Cloud Computing with the Windows® Azure™ Platform

Chapter 7. Optimizing the Scalability and Performance of Azure Tables

SQL Azure Database (SADB) offers most of the relational database management features of SQL Server 2008 Enterprise, but Azure Tables come to the fore when scalability is the criterion. The SDS team classified SDS v1 as suitable for most "departmental" database applications in its "First round of Questions and Answers" blog post of March 12, 2009 (http://bit.ly/cTWjP, http://blogs.msdn.com/ssds/archive/2009/03/12/9471765.aspx), and SDS architect Nigel Ellis announced at the MIX 09 conference (http://bit.ly/4v0B2y, http://videos.visitmix.com/MIX09/T06F) that database size will be capped at 5GB to 10GB. SDS is a premium service; therefore, users can expect a significant surcharge to hourly instance, data storage, and, potentially, data ingress and egress fees. These limitations and, especially, surcharges mean that financial executives will require .NET architects and developers designing and writing Azure application to justify substituting SDS for plain-old Azure tables (POATs).

Earlier chapters discussed how tables based on the Entity-Attribute-Value (EAV) data model, such as Azure, Amazon SimpleDB, and Google App Engine (GAE) tables, differ from their relational cousins. This chapter concentrates on helping .NET architects and developers get the most out of Azure tables' scalability benefits and maximize their performance in enterprise-scale applications.

Assigning Primary Key Values to Entities

Azure tables' composite ...