How to stress test a network without shutting down productivity

Author: Adam Kimmel

Stress tests intentionally overload a network with extreme (though simulated) amounts of data from as many media forms as possible. This limit test shows system designers where the issues are, what the outcome of a failure at a certain point due to excess data would be and what system signals change in response to the overload. This type of stress test ultimately can help you diagnose problems that would otherwise deteriorate applications performance which in turn could negatively impact end-user experience and user productivity.

There are several approaches for how to stress test a network; although they all aid in optimizing network performance, the project return on investment (ROI) must consider the opportunity cost of shutting down the test equipment. Because this may be highly disruptive, IT leaders need to know how to stress test a network without shutting it down.

Engaging a managed services partner can offer a cost-effective alternative to in-house network monitoring and management by providing the people, processes and tools to establish and execute a successful network. This approach can also improve ROI by conducting the network monitoring in parallel, allowing your employees to work on more business critical tasks.

Here's a guide to network stress testing that will allow IT leaders to get the most out of their networks without compromising operations or productivity.

Why you need to stress test a network

The most efficient use of a network budget is to push as much traffic through it as possible without creating a backlog. To achieve this optimized state, you have to know the failure point when traffic exceeds capacity. Conducting a stress test will prevent downtime due to crashes and identify weak points in the infrastructure. This downtime could increase operating costs and raise the risk of unbudgeted, emergency maintenance in response to an equipment failure. At a high level, a stress test serves to help you understand and define the guardrails of your network capabilities in order to design information flow around it.

How to perform a network stress test

A network stress test can be manual or automated. Depending on the application of the network, a manual test might be sufficient. If there is a one-time data transfer or slug of information coming across due to an integration or software update, you can test each link independently. For a more comprehensive test, an automated tool may be a more efficient option. For either approach, here are the steps for how to stress test a network.

Ping check the network

Sending a ping across your network helps to identify bottlenecks. Pinging devices on either side of a firewall could uncover a potential issue and provide data on latency issues or data packets that do not get transferred properly.

As this is a manual test, it is time-consuming and does not allow for a more robust network check. For this approach, the best option is a customized, automated network stress test.

Create test packets you anticipate transmitting

If you know your data sources are a specific size and file type, create test packets meeting those parameters (adding some safety margin to the size). During the test, you may run various combinations of data sizes and types, so it may be best to start with all conceivable building blocks to create the most effective test.

Define the test sequence

You can evaluate user data to derive a sequence to send the data, creating a type of design of experiments to send the most common type and order combinations. An automated tool is more effective than a manual tool for this step and may help reduce the number of test iterations while ensuring technical robustness.

Check results and define a response plan

After you perform the network stress test (depending on your objective), you then have the data to act. If you are looking to expand, you will know whether you need to upgrade your systems or whether your capacity is sufficient. A data audit and subsequent network test can also indicate where high-use data packets are, so you can design in workarounds and redundancy for mission-critical transmission.

Maintaining productivity

As with any limit test, pushing a system to its extremes could create issues. Implementing a pragmatic, step by step approach will slowly push the system toward the limit. It is important to define success criteria to determine the limit and avoid potentially breaking network equipment. Implementing automatic stops and safeguards into the test will prevent crashes, ensuring operational productivity during testing. The test should also be conducted outside of core working hours to maintain sufficient control over the transmission load.

Having the knowledge about how to properly and effectively stress test your network can give you valuable insight into optimizing your system. Partnering with an industry expert for network monitoring can deliver these crucial results without taking time away from your team.

Learn more about how Verizon's Process Application Monitoring services can help you better monitor the performance of your applications.