1. “Load Testing Early and Often” By Donald Doane Presentation to the Rockville MDCFUG

2. 2 Session Overview OpenDemand Systems Introduction to Load Testing Web Applications Testing from Development to Deployment Load Testing Over the LAN vs. the Internet Creating a Real-World Example in ColdFusion Load Testing Best Practices Identifying Common Performance Bottlenecks ColdFusion Performance Tuning Hints and Tips Questions & Answers

3. 3 The ColdFacts OpenDemand Systems Most developers are familiar with the “concept” of load testing, but very few actually do it! Most common reasons for not testing are Time, Complexity & Cost ($$$ can be very expensive) Most sites run into performance problems with only a few users (you don’t have to be Victoria’s Secret) Most performance problems are discovered too late in the game to properly address the issues

4. 4 What is a Load Test? OpenDemand Systems Simulates multiple users to measure, define, validate and maintain optimal application performance, scalability and reliability Serves as a compass for understanding the limits of any Web-enabled application and, subsequently, for managing its performance and growth

5. 5 Why Load Test from Development to Deployment? OpenDemand Systems  Identifies problems early on before they become costly to resolve (avoid the pyramid effect)  Prevents the “Onion-Peel” syndrome  Reduces development cycles  Produces better quality, more scalable code  Prevents revenue and credibility loss due to poor Web site performance  Increases customer satisfaction and retention  Enables intelligent planning for future expansion

6. 6 What does it mean for your site to perform or scale well? OpenDemand Systems Delivers page views to users in 8 seconds or less (industry standard) Performs consistently across multiple page view requests (avoid the wack-o- mole effect) Response times scale linearly as user load increases as opposed to exponentially

7. 7 Virtual User Simulation OpenDemand Systems One-to-many relationship between user and browser requests Browser opens multiple concurrent connections to Web server HTTP Request-Response Thread: GET / HTTP/1.1 Host: www.mysite.com ------------------------------- HTTP/1.1 200 OK Content-Length: 327 Content-type: text/html [BODY]

8. 8 OpenDemand Systems Your Browser Your Web Site/App Virtual Users Virtual Users OpenLoad Appliance Virtual Users

9. 9 Load Testing over the LAN vs. the Internet OpenDemand Systems LAN Testing:  Eliminates network variability due to the Internet  Saturates system rather than the network Internet Testing:  Uncovers hidden delays due to Internet weather  Exercises your pipe, routers, firewalls and load balancers in addition to the system under test  Ensures Service Providers are meeting SLAs

10. 10 Load Testing Do’s OpenDemand Systems  Do test early and do test often  Do establish what is and is not acceptable performance for your application  Do test from the user’s perspective - it’s the only one that counts  Do baseline and compare your findings  Do monitor your system while you test  Do test whenever there’s a change in your site’s content, code or infrastructure

11. 11 Load Testing Don’ts OpenDemand Systems  Don’t wait until the last minute to test  Don’t depend on your customers to do your testing for you  Don’t test under unrealistic conditions  Don’t forget that increases in table sizes, disk usage and network traffic will degrade your application’s performance over time  Don’t be so quick to throw hardware at the problems you turn up - it doesn’t always help

12. 12 Creating a Real-World Example in ColdFusion OpenDemand Systems Customer Case Study:  Set up a User Profile  Configure a Load Test Scenario  Schedule a Load Test  Analyze Test Results

13. 13 OpenDemand Systems

14. 14 OpenDemand Systems

15. 15 OpenDemand Systems

16. 16 OpenDemand Systems

17. 17 OpenDemand Systems

18. 18 OpenDemand Systems

19. 19 OpenDemand Systems

20. 20 OpenDemand Systems

21. 21 OpenDemand Systems

22. 22 OpenDemand Systems

23. 23 OpenDemand Systems

24. 24 OpenDemand Systems

25. 25 OpenDemand Systems

26. 26 OpenDemand Systems

27. 27 OpenDemand Systems

28. 28 OpenDemand Systems

29. 29 OpenDemand Systems

30. 30 OpenDemand Systems

31. 31 OpenDemand Systems

32. 32 Identifying Common Performance Bottlenecks OpenDemand Systems  Over design of Web pages (too many graphics)  Bad code (avoid recursive function calls)  Web server connection limit too low (IIS is set to 10 by default)  Poorly tuned Web application server (play with your Administration Server Settings)  Excessive queue lengths (more than one) & wait times (more than a few milliseconds)  Memory contention issues (paging in)  Heavy disk I/O (high service times)  Network latency (packet collisions & lost)

33. 33 Performance Tuning Hints & Tips OpenDemand Systems  Observe design & coding best practices (Fusebox)  Pre-load images before they are needed  Validate on the client whenever possible (JavaScript or ActionScript)  Strip formatting white space, tabs, carriage returns and line feeds from HTML (speeds up response)  Upgrade to CF5 (Up to 5x faster than CF4.5)  Tune simultaneous requests per CPU (cfstat)  Turn on “Trusted Cache” & increase cache size  Restart server at 5-10 unresponsive requests

34. 34 Performance Tuning Hints & Tips Cont’d OpenDemand Systems  Log slow pages (remember 8-second rule)  Enable Debugging (check processing time)  Replace frequently referenced Custom Tags with User Defined Functions (less overhead)  Bump up your cached query limit (if you can spare the memory)  Let ColdFusion work smarter and your Database work harder (write smart SQL - don’t do SELECT *)  Use stored procedures for faster execution

35. 35 Test your app for FREE OpenDemand Systems  Register at http://www.opendemand.com/cf/  Test for up to 5 virtual users for FREE!  Try-out performance tuning tips  Test again and compare results to your baseline

36. 36 OpenDemand Systems Q&A ddoane@opendemand.com http://www.opendemand.com