Introducing NeoLoad 9.0

With the release of NeoLoad 9.0, we are pleased to offer our new RealBrowser technology for all users

Modern applications are constantly evolving, and with the advent of integrated DevOps toolchains applications can be built and deployed at ever-increasing velocities. This puts pressure on quality assurance teams to increase their velocity as well. With NeoLoad RealBrowser technology, we now offer a rapid way to record and immediately execute user paths to validate the performance of application builds as they move through their pipeline.

Complex single-page applications can be extremely difficult to create user paths at the protocol level. During development and test cycles, the importance of delivering high-quality, relevant results in a timely fashion is critical to maintaining high-velocity delivery. RealBrowser allows developers and testers to record a scenario and then execute using a method that abstracts away all protocol scripting complexity, creating a test asset that is highly resilient to changes in the application itself.

Not only can teams now create tests in minutes, but the test artifact itself requires minimal maintenance. NeoLoad has always had a significantly lower script maintenance cost for organizations, and RealBrowser now reduces this maintenance even further, allowing teams to focus on delivering software that is highly performant.

What is RealBrowser and how does it work?

RealBrowser technology allows you to record a user path based on end-user browser actions that take place, like mouse clicks and keystrokes. Unlike a protocol-based script that sends the messages directly from the network interface to the servers hosting the application, RealBrowser allows users to interact with the rendered application inside a browser object, offering two paths to end-to-end performance testing. We accomplish this using a browser-based testing framework that allows us to support any chromium browser, at a significantly smaller memory and CPU footprint compared to our competitors.

RealBrowser technology is available to all current NeoLoad customers who upgrade to version 9.0. Our customers will need to migrate their licenses to the new license model we released at the beginning of 2022. RealBrowser technology is an addition to all existing functionality and offers new ways for you to record your applications and execute performance tests.

Should I always use RealBrowser?

Not at all! Even though all existing NeoLoad customers will receive full access to RealBrowser technology once they update to the current license model, this in no way eliminates the protocol-based user paths that our customers have been using for years.

RealBrowser technology allows us to render the HTML page directly, and this comes at a cost of additional memory and CPU consumption for each virtual user in the scenario. As you move from rapid development cycles to more stable performance engineering and pre-deployment testing, the scope of your test scenarios will increase, meaning there are potentially enormous cost savings for continuing to leverage protocol-based user paths.

Pros of RealBrowser

• Extremely rapid test development
• User paths are highly resistant to application volatility
• Skill requirements for team members creating test assets are substantially reduced, opening a much larger pool of team members who can participate

Cons of RealBrowser

• Higher memory and CPU cost for execution

So, when *do* I use RealBrowser?

Use Case 1: Complex applications

Many applications are written to provide a highly dynamic user experience, relying on complex JavaScript to provide a seamless user experience. This can cause extremely difficult correlation and parameterization activities to be required on every user path you create. With RealBrowser, instead of spending days or weeks trying to create a test for these complex applications, you can instead have a test up and running within minutes, providing meaningful feedback while it is still timely and relevant.

Use Case 2: End user response time

Traditional performance test scenarios leverage protocol-based execution, which sends the message traffic directly to the server, and catches the response as soon as it arrives on the network interface. This is known as transaction response time, and includes the time taken to send a message across the network to the servers, for the servers to process and respond to the request, and for the message to arrive back at the NeoLoad load generator.

End user response time is an increasingly relevant metric and encompasses the additional time that is spent by the end-user’s computer to render and display scripts, HTML content, and images. This metric is precisely what the real users will experience, and being able to collect and report on this is something that many of our customers require. With RealBrowser technology, you can now easily collect this information while staying entirely within the NeoLoad solution.

Use Case 3: New to performance engineering

Team members who have never created a performance test asset are increasingly being pulled into performance engineering activities. Users who do not have this background still need to create performance assets and execute them against new build of an application extremely rapidly, often needing testing to be complete within hours of the build finishing.

With RealBrowser’s significantly higher ease of use and lower skill barrier to entry, the pool of team members at your organization that can participate in performance testing activities is substantially greater, and they can deliver performance data directly back to the development team within the same agile delivery cycle.

Use Case 4: Rapid performance testing requirements

If you are working with a development organization that leverages an integrated toolchain to build and deploy application features rapidly, they will require timely feedback on the performance impact of adding additional functionality. Traditional performance testing activities can take days, by which time the development teams have moved to the next features. This delay adds costly refocusing of effort as you pull a development team from new features back into existing work they have already completed.

With RealBrowser, you can now rapidly create performance tests in minutes and provide meaningful performance test results back to the development team within hours of a build being completed. Better still, NeoLoad comes with a robust Command Line Interface (CLI) that allows full automation of performance tests as part of a pipeline. The robust nature of RealBrowser test scenarios means that existing assets can be run without changes in many cases.

Any practical advice for us?

First and foremost, RealBrowser is still NeoLoad. All the skills you have developed leveraging our solution are still 100% relevant while you take advantage of the RealBrowser technology.

To determine the size of your RealBrowser virtual users, you need to look at your application. The CPU and memory consumption (the footprint of the virtual user) on a per-virtual user basis is determined by many factors, notably the amount of content that needs to be executed on a webpage. If your web application contains significant dynamic JavaScript and other page elements, then the size of your RealBrowser virtual user footprint will be larger. Determining this value can help you size and scale your load generators appropriately.

You can also freely mix and match RealBrowser user paths with protocol-based user paths. This allows you to drive load onto the system, while still giving you a subset of users that are capturing true end-user response time. More than that, you will be able to compare in your results reporting the difference between end-user response time and transaction response time, gaining additional insight into how server load impacts client responsiveness.

We are excited to release RealBrowser technology for all our existing NeoLoad customers. To learn more, check out our webinar on Nov. 10. We encourage you to upgrade today to take advantage of this new capability!