By David Postlethwaite
On the Database Scale page we can decide on the performance level that we want our database to have.
There are three performance levels Basic, Standard and Premium

• Basic is best suited for a small size database, supporting one single operation at a time such as for development or testing, or small scale infrequently used applications.
• Standard is the best option for most cloud applications, supporting multiple concurrent queries such as workgroup or web applications.
• Premium is designed for high transactional volume, supporting a large number of concurrent users and requiring the highest level of business continuity such as mission critical applications.

There used to be two types called Web and Business they have now been retired but you’ll still find them mentioned in the documentation and it is still possible through SSMS to create a web level database. (Upgrade SSMS 2014 to CU5 to get the latest) They are due to be removed in September 2015.
Unlike your traditional on premises database server you cannot choose the number of CPUs, amount of memory or disc configuration with SQL Azure DB.
Each Service tier has different Power levels which are measured in Database Throughput Units or DTUs. A DTU represents the power of the database engine as a mix of CPU, memory and read and write rates.

The table below show you the power levels available for each tier. Expect these to increase over the coming months and years.
The idea of a DTU is to show relative power that there is between the different levels So S1 will give you 4 times the power of Basic and P1 is 10 times more powerful than S0.
To give you some idea of what you might expect from each level Microsoft have published a benchmark transaction rate.

You decide on the power level you think you require and then monitor and tune the database as required. It is very easy to change from one tier to another if you decide you need more or less power but changing the level will cause the database to stop and restart. Remember, each server can have a maximum of 150 databases and a maximum of 2000 DTUs

Visit our SQL Server and Azure Consulting page