How Team Testing in Spirent CyberFlood™ will Accelerate your Testing

CyberFlood dashboard

In the real world, how do most people create a test case? In most instances, they open the tester GUI, configure the test, debug the test, run the test, print the report, and are done. The question when we drill deeper is “Is this the most effective way for us to be performing our testing?” Let’s drill down into the problems this technique creates for organizations. Generally, many test engineers are given a test set which may include a full DUT or specific test ports on a shared DUT. The configuration of specific attributes of these ports such as routing/IP information, firewall rules, ALCs typically do not change unless they are specifically being tested. In the classic model, the user keeps rebuilding attributes of the test over and over again potentially building the same test hundreds of times.

This redundant action has no testing value and is wasteful and costly when considering the users time, depreciation of the tester, reduced time-to-market, and potential user input error debugging time, etc. In addition, this approach does not reflect modern test practices such as multisite or follow the sun testing.

For example, how would another engineer halfway around the world know how specific local tests should be configured? The problem is also compounded when the component levels of the test case are multilayer and dependent upon the previous layer being correctly configured and interpreted. Even with fully configured test cases, the act of exchanging configuration files can lead to hidden costs of managing files, versions, time spent explaining how to import and use the files, etc. From a management perspective, classic approaches do not allow a manager the ability to see an overall snapshot on how testing is progressing, nor does it encourage a manager to access the results in a centralized location. Typically, a manager has to manage many test results in multiple formats ranging from PDF, CSV, HTML, to even printed paper copies.

When measured, there is tremendous undocumented time and energy spent managing project planning and results interpretation. Finally, as the scope of projects scale there are hidden scaling costs to all of the aforementioned issues. Simply put, the larger the projects scale, the more hidden cost is associated with day to day and top down management. The energy spent processing these costs is inefficient and would not lead to a better and timelier product.

Spirent CyberFlood is designed from the ground up to support what we call Team Testing. In this model, enabled by our modern true HTML5 web GUI, CyberFlood becomes the hub of test objects, users, test cases, results and management.

At a fundamental level, test cases are built up by using various sorts of profiles. By allowing the user to create profiles independent of the test case, we allow the user to create once a reuse many times. This saves time and improves efficiency in many ways. First, when a user is creating a profile, they are just working on that part of the test and are less likely to misconfigure settings because they are not surrounded by a complicated GUI.  Even if the user misconfigures the profile, they are given the option in the test to correct the profile without jumping out of the context of the test case. The changes are then saved back to the profile. Once the profile, say the IP addressing scheme for a port, is perfected then future test cases can reuse the profile object by dragging and dropping the profile into the test case. Therefore, the more test cases you build, the easier and faster it is to get future tests up and running. In addition, profiles are exposed to all users in the test ecosystem, so if an another user in another lab wishes to used that profile, they get the same economies of scale.

Fundamentally, as your users use CyberFlood, they are not only testing the current test case but also simultaneously building parts of future test cases. Because CyberFlood uses a shared HTML5 web GUI, there is no longer a need to email configuration files to colleagues. Now, a user halfway around the world can take a test case and either run it on their own ports or relocate test ports and continue testing 24x7x365. They can even clone the test case and make their own modifications creating a test case variant.

Additionally, we help the testing process by syncing new content continuously down to CyberFlood. This may include new protocols, new malware/attacks, or code improvements. By not forcing our users to create administrative windows for downloading and installing new code or content, we preserve test time. Of course, for our customers that have restricted labs, we do allow you to sync those files as well.

Another way that CyberFlood helps you optimize testing is through test queueing. Classically, we wait for one test case to finish then load up the next test and run. When this occurs in the middle of the night or over the weekend, which can be typical, hours to days of productivity are lost. With CyberFlood, users can queue up many test cases behind a running test. This helps optimize the testing utilization rate and promotes more testing.

From a project management perspective, CyberFlood allows a manager to, at a glance see test queues in play and have access to results in a centralized, web-based location. This promotes a better and more accurate snapshot of the state of testing and overall a more efficient testing process.

Learn how CyberFlood supports global team testing for application layer performance by visiting http://www.spirent.com/Products/CyberFlood.


comments powered by Disqus
× Spirent.com uses cookies to enhance and streamline your experience. By continuing to browse our site, you are agreeing to the use of cookies.