Developer Portal for testing cloud storage with local Worker Roles and deploying the finished project to a production data center followed. The Developer Portal has changed since Azure moved from free CTP to paid status on January 4, , but Roger’s December and January blog posts announced in the publisher’s online forum for the book describe the differences in detail.

Security is the chief concern of CIOs and CISOs when considering cloud data storage and compute operations for confidential information. NET System. Cryptography namespace and AES for storage encryption. Chapter 6 takes you through the process of implementing ASP. NET authentication and role management for Azure Worker roles. Chapter 7 explains how to make Azure tables more scalable, use entity group transactions and display table data in grids.

Chapter 8 introduces you to Worker Roles and Azure queues. Book description Learn the nuts and bolts of cloud computing with Windows Azure, Microsoft’s new Internet services platform. Learn how to build applications using the Windows Azure toolset Discover how Windows Azure works under the hood, and learn the how and the why behind several features Choose to write application code in. Show and hide more. Table of contents Product information.

Native and Non-. ISBN: You might also like book 40 Algorithms Every Programmer Should Know by Imran Ahmad Learn algorithms for solving classic computer science problems with this concise guide covering everything from fundamental … book Data Science from Scratch, 2nd Edition by Joel Grus To really learn data science, you should not only master the tools—data science libraries, frameworks, modules, … book Implementing Azure DevOps Solutions by Henry Been, Maik van der Gaag A comprehensive guide to becoming a skilled Azure DevOps engineer Key Features Explore a step-by-step approach … book Designing Data-Intensive Applications by Martin Kleppmann Data is at the center of many challenges in system design today.

Network security Network protection comes from a wide collection of network devices, including load balancers, specialized hardware firewalls, and intrusion prevention devices.

Applications and services are split into segmented virtual local area networks VLANs to ensure that traffic doesn t cross application boundaries. Additional controls are also in place to lessen the impact of distributed denial of service DDoS attacks.

Compliance Compliance is a hot topic in cloud computing, especially for those looking to move enterprise applications to the cloud. However, note that this is compliance as far as the data centers go. As of this writing, Microsoft hasn t made any public comment regarding which standards Windows Azure is compliant with. This is one area where users should expect future statements and clarification. The Hypervisor With all of the machines in these data centers, the next obvious question concerns how an operating system actually runs on these machines.

The default option would be to install Windows Server directly on these machines and run them. This is, in fact, how most people run Microsoft software in their data centers today, and indeed is how Microsoft runs several of its services in its data centers.

Microsoft doesn t reveal how many machines it uses to run Windows Azure. The exact hardware used in Microsoft data centers also has not been specified, but as with most modern data center deployments, it is safe to assume that it is fairly homogeneous.

When managing thousands of machines, you want as much uniformity as you can get away with. However, there are two key problems with this approach: Sharing resources Henry Ford once said, Any customer can have a car painted any color, as long as it is black. Running operating systems directly on the hardware means the same holds true for machine size. Every application gets to use the same type of machine in the data center. However, if you have beefy, eight-core machines with tens of gigabytes of RAM, and several users each have tiny websites with no traffic, it makes little sense to give each user his own machine and make the user pay for the entire machine.

Migrating workloads You cannot move applications and workloads around easily. If Website A were experiencing low traffic and Website B suddenly goes viral and is taking a pounding, you cannot quickly take A s machines and give them to B. You would need to wipe A s machines to get rid of any custom installation and changes A might have performed, install Windows again, and then install B on top of it.

The industry solution to these and several other problems with running different workloads in the data center is to use virtualization. This refers to the capability to run several guest operating systems simultaneously on the same machine under a host operating system. Each operating system sees an entire machine typically called a virtual machine , and is unaware of the existence of the other operating systems. These virtual The Hypervisor In general, virtualization refers to abstracting away some hardware resource.

Most modern operating systems have a degree of virtualization built in. Different processes on the same machine all see the same amount of RAM, and can t get at virtual memory from another process. Multiple disks might be bound together under a RAID configuration and appear as one logical disk. However, this isolation isn t sufficient for today s cloud computing needs. However, this frequently proves insufficient.

The best way to fix that is to use one of the virtualization techniques described in this chapter. Hypervisor Architecture To create several virtual machines on one physical machine, a thin piece of low-level system software called the hypervisor or Virtual Machine Monitor VMM is used. The hypervisor is responsible for fair allocation of resources between the various virtual machines. To perform these tasks efficiently, the hypervisor relies on hardware support.

Though they are incompatible, fortunately for you and for all the developers building hypervisors they both work using similar concepts. In general, code executing in Ring N has access to all the same functionality as code executing in Ring All modern operating systems follow a model where the operating system kernel the core or heart of the operating system runs in Ring 0, and all user-mode applications run in Ring Chapter 2: Under the Hood.

For example, in Windows, ntoskrnl. This, of course, describes the execution of a normal operating system. Where does the hypervisor run if Ring 0 is the most-privileged level and all kernels expect to run there?

This is the problem Intel and AMD solved with their extensions. They added a new privilege level called Root mode better known as Ring 1. By adding an entirely new privilege level, all guest operating systems can execute unmodified as they used to in Rings 0 and 3.

This description may not be completely accurate. Some virtualization products used to move the guest operating systems into Ring 1 which isn t used for any other purpose and trap any privileged instruction in Ring 0. This has issues, though. There are several x86 instructions and architecture quirks that the VMM can t intercept running in this mode. For example, the instruction HLT immediately causes a generalprotection fault. This is typically worked around by modifying the guest operating system kernel to call through to the VMM in this case.

How does all this work? Let s say the guest operating system kernel is merrily executing away. By default, most x86 instructions don t require special privileges, so everything is executing as normal.

Since there are several guest operating systems executing in a data center, you must ensure that one doesn t trample over another operating system s page table. This sounds like a job for the hypervisor!

The processor recognizes that this is a privileged instruction and transitions from the guest operating system to the hypervisor, and the hypervisor executes some privileged code for the operation.

Since the hypervisor has knowledge of all guest operating systems on the machine, it can ensure fairness and security. Once the hypervisor is finished, it returns control to the guest kernel, which continues along its merry path. The Hypervisor However, let s say the operating system s creators knew beforehand that their operating system might be run on a hypervisor.

Could they take advantage of that somehow? As you might have guessed from the fact that the question was asked, yes, they can. Guest operating systems can communicate with the hypervisor they run on using a mechanism called hypercalls. The relationship between guest operating systems and a hypervisor is roughly analogous to a normal user-mode process and the kernel.

Just as a process can call an API to get the kernel to do something, the guest operating system kernel can invoke a hypercall that will result in code executing on the hypervisor.

An enlightened kernel is one that uses these hypercalls to notify and work along with the hypervisor to improve performance. The enlightening here refers to the knowledge the guest operating system kernel has of the hypervisor underneath it.

A canonical example of enlightenment is that of a long busy-wait spin loop. Typically, the processor would sit in a tight loop waiting for the spin lock to be released. If the spin lock code notifies the hypervisor with a hypercall, the hypervisor can schedule another virtual processor to execute, instead of wasting resources on the tight loop. Because of optimizations such as these, enlightened kernels can greatly outperform unenlightened kernels in several scenarios.

Windows Azure Hypervisor Architecture What you have learned thus far is relevant to all the hypervisors on the market today. Let s now look at one of them specifically: the Windows Azure Hypervisor. This hypervisor was designed with the following three principles in mind: Fast The Windows Azure hypervisor has been designed to be as efficient as possible. A lot of this is achieved through good old-fashioned, low-level optimizations, as well as pushing workloads onto hardware whenever possible.

Since Windows Azure controls the hardware in its data centers, it can rely on the presence of hardware features, unlike generic hypervisors designed for a broader market. Small The hypervisor is built to be lean and mean, and doesn t include any features that are not directly relevant to the cloud.

Smaller code not only means better performance, but also means less code to fix or update. Tightly integrated with the kernel Earlier, you learned how enlightened operating systems that integrate tightly with the hypervisor running underneath them can achieve better performance than unenlightened operating systems. In reality, both are different and are built for different purposes. Hyper-V ships as a part of Windows, and is meant to run on a broad variety of hardware for a broad variety of purposes.

The Windows Azure hypervisor runs only on Microsoft data centers, and is optimized specifically for the hardware that Windows Azure runs on. As you might expect with two similar products from the same company, there is sharing of code and design. In the future, new features from the Windows Azure hypervisor will make their way into Hyper-V, and vice versa, where applicable.

Figure shows the architecture of the Windows Azure hypervisor. Those familiar with Hyper-V will recognize many elements right away. Windows Azure hypervisor The Hypervisor User applications go into guest partitions, each of which runs a guest operating system instance. A guest partition has visibility only over its own address space, but doesn t have direct access to the hardware, and can t see, manage, or modify other partitions.

When the system boots the boot process for Windows Azure is covered a bit later , the booting instance is designated as the root or the host partition. This partition is special in several ways. It also performs one critical function: it is used to perform all hardware device access using standard Windows device drivers. This might sound slightly surprising.

Why doesn t the hypervisor deal with devices itself using its own device drivers? Indeed, this is the model used by other hypervisors in the industry. This lets the Windows Azure hypervisor and Hyper-V run on a wide array of hardware since Windows enjoys good driver support , and doesn t force hardware vendors to write new device drivers. A typical guest operating system, without enlightenments, could naively perform many discrete hardware accesses for a single logical operation.

This would involve several transitions between the guest and the hypervisor, and into the host partition. Using this mechanism, several chatty calls are consolidated into a few chunky calls, greatly improving performance. Even with all this discussion about operating system enlightenments and integration with the hypervisor, most developers only really care about what environment their code runs in the guest operating system.

The answer to that is simple. Applications run on Windows Server x64 Enterprise Edition. Well, almost. Microsoft calls it a Windows Server compatible operating system, which refers to the fact that it is Windows Server in almost every respect except the low-level changes to optimize for the hypervisor.

However, applications are abstracted several layers away from these changes, and shouldn t notice anything different from running on a normal, plain-vanilla Windows Server machine. Because some of these are obviously useful for Hyper-V, you might see them make their way into future versions of Hyper-V, or even Windows itself.

Performance features As mentioned, the hypervisor takes advantage of the homogeneous hardware environment to get better performance. One of the functions of the hypervisor is the capability to translate the guest s physical address to the system s physical address.

This was typically done in the hypervisor by using something called shadow page tables. Every guest had a mirror of its page tables created that kept information about the physical location of guest memory.

Every time a guest operating system changed its page table mapping, these shadow page tables had to be updated as well. This constant updating consumed CPU cycles and added to performance overhead. Newer processors such as the one used in Windows Azure s hardware allow some of this work to be done by the hardware itself. This allows identification of which guest virtual machine the TLB entry came from. Because the mapping is now stored in hardware, this greatly reduces the work the hypervisor must perform, and removes the need to update shadow page tables.

One side benefit of using SLAT is the support for large pages. Larger page sizes bring a host of performance benefits, since they can reduce the number of cache misses and deal with memory in larger chunks.

Apart from memory improvements, the scheduler was tweaked to be tuned for data center workloads. For example, scenarios involving ASP. Apart from this, improvements were made to reduce the number of intercepts into the hypervisor, as well as add enlightenments into the guest operating systems to make spin locks perform better. The risk of the installation failing on some of the machines and leading to these machines being in an unknown state is too high anyone who has had to administer a large server environment can attest to this.

To avoid these issues, the Windows Azure hypervisor uses an image-based deployment model. Operating systems are never installed on Windows Azure. An operating system image is constructed offline. Whenever there is a patch or an update, this image is updated. Separate images are constructed for the root and for the guest partitions and different flavors are offered for different services on Windows Azure.

Instead of having to install an operating system on thousands of machines, Windows Azure simply boots this VHD directly. Rolling out a new patch becomes a simple process of using XCOPY to create a new VHD, pointing the bootloader at it, and then rebooting the machine into the new updated operating system.

After discovering a bit about the hypervisor, you might be asking some or all of the following questions: How does the machine know which image to pull down, and how does it get started at the first boot? What happens when a new machine comes to life in the Windows Azure data center? What does the boot process look like? How does user application code wind up on the machines?

To understand the answers to these questions, you must first understand one of the most important components of Windows Azure: the fabric controller. The Fabric Consider how the operating system and programming frameworks abstract over memory. Instead of having to deal with individual RAM cells, the operating system and your programming platform provide several abstractions on top: With raw physical memory, programmers can allocate memory and deal with memory addresses, rather than individual memory cells.

With garbage collection, programmers don t have to worry about allocating or freeing memory, since that is done automatically for them. With the hot-adding of RAM, use of modern operating systems may allow for the addition of extra memory on-the-fly without having to shut down the machine. Now, consider how computers are managed in a typical infrastructure setup. Machines are usually referred to and named individually. Adding and removing machines often means some changes to code, or at least to configuration.

At a minimum, some network configuration changes will be needed. Relocating applications and data machines is a tedious, error-prone task, and is typically avoided as much as possible.

This would allow programmers to move from describing how they want things done to a model where programmers describe what they want done and the system does it for them. The fabric itself is a massive, distributed application that runs across all of Windows Azure s machines. Instead of having to deal with thousands of machines individually, other parts of Windows Azure and the users of Windows Azure can treat the entire set of machines as one common resource managed by the fabric.

The Fabric Controller Though fabric code is running on all machines in the Windows Azure world, almost all the heavy lifting is done by a small core set of machines known as the fabric controller. The fabric controller is often called the brain of Windows Azure, and for good reason: it controls the operation of all the other machines, as well as the services running on them.

The fabric controller is responsible for the following key tasks: The fabric controller owns all the data center hardware. This ranges from normal machines on racks to load balancers, switches, and so on. The fabric controller knows the state of each of them, and can detect failures on them at which time, a human being is notified if the failure can t be fixed programmatically.

Just as a normal operating system has drivers to deal with video cards, sound cards, and so on, the fabric controller has drivers that encode the knowledge of dealing with specific types of hardware. When new types of hardware must be brought online, a new driver is written to give the fabric controller knowledge of how to work with that hardware.

The Fabric We will discuss how these provisioning decisions are made a bit later in this chapter. The fabric controller maintains the health of all the services.

It monitors the services and tries to take corrective action upon detection of a failure. It also deals with upgrades and configuration changes to services. As you can imagine, a failure in the fabric controller can be bad for overall system health.

To help ensure reliability, the fabric controller is built as a set of five to seven machines, each with a replica, as shown in Figure All replicas share state with a Paxos-based replication system. Fabric controller nodes and replication Paxos is a distributed consensus algorithm, typically used to share some form of state between a distributed set of machines. See pedia. At any time, a primary machine is responsible for executing all key tasks. Any changes in state to the primary are automatically replicated to all the nodes in the fabric controller cluster.

If the primary goes down, a new primary is picked up seamlessly from one of the other nodes. Even in the unlikely case that all five to seven replicas go down, all user services will still continue to run until the fabric controller can be brought up. Because of the way the replicas are spread out in the data center, and how the code in the fabric controller is written, the chances of this actually happening are very remote.

The state on these replicas is quite valuable, since it contains knowledge of the running services. To avoid or recover from corruption, the state on each replica is checkpointed regularly.

This helps to prevent code or operator errors. How does the fabric controller get upgraded, changed, and configured? In the Windows Azure world, anything that needs to get upgraded goes through a fabric controller. So, in the case of the fabric controller, it is itself managed by another fabric controller called the utility fabric controller.

To understand the workings of the fabric and the fabric controller, let s walk through the life cycle of a service, from the time code is written on a developer s machine to the time it is running in a machine in a data center. Coding and Modeling In the Windows Azure world, you model your service and produce a service model with such things as worker roles, web roles, service configuration, and so on.

These are all nothing more than ways to define the topology of your service. To better understand a service model, consider how a website or service s architecture typically gets described in a setting. You start scribbling on a whiteboard or paper. You draw boxes to refer to the various machines or logical components that make up your service. You draw arrows between the boxes to denote how these machines talk to each other, and you might annotate these arrows with the protocols and ports used.

Now, consider the effort it takes to actually get that diagram working. A build gets generated containing the binaries for the various parts of your service, your SQL database scripts, and whatever else your service needs. Getting the right parts of the build installed and configured on the right machines usually involves some voodoo.

Everything from off-the-shelf tools to homegrown scripts is used to copy bits over and set them up. If you re lucky, you won t have to do any network changes, and your scripts are written in such a way that they are resilient to changes in machine names and environments. In essence, it is a lot of hard work to get from your code to that whiteboard diagram.

That is exactly what the service model does for you. In Windows Azure, each service from the simplest Hello World website to the most complex services has an associated service model. This service model is just a giant XML file that contains the same elements that your whiteboard diagram does. However, it describes them using well-defined elements such as roles, endpoints, The Fabric Once this service model has been defined, the fabric controller can basically automate what you had to do manually: copy bits to the right machine, perform the right networking configuration, and so on.

Moreover, the fabric controller maintains that service model. If some part of the model fails such as a machine going down , it takes corrective action, brings up a new machine, and configures the new machine with the right binaries and network settings so that your service model is restored. The key advantage of having such a model is that it defines what you want as opposed to how it should be done.

The service model only specifies the final configuration, and it is up to the fabric controller to map that to a set of procedural actions. For example, if the service model specifies a simple ASP. NET website with a single. Specifying things in a model-driven manner not only is easier and less error-prone, but also frees up the fabric controller to optimize tasks and parallelize execution of various such tasks across the system. Service configuration files and service models You might be wondering how to generate one of these service models yourself.

Generating one of these service models directly is a complex process, and the generated service models are quite gnarly beasts. Instead of forcing users to deal directly with the complexity of creating service models, Windows Azure ships with a few service model templates that users can use.

The web role, the worker role, and the CGI role are all examples of these templates. When you build a package, the CSPack tool uses your code and your service configuration to generate service models for your applications based on these templates. As discussed, the. If you extract the contents of the. Note that this file is useful to look at only as a curiosity it is complex and undocumented, changes from release to release, and should never be modified directly. In a sense, think of it as the raw output that the platform executes.

Following is a snippet from ServiceDefinition. You can see how elements from the service definition and configuration files show up here: 38 Chapter 2: Under the Hood.

This is the processed form of the configuration values defined in the ServiceConfiguration. They re used for everything from storing credentials to storing trivial application-specific properties.

Before wrapping up the discussion on service models, there are two final concepts worth covering: update domains and fault domains. Update domains Consider how you update a multimachine service. You typically update one component first say, the database or the web frontends. You then walk through the rest of the components in some specific order.

Update domains let you achieve the same in the service model. You can partition your service into a number of update domains. When you want to upgrade your service, the fabric controller will walk through each update domain in turn, and will start with a new one only when the previous update domain has been completed.

Update domains are also used in updates initiated by Windows Azure. When a new operating system image must be rolled out, each service is upgraded, update domain by update domain.

Fault domain A fault domain is a way of splitting up your service so that it is spread out across the data center.

No two critical resources are present within the same fault domain. Suppose that all your web roles are within the same rack in the data center, and the switch connected to the rack bursts into flames a very unlikely scenario, but play along here.

You would instantly see downtime as the fabric controller tries to bring up your web roles elsewhere. Unfortunately, as of this writing, there is no manual way to specify how your service should be partitioned into fault domains, and you must rely on Windows Azure to pick fault domains for you. This might be a feature exposed in a future release. The complete service model, along with the configuration settings and your application binaries, make up the package that you upload to Windows Azure.

The trio has all the information required by the fabric to deploy, run, and manage your application. Provisioning and Deployment Once the package reaches the fabric controller that is, the. This is essentially a constraint-solving problem the service model expresses constraints such as the number of role instances, the fault domains required, the local disk and machine size required, and so on.

The fabric controller looks across its machine pool and finds the right set of nodes to act as homes for these role instances, based on these constraints. Once it has made up its mind on the location where these role instances will live, the fabric controller starts the process of getting the bits onto the machine and getting the service running.

This is done in tandem with network hardware and the hypervisor running on these machines. How this is accomplished behind the scenes is fairly interesting.

Following is a brief synopsis: 1. In the beginning, all servers have nothing on them, and the fabric controller powers them on programmatically. PXE is a standard mechanism to boot computers using a network interface. It downloads a small operating system image called the maintenance OS and boots into it. That name is a bit of a misnomer, since it actually doesn t do any maintenance per se; rather, it bootstraps the machine and is the first bit of fabric code that runs on the target node.

All communication with the maintenance OS is through a fabric agent that lives on the maintenance OS. The maintenance OS talks to the fabric controller. After some secure handshaking, it sets up the host partition. The fabric controller tells the agent how many guest partitions to set up, and what VHDs to download. These are cached so that this download must happen only the first time.

For each guest partition, there s a base VHD that contains the operating system, and a differencing disk that contains any changes to disk. This is a standard practice used with virtual machines that lets you destroy the differencing disk to get back to the pristine base VHD state.

In the case of Windows Azure, resetting a guest virtual machine to its original state is as simple as deleting the differencing disk. Once the guest virtual machine is up and running, the specified role instance s files are copied onto the machine. Depending on the kind of role, different actions are taken. For example, a web role is launched in an http.

Currently, there is a strict enforcement of one role instance to one virtual machine. This helps ensure that the virtual machine is an isolation boundary both from a security and a resource utilization perspective. The fabric controller repeats this process for each role instance in the service model. Typically, only the last step must be performed, because most machines would have had the initial bootstrapping done already. The fabric controller programs the load balancer and other network hardware to route traffic from the external address assigned to the service say, for example, foo.

The role instances are placed behind a load balancer, which performs a simple round-robin algorithm to route traffic between the instances. Once the right networking routes are in place, traffic from the outside world flows to and from the role instances. Note how there s no installation in these steps. With everything relying on an XCOPY model, the chances of misconfigured or failed installations are lowered dramatically. Management and Monitoring The fabric controller maintains information about every role instance in the form of a state machine.

Every role instance has a goal state that it must get to, and the fabric controller tries to keep moving the role instance along to the goal state. This is important to note, because this forms a key part of how the fabric controller implements some critical functions in Windows Azure.

Updates don t have to be initiated by the user alone operating system updates security patches, fixes, and so on are rolled out periodically using the same mechanism.

Monitoring uses the same goal state features as well. The fabric controller monitors the health of all roles and quickly detects when a role dies. A role can also inform the fabric controller that it is unhealthy. When the fabric controller learns that a particular role is unhealthy, it updates its internal state machine and tries to push the role back to its goal state. This could range from restarting the role, to taking specific corrective action.

The fabric controller also monitors the health of the nodes in the data center. If the fabric controller detects that the node is offline for example, because of not hearing from the agent on the machine , the fabric controller migrates all role instances on the machine to a new node.

This new node is picked based on the constraints in the service model. For example, if the role instance in a failed node was part of a separate fault domain, the new role instance will be placed on a node in the same fault domain as well. The fabric controller then fixes up the network settings to restore the service model to bring the service back to its original state. Summary Frankly, you ll rarely be exposed to any of the details covered in this chapter. However, knowing how the system works underneath will help you understand how the various pieces of Windows Azure fit together.

For example, when you see a FabricInternalError in the portal, you can now guess at what is going on. Whether you re creating a new deployment using the management API, or optimizing your application for the hypervisor, knowing how the fabric controller and the Windows Azure hypervisor work their magic provides you a distinct advantage when programming in the world of Windows Azure. If you were off on an interstellar adventure, this would be the book you would want by your side at all times.

However, since the device itself looked insanely complicated to operate, and partly because intergalactic adventures tend to get scary at times, the book had the words DON T PANIC written in large, friendly letters on the back cover, always in uppercase.

In this chapter, you ll start building applications on Windows Azure. You ll begin by downloading the SDK, exploring it, peeking around its corners, and getting a basic application running inside a Microsoft data center somewhere.

Windows Azure is a vast, sprawling platform, and exploring it can look a bit daunting at first. But rest assured that it is all familiar Microsoft technology, with standard, easy-to-use APIs and friendly developer tools. To sign up, head to You ll be asked to create an account by signing in with your Live ID credentials and providing billing information.

The entire accountcreation process is quick and painless. You ll be using the portal throughout this book: for creating new hosted services and storage services, uploading new deployments to the cloud, managing your code and data once it is in the cloud, and other administrative functions.

First, to run code, you need tools to package your application into a format that Windows Azure can understand and run. Importantly, you ll probably need some tools to help you develop all this on your local development machine. Remember that services hosted in Windows Azure are all about running existing code inside Windows in a Microsoft data center Windows Azure is not a new programming language or programming framework.

Most of the work in running code is in packaging your application correctly and ensuring that it works in the cloud. When it comes to hosting applications, almost all the new concepts that Windows Azure introduces deal with how your application is packaged. There are very little to no new language features or fundamentally different programming idioms.

The idea is to let you run as much of your current code as possible. Doing anything on the cloud costs money, albeit in small incremental amounts. Doing as much development as possible locally can save you quite a bit.

You ll learn what these tools do later in this chapter. The SDK typically has a new version released every few months. The biggest add-on that the Visual Studio tools provide is a set of Visual Studio project templates, as well as the ability to debug locally.

Satisfying the Prerequisites Before installing the SDK and the Visual Studio tools, you must ensure that you have installed several prerequisites on your machine. You must have at least Windows Vista SP1 and. The SDK installs a mock version of the storage services in the cloud meant for local development. You ll need SQL Server either the full-blown version or the Express edition installed to make that work.

Installing these prerequisites separately can be a hassle. One easy way to install all of them with one tool is through the Microsoft Web Platform installer, available from Figure shows the Web Platform installer with the Windows Azure tools an early version set to be installed. Note how the necessary dependencies are detected and installed.

Note that this figure is reproduced from an early build of the software, and you ll almost surely see a different icon. Apart from the. Whether you call this binary directly, or whether it is called for you by a tool such as Visual Studio , this is what you use to package up your services. The letters stand for Cloud Service.

The actual acronym to use was picked close to the date of Windows Azure s initial launch, and there was a bit of discussion on the exact extension to use. Development Fabric and Development Storage These provide a simulated version of Windows Azure to help you develop code on your own machine, and test it before deploying against the actual cloud. We examine both of these in greater detail later in this chapter. Samples The SDK ships with the obligatory samples, as well as some useful tools masquerading as samples.