Load testing is one of the tools we leverage regularly at Tag1. It can help prevent website outages, stress test code changes, and identify bottlenecks. The ability to run the same test repeatedly gives critical insight into the impact of changes to the code and/or systems. Often -- as part of our engagements with clients -- we will write a load test that can be leveraged and re-used by the client into the future. In some cases, our clients have extensive infrastructures and multi-layered caches, including CDNs, that also need to be load tested. In these instances, it can take a considerable amount of computing power to generate sufficient load to apply stress and identify bottlenecks. This ultimately led us to write and open source Goose, a new and powerful load testing tool. # Locust Is Fantastic, Python Is Not Ideal Discovering Locust was a breath of fresh air, solving so many of the frustrations we used to have when load testing with jMeter. Instead of working with a clunky UI to build sprawling, bloated JMX configuration files, Locust allows the writing of truly flexible test plans in pure Python. This allows code to easily be re-used between projects, and...

Read more
Jeremy
Tue, 05/12/2020 - 04:54