Tuesday, January 13, 2009

Requirement gathering for Performance testing: Part-3

This posting explains about "Requirement gathering for Performance testing - Business Process"

Business Process:

1. What are the user initiated activities that you want to simulate in the script? Please list.
• Will you be providing the use cases?
• Will you be providing the load testing scenarios?
2. What is the user ratio for each activity?
(i.e. activity1 – 70% of load, activity2 – 15%, and activity3 – 15%)
3. Are there any business process dependencies that impact the user activities you defined above?
• Does one process need to run before another process?
• Can the activities run concurrently?
4. We generally measure each transaction sent/received. Are there other requirements that you are looking for?
• Measure each page or groups of pages that comprise a business flow?
5. What are the expected response times?
• What are the maximum acceptable response times?
6. How do you see your users accessing the application?
• Do they all log in at the same time?
• Sporadically throughout the day?
• Certain number at specified intervals?
7. What is the activity execution frequency?
• How often are the users doing each activity in an hour, a day, etc.
• How many transactions occur in an hour, a day, etc.
8. What is the expected throughput of the architecture?
• Throughput can be expressed in terms of hits per hour, day, etc.
• It can also be expressed in terms of how many bytes per second or transactions per hour.
9. Are there specific requirements for the input data?
• Data validation (credit cards, zip codes, etc.).
• Data uniqueness (you cannot enter the same data more than once).
• Is it time sensitive?
• Is the data reusable within the test?
• Is there cleanup—and how often?
10. How long will it take to obtain all input data?
• Customer records
• Logins and Passwords
11. Are there any input data dependencies?
• Is data prepping required?
• Does form input data have to be unique? (I.e. can all users have the same address?)
12. What is the data refresh strategy? Is one required?
• Does the DB need to be restored between runs?
• Is it possible to do a rollback?
13. Has disk space been evaluated to assure that there is sufficient disk space to support the high volume of users and activities? (Ex. log files, other flat files)

14. Will batch processes be executing concurrently with the performance testing? Are there any you think should be included on at least one test run?

15. Is there sufficient Database space to support high volume/users?

Related Articals:

Requirement gathering for Performance testing: Part-1

Requirement gathering for Performance testing: Part-2

post your comments to testingblog4u@gmail.com

No comments:

Performance Testing News:

Loadrunner v12:
- Targeting an April 2012 Development release. Not sure when it will be available for public.
- Customers can participate in the beta program after legal procedures are completed.
- Availability of LR Mobile protocol. Performance testing for mobile (both web browser based as well as native mobile device) – Supports HP Palm Pre, iPhone (3GS, 4), Android, Blackberry phones and many other actual mobile devices.
- Support for Windows7, TruClient IE9.
- VUGen will be completely redesigned and new IDE (called Sharp Develop) will be built.

Performance Testing,Monitoing & Analysis Useful Documents

Download Performance Testing,Monitoing & Analysis Useful Documents


Loadrunner

Design scenario - Download

Record & Replay Issues with VUGen - Download

LR8.1 Protocol Footprint - Download

Monitoring & Analysis

HP Performance monitoring best practice - Download

Others

Load test plan - Download