Database Reference
In-Depth Information
common fields are stored in one database and the field containing documents is stored in another
database.
A VPS does bring it own set of challenges, such as the difficulty of keeping strong RI between
databases and, in the case of SQL Azure, ensuring transaction consistency. These considerations are
only meant to bring certain design issues to the surface when you're thinking about a VPS. Depending
on your database design, a VPS may work very well and may be simpler to implement than an HPS.
Conclusion
As you've seen, building databases with SQL Azure can be complex if you need to scale your database
design or if you're looking to develop a high-performance system. This chapter introduced you to the
design and implementation of a shard library that gives you the necessary building blocks to experiment
with a flexible and scalable architecture.
You saw how to use caching in a data access layer and how parallel processing of SQL statements
can increase performance of your applications. You also explored more advanced topics, such as
referential integrity and exception management. This chapter provides you with the necessary
background to create your own shard and create high-performing applications against SQL Azure
databases.
