Load Testing Blog
How Many Accounts do you need for Load Testing?
Eventually in every planning meeting for web load testing the question comes up: can we test with just one user account?
The short answer is “no”.
The longer answer is “heck no”.
The best practice is always to simulate real users as closely as practical, which in this case means concurrent simulated users should login on separate accounts, just like real users. If the test cases are idempotent, i.e. running them doesn’t affect the state of the account, and user accounts aren’t cached, then the answer is easy: at a minimum you need as many test accounts as the maximum number … Continue reading »
How to Find Hidden Web Server Load Testing Bottlenecks in Scatter Plots
If you’ve been using average page load data to evaluate your website’s performance, you could be missing important bottlenecks that are reducing your conversion rates and causing customers to have a frustrating experience. While averages are easy to understand, they cover up all of the details. After all, if the average load time of a page is 2 seconds, many users are seeing load times slower than that, but we don’t know exactly how slow. The industry standard of expressing site performance in percentiles is technically accurate, but also doesn’t give a full picture.
Here’s a good example. A recent client, … Continue reading »
Load testing with OAUTH authentication
There has always been a race between hackers and web security experts. Performance testing professionals are caught in the crossfire, since the same techniques used by hackers are the ones we use to configure test scripts.
OAUTH 2 is the de-facto authentication standard these days, and while thanks to Load Tester’s advanced editing it’s not too hard to configure a Load Tester script, there are a few extra steps you’ll need to take. Unfortunately there’s no automatic way of configuring a script since each implementation can be different. This article will show how to … Continue reading »
Recording with Brotli Encoding
Load Tester 6.7 and earlier does not support Brotli encoding. This is obviously planned for future releases, but for now you may need to use this workaround. If the site just has brotli on static pages there’s no need to do anything special, but if you are testing a site that is using Brotli encoding on security API calls that return session variables you’ll want to read on.
Brotli encoding has been gaining in popularity because it was designed from the group up for streaming. As a result, it reduces latency over gzip, which was originally designed for files … Continue reading »
How to Disable Unneeded Extractors and Better Handle Web Page Variations
ASM or Automated State Management is one of Load Tester’s best features, saving loads of time by automatically detecting and configuring dynamic web fields. This makes Load Tester many times faster than trying to configure a test case using a manual tool like JMeter.
The downside is that sometimes the logic picks up fields that don’t really change dynamically, which can have a downside of generating parsing errors when the page layout is different for different users.
Long term you’d probably want to adjust the detection rules to avoid configuring too many fields, but in the short term the Fields … Continue reading »
Load Testing Salesforce Apps
CRM functionality is so integral to many business workflows that writing a Salesforce app is often cheaper and faster than starting from scratch. Why reinvent the wheel with 50%+ of the functionality you need is already in Salesforce? But just like a 100% custom app, Salesforce apps must be tested for performance, not just functionality, especially when integrating into an existing customer back-office or 3rd party system. And Salesforce is one of the more difficult platforms to test because of the complicated field naming conventions typical of auto-generated front ends.
The first challenge is just the sheer overall complexity of a … Continue reading »
Web Performance Announces Management Reorganization
Due to overwhelming customer support for our services, Web Performance has undergone a management reorganization, which means we are in a strong position to continue operations for the foreseeable future.
Load Testing with Microsoft Edge
Recently Microsoft has been pushing their new Edge browser release by forcing users to start it upon login. If you want to start recording with Edge then Load Tester has you covered with just a few configuration steps.
First, go to the Browser settings in Load Tester: Window->Preferences->Web Performance->Browser
Click the Add Button on the right and configure a new setting for Edge. The path is as shown above for Windows 10.
Be sure to check “Use Operating System-wide Proxy Settings”!
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Click Apply, and then OK.
When recording you’ll now see a new option to record with Edge:
You’ll probably … Continue reading »
Generating Dynamic UUID with a Javascript Data Source
We recently had a customer question on how to dynamically generate UUIDs during a load test, and the answer illustrates two interesting techniques in Load Tester:
How to use javascript to dynamically generate data during a test.
How to make call outs to the more extensive Java functions within javascript.
In this example let’s assume the UUID is being used as a path element that you want to dynamically replace during a test like this:
http://mysite/path1/123e4567-e89b-12d3-a456-426614174000
Like much of the test case configuration in Load Tester, the process to edit the dynamic behavior starts in the Fields View. Click on the URL to select … Continue reading »
Distinct Think Time setting for Virtual User Pacing
Load Tester 6.7 includes a new option for Virtual User Pacing. When configuring Think Time Generators (Window > Preferences > Web Performance > Think Time), the new checkbox is labeled Distinct – Allow users to have distinct paces. This applies only to the random variation generators. When turned on, it changes how the random variation range is applied to think time.
When turned off, the think time generator will function as it always did – for each think-time encountered, it will vary the think time (as configured in the testcase) by a random value within the configured range. When turned on, … Continue reading »