
Migrating to Sybase Adaptive Server Enterprise 11.5 7-9
Adaptive Server Enterprise 11.5 Test Cycle: Summary of Tests
Using a separate file for each transaction can prevent skewing;
however, it requires additional code to sync users. Using a separate
file for each user is more work to create but also prevents skewing
Using a memory generator works well to prevent skewing and
makes the test easy to administer. However, the benchmark is then
not 100% repeatable.
Test Cycle: Summary of Tests
This section summarizes a complete test cycle, with tests that target
specific issues, including old and new functionality, performance
under multi-user loads, integration, and user acceptance.
Table 7-2: Stages of Testing
Stage Purpose Best Technique
Functional
testing
For each application or process, addresses the
following questions:
• Are there any obvious bugs?
• Do transactions return the same results?
• Will the application break anywhere?
If you decide to use new functionality:
• Does the release provide the functionality it
markets?
• What are the limitations of the new features of
this release?
Single-user:
• Ad hoc test
• Manual test scripts and cases
• Existing application test
suites
Stress testing
(benchmarking)
Using very heavy loads, addresses the following
questions:
• Are there any bugs related to multi-user
loads?
• Is the performance of critical transactions as
good or better?
• Is the new release stable under load?
Multiuser:
• Keystroke capture
• Transaction generator
• Production load capture
Kommentare zu diesen Handbüchern