Azure SQL Database supports large numbers of databases for SaaS providers
Опубликовано 29 апреля 2015 г.
Software-as-a-service (SaaS) providers face many challenges. One architectural pattern for SaaS applications involves creating one database per end customer, but purchasing and overprovisioning individual databases to meet variable and peak demands for each database is often not cost efficient. In addition, managing many databases presents its own set of issues. Now instead of overprovisioning to meet peak demand, you can allow hundreds or even thousands of databases within an elastic database pool to spike and dynamically consume resources within a prescribed budget. Elastic database pools provide you with the ability to combine resources to be allocated among a group of these databases. The elastic database pricing model, new in public preview, enables predictable pricing over a set of unpredictable databases. You get the flexibility to support explosive growth and profitable SaaS business models, while gaining cost efficiencies. And with elastic database queries (in public preview starting May 29, 2015), you can run centralized query operations (such as reporting and data extractions) on multiple databases, while getting unified results from the entire database set. Also in public preview now, you can manage and monitor a large number of databases by running T-SQL script-based jobs from the portal, including permissions, schema changes, and maintenance. To help developers scale their applications across a large number of databases and improve time-to-market, the elastic database client library and split-merge tool (previously code-named “Elastic Scale”) are now generally available. The elastic database pricing model is available in public preview for the Standard tier. For more information, please visit the SQL Database page, and read this blog post: Azure SQL Database previews major updates for BUILD. For more information about pricing, please visit the SQL Database Pricing page.