Above The Cloud

If you are new to Azure you might be overwhelmed by what the various services do and don’t do. I don’t think the Azure site has done a good job of explaining how each service is used separate unto others.

Then today when I was searching the Azure Forums for something else, I came across this thread where Yi-Lun Luo explains what each of the four services is used for and some basic information about each.

Quoted:

Windows Azure:

• Write, read, and delete resources (in the form of Blob/Queue/Table) containing from 1 byte to 50 gigabytes of data each. The number of files you can store is unlimited.
• Each resource is stored in a container and retrieved via a URI in the form of <http|https>://<account-name>.<service-name>.core.windows.net/<resource-path>.
• Currently all data is stored in the United States, but can be accessed from anywhere. We’re building more and more data centers around the world.
• A container can either be public (the requests do not need to be authenticated) or private. Currently you can’t specify access control on a specific resource.
• Uses standards-based REST interfaces designed to work with any Internet-development toolkit. We’ve also built a wrapper library on desktop .NET platform (not Silverlight or .NETCF) for easier working with the APIs. In the future, maybe SOAP APIs will also be provided.
• Currently only http/https are supported. Additional protocols will be supported in the future.
• A resource is at least duplicated for three times to prevent data loss.
• Ability to host server side partial trust web applications in the cloud. Currently the only supported technology is .NET. Additional technologies, such as Java and PHP, will be supported in the future. Also less security limitations will be forced in the future. Client side applications can either be embedded in a web page (Silverlight/Flex) in a web role, or deployed on the user’s desktop and work with the storage APIs.

Live Services (AKA Live Mesh):

• Synchronize any kind of file on any device a user owns. The data will be stored in the Mesh as well as on the devices. Up to 5 GB data can be stored in the Mesh. Larger space will be provided in the future. Each file doesn’t have a size limitation.
• The device can be PC running Windows, Macs, Windows Mobile Devices, and Mesh Web Desktop. More devices will be supported in the future.
• All devices, files, applications, and other kinds of data stored in the Mesh are considered to be resources, and can be accessed in any authenticated applications.
• The end user controls which resource to be private, and which to be shared with friends.
• Resources can be accessed and updated through Atom Feeds in the form of REST APIs. A .NET library, a Silverlight library, and a JavaScript library are also provided for easier working with resources.
• Provides mechanisms to run web applications (Silverlight or AJAX) offline, outside of the browser. Modification of the data will be synchronized the next time this device goes online.
• Built upon Windows Azure. Can be taken as a case study of services hosted in Windows Azure.
• Part of the Live Platform which contains a lot of other services, such as Live Search and Virtual Earth. But I don’t know much about them…

.NET Services:

• Provides Service Bus to connect existing services behind firewall and NAT, and integrate applications.
• Provides Workflow Service to automate business process which involves calling a lot of services.
• Provides Access Control to determine which application/service can call a specific service or participate in the workflow.
• A desktop .NET SDK and a Java SDK have already been provided. More technologies will be supported in the future.
• Service Bus supports the http/https/tcp protocols. Services can communicate either one way or two way in any format (xml, binary, etc). It also provides NetEventRelayBinding to support multicasting.
• Access Control supports username/passwrod, CardSpace, and X509 authentication. More will come in the future.

SQL Data Services:

• Provides a cloud database to store data in the relational architecture.
• Supports CRUD operations over entities through standard-based REST and SOAP interfaces designed to work with any Internet-development toolkit.
• Each entity belongs to a container. Each container belongs to an authority. Entities do not have a specific schema. Two entities in the same container can contain different properties. This is called flexible schema. In the future, strict schema (similar to columns in traditional databases) will also be supported.
• Supports cross entity join queries in the same container. Cross container queries will be supported in the future.
• Provides blobs to store large data up to 100MB. Larger data will be supported in the future.
• More features, such as data warehouse and data mining, may come in the future.

Thanks for such a great explanation Yi-Lun!