Licensing options for this format begin with FME Professional Edition.
Microsoft Windows Azure™ Tables is a cloud storage service that is part of the Windows Azure platform.
Azure Tables is designed to store simple tabular data in the cloud. In many aspects it behaves like a database. Azure Tables can store terabytes of data providing and is very is scalable. The key limitation of Azure Tables is that they are not relational, however Tables are not limited to a specific set of fields. Two different entity sets within a table may have completely different fields thus schema is a fluid concept within Azure Tables. At the row level, every entry has a Partition key and row key. These two keys together form the identity of the entry. The partition key is also use to load-balance the table across multiple servers.
The Azure Tables reader and writer does not include any native geometry support but provides a mechanism for custom storage. For example, a related format based on Azure Tables is Microsoft Windows Azure OGDI which adds its own geometry and metadata support. Currently, Azure Tables connects with the Microsoft Windows Azure architecture and retrieves data from the Table storage type. Reading and writing from or to Blob or Queue storage types is not supported.
As it is not a relational database, the Azure Tables reader and writer does not support a SQL interface. Instead the Microsoft Azure over SQL Server should be used if this is desirable. The Azure Tables reader and writer communicate with Azure servers directly through the REST API.
This chapter assumes familiarity with the Microsoft Azure Platform, tables, column types, connection parameters and cloud computing in general.
Connecting to Azure Tables requires knowledge of a Storage Service name or URL and a Primary Access Key that is granted access to that service. The Primary Access Key is usually provided when creating an Azure account for tat service and should be stored by the user as it is not easily human-readable but must be provided for new connections using the reader and writer. Usage of FME defaults is strongly recommended.
Additional information on the Microsoft Windows Azure Platform is available at http://www.microsoft.com/windowsazure/.
For more information on the Azure Table storage service, see the Summary of Table Service Functionality.
Reader Overview
The Azure Table Reader produces an FME feature for each row in a table in the provided source.
Reader Dataset
The dataset is the Windows Azure storage service. This value is sometimes referred to as the account name and is the first part of the URL used to access the service from a browser. For example,
http://<storage service>.cloudapp.net
Writer Overview
You can choose your own keys as long as every Partitionkey-RowKeypair is unique, or you can let FME create them for you.
If FME creates the keys, it will assign a unique Partition key per translation. This Partition key will be the same for all the features on the same translation to optimize batch transfer to the server. The row key will be a random UUID for each feature.
Writer Dataset
The dataset is the Windows Azure storage service. This value is sometimes referred to as the account name and is the first part of the URL used to access the service from a browser. For example,
http://<storage service>.cloudapp.net