Load Testing Blog
Easily find an element with locator suggestions
Accurately and reliably locating the right element to interact with is one of the biggest challenges with real-browser testing, both in our products (QA Tester and Load Tester) as well as when coding tests to the Selenium/WebDriver APIs. Our upcoming 6.4 release provides suggestions for a variety of locators that may be a suitable replacement for the locator that was chosen during the testcase recording. If you are accustomed to using SeleniumIDE, you find that our implementation operates provides a familiar experience.
To access these suggestions, look for the light bulb icon next to the locator edit field. To … Continue reading »
Best Practices for Recording a Testcase to Work the First Time
Sometimes, HTTP testcase don’t work immediately after being recorded. Your application may require special configuration, or your workflow may need some special data entry in order to work in a repeatable fashion. However, sometimes the problem can be compounded by easily avoidable conditions.
Recommendation 1: Close unnecessary applications while recording
During recording, Load Tester will capture HTTP and HTTPS network traffic from your workstation as you record. This allows Load Tester to observe your recorded browser window, and child windows that may be spawned from it. If you have other browser windows open, e-mail clients, etc, these can all interfere with the … Continue reading »
Load Tester 6.1 Release
Load Tester 6.1 adds the ability to perform file uploads within real-browser testcases. See the link for instructions on using this new feature.
The release updates support for Chrome to include versions 33-36.
In addition, we have improved the user distribution algorithm specifically for real-browser tests. The real-browser replay logs now include annotations about datasource usage, so you can easily see, for example, which login was used from a dataset when an error occurred. See the change log for the additional bug fixes contained in the release.
How do you know if a load testing tool is accurate?
There are several different aspects to a question like this. The aspect that I’m going to address is: “How do I know that the load generator is accurate when it’s running a lot of users?” Or put another way, “How do I know that the load generator itself isn’t overloaded and is reporting inaccurate results?”
I could get all theoretical on you, citing a variety of technologies and design approaches used in our software to ensure it is accurate. I’d show some nice, official-looking white papers and report all the successes that our other customers have experienced using our software. But … Continue reading »
Load Tester 6.0 Beta 2 – Real Browsers
Load Tester 6 introduces an exciting new capability: generating load with real browsers. This technology, leveraging the Selenium/WebDriver project, brings a number of important advantages:
Testcase configuration and debugging is more intuitive
Test results accurately describe the end-user experience
Testing very complex and sophisticated AJAXy applications is much easier
Here is a pic of what a testcase looks like when using real browsers for testing. We hope you’ll agree that it is much more intuitive:
We are confident these improvements make Load Tester a more efficient and pleasant tool to use – but now we want to hear what YOU think about it! Are … Continue reading »
Accuracy of Load Testing with Real Browsers vs Virtual Browsers
Comparing the two directly is not an apples-to-apples comparison, because they measure different things. Actually, Load Tester measures a lot of the same things for each, but what is generally considered the most important metric – Page Duration – is actually measuring a different aspect of performance in each case.
Virtual Browsers
Virtual Browsers work at the HTTP layer – they send the same HTTP messages to the server that real browsers would send. The Page Duration measures the time from the beginning of the first request that is sent to the server to the end of the last response for a … Continue reading »
DynaTrace Support in Web Performance Load Tester 5.4
DynaTrace (now known as “Compuware APM for Enterprise Tiers”) is a tool to analyze performance as requests pass from the front-end webserver back to various application servers, database servers, and web services. It’s marketed as a way to obtain deep insight into performance problems without the overhead and inconvenience of other more intrusive profiling tools. During a load test, DynaTrace works best if requests are annotated with metadata. We added a feature to emit DynaTrace-compatible metadata to Load Tester 5.4.
DynaTrace support is a premium feature sold separately from Load Tester PRO or our other offerings, but you … Continue reading »
Scheduling Load Tests with Load Tester 5.4
Many of our customers want to run tests off-hours to minimize collateral inconveniences. For example, if the rest of your QA team is working on the same test server as you, it might behoove you to run a load test at 2:00 AM when the team is asleep. Load Tester has had the ability to schedule off-hours tests for years, but the feature remains a frequently asked question among potential testers.
In Load Tester 5.4’s streamlined user interface, you can schedule a test from the ‘Control’ menu by choosing “Schedule Load Test.” A “Scheduled Operation” dialog will appear, and you can … Continue reading »
How Much Bandwidth Do You Need To Do a Load Test?
One of the common questions people are interested in is finding out how much a test is going to cost. There are a lot of factors that go into this equation, such as getting an appropriately sized testing license, configuring a test server environment, reserving hardware for Load Engines, and bandwidth costs between the Load Engines and the content delivery servers (such as a CDN or origin servers). Let’s take a look at just how we might calculate the bandwidth charges that are involved in a single test.
If your site is only needs to support a few hundred users, then … Continue reading »
Measuring Web Page Load Times using JMeter
At last count there were over 65 separate commercial load testing tools out there, but few with the name recognition of the open source program JMeter. Often people will call us up and ask to compare Load Tester with JMeter, but I only had a cursory look at it many years ago, and couldn’t speak from recent first-hand knowledge. So, when someone called me last week asking about JMeter, it seemed like a good opportunity to give it another look.
Web Performance Consulting