I am currently performing performance and load testing of a complex tiered system, investigating the impact of various changes, but I have problems tracking everything:
- There are many copies of various assemblies.
- Roughly released builds
- Officially released fixes
- The meetings I created contain additional additional fixes.
- Assemblies I created that contain additional diagnostic logs or traces
- There are many database patches , some of the above builds depend on the specific database patches that apply.
- There are different levels of logging at different levels (application logging, application performance statistics, SQL server profiling).
- There are many different scenarios , sometimes it is useful to test only one scenario, in other cases I need to check combinations of different scenarios.
- The load can be divided into several machines or only one machine
- The data available in the database can change , for example, some tests can be performed by the generated data, and then with data taken from a living system.
- There is a huge amount of potential performance data that will be collected after each test , for example:
- Many different types of application logging
- SQL Profiler
- DMVs
- Perfmon
- Gb , , , , , .
, ( , , ), - . , , , , , , . , , !
, , .
, , .., , - , , , .
/ , , , , -