Introducing Windows Azure Storage Architecture, here is a new kind of Windows Azure Storage. Just remember it is windows azure storage because the data management service on windows azure has several types of storage.
Windows Azure Table Storage is another type of windows azure storage, you might think of it as table but however I would prefer to think of it as special type of storage. You might think of it as table and comparing it to the SQL Azure, there are a lot of points of points that you may want to differ and others in common. You will find all the require things about the SQL Azure in this post. So the table storage is composed of tables and has something like the primary key for SQL Azure, called the partition key.
The partition key is not for the whole table it is made to partition your table along the whole infrastructure. In other words, let’s imagine a table with hundreds of names, when searching for a specific name; you will need to write a query that goes all along the table name by name. When using the partition key you will have the opportunity to partition the table into several partitions, for the previous example, let’s assume that you will partition the table using the first character of each name.
The second main thing is called the row key. That’s what defines the entity; it must be unique along the whole application. For example when adding a new entity, you can define the entity with a string, like its id or email.
So for the development, there are main two classes that must be created, one for defining what the entities will contain and the second one for the operations done on these entities like adding new entities, searching the table for entities or a specific one.
You can find the source code for an application I developed that adds and searches entities in the table in this link.