How many customers can you handle at once? How many orders could you process? What changes should you make to serve customers more quickly? Will your infrastructure support a sales event—or crack under pressure? Load testing gives you concrete answers to all of these questions and more.

Multi-user load

We often think about one user when considering whether a site is fast or slow. In reality, you have multiple users navigating your site at once. The more users you have, the slower your site’s response time. Simulating multi-user scenarios across numerous threads and different transactions gives you an accurate measurement of your site’s current performance.

Analyzing traffic flow, we identify which resources affect rendering time. For example, international users have to go through multiple hubs, which results in latency and a slower response time. We review your firewall, web server, application layers, and third parties, like Google search platforms—all of which add up to impact response and user experience.

Common load tests include:

• Load test: Establish your site’s threshold to ensure you never overstress your system and risk a crash. For example, setting the desired response time, we might run 5000 users over two hours to establish an impact on response time.
• Endurance test: An endurance test is conducted over longer durations to mimic big sales events like Black Friday to pick up on memory leaks that slow response time.
• Stress test: Where is your system most vulnerable right now? Is the traffic load balanced between your servers? A stress test reveals how much of a buffer you have and whether you can support more traffic with confidence.
• Spike test: A sudden flood of traffic from advertising or flash sales can cause systems to struggle or crash. A spike test shows whether you can handle traffic steadily and, if not, what you need to do to fix it.

Using various tools, load testing measures everything from memory usage and response time to active sessions, and network bandwidth.

Build your traffic map

We also help you understand your traffic by reviewing historical data and aligning your business and user needs. For example, if you have a trading app, you need to allow for high traffic on a Monday morning, while users of a news app may check in over the day. You’ll also need to consider the network conditions for users in different locations.

Using scripting and various toolsets, we model load traffic, simulating different user groups across different networks with a range of scenarios and actions from search to resetting a password.