Stress-test an API endpoint before launch to measure how many requests per second it can handle.
Compare server performance before and after an optimization by running plow with identical settings.
Simulate concurrent users hitting your web app and watch live latency charts update in a browser dashboard.
Find the breaking point of a server by gradually increasing concurrency until error rates climb.
Plow is a command-line tool that stress-tests web servers by sending a large number of HTTP requests and measuring how fast the server responds. If you have a website or API and want to know how it holds up under heavy traffic, plow simulates that load and reports the results in real time. It shows you things like how many requests per second the server handled, how long each request took, and what percentage of requests succeeded or failed. While the test runs, plow displays live statistics directly in your terminal and also opens a browser-based dashboard where you can see charts updating as the test progresses. This makes it easier to watch performance change over time rather than waiting until everything is done to see a summary. You control the test with a few simple options. You can set how many connections to open at once (more connections means more simulated users), how long to run the test, or how many total requests to send. You can also configure the HTTP method (GET, POST, etc.), add custom headers, send a request body from a file, and set timeouts. There is support for HTTPS, proxy servers, and even Unix sockets. Installing plow is straightforward if you have Go or Homebrew on your machine, and there is also a Docker image if you prefer containers. Once installed, you point it at a URL and add whatever options you need. The output includes a breakdown of latency at various percentiles, which tells you not just the average response time but also how the slowest requests compare to the typical ones. Plow is suited for developers and system operators who want a quick way to check a server's performance before or after making changes. It does not require a complex setup and produces detailed output without needing a separate monitoring system running alongside it.
← six-ddc on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.