HOTSPOT
You have an Azure SQL Database named Contosodb. Contosodb is running in the Standard/S2 tier and has a
service level objective of 99 percent.
You review the service tiers in Microsoft Azure SQL Database as well as the results of running performance
queries for the usage of the database for the past week as shown in the exhibits. (Click the Exhibits button.)
For each of the following statements, select Yes if the statement is true. Otherwise, select No.
Hot Area:
Answer: 
Explanation:
The P1 performance level has 100 DTUs compared to the 200 DTUs of the P2 performance level. That means
that the P1 performance level provides half the performance of the P2 performance level. So, 50% of CPU
utilization in P2 equals 100% CPU utilization in P1. As long as the application does not have timeouts, it may
not matter if a big job takes 2 hours or 2.5 hours to complete as long as it gets done today. An application in
this category can probably just use a P1 performance level. You can take advantage of the fact that there are
periods of time during the day where resource usage is lower, meaning that any “big peak” might spill over into
one of the
http://msdn.microsoft.com/en-us/library/azure/dn369873.aspx
How do we calculate the actual DTU from the data provided?
Current tiers. https://azure.microsoft.com/en-gb/pricing/details/sql-database/#single-database
You can’t calculate the DTU from the info provided as it is a magic number Microsoft come up with.
What they are getting at is that the “Maximum CPU Usage” is high, but the average is low. This suggests there is one job that runs and take the CPU high for a short time, and then finishes. Maybe it happens one or twice a day.
Assuming this job or query isn’t needed immediately then if it take twice as long to run this is fine. So it could be set to S1 and the database still run well.
Basic has only 5 DTU and the Premium start at 125 (used to be 100) DTU, so these are too little and too much.
https://docs.microsoft.com/en-us/azure/sql-database/sql-database-single-database-monitor#monitor-databases-using-dmvs