Pylot for unit testing, monitoring and load testing

Pylot, written by Corey Goldberg, is a simple open source web performance tool written in Python. It is easy to setup and get basic benchmarking metrics about your web services. You can also use Pylot for executing unit tests on REST-ful web applications.

I recently had to plan the roll-out of a new ad/image server on a tight timeline. The image server uses OpenX and I wasn’t sure what to expect in terms of performance, I was new to OpenX, the virtual private server that was to host the application hadn’t been benchmarked and I needed some basic benchmarks to ensure that the rollout would be successful for the limited load scenarios we were expecting.

Within 20 minutes of downloading Pylot I had my first results. A few Apache configuration changes later and I was comfortable that the new image server would easily handle our traffic needs.
Since Pylot supports checking the Response body of an HTTP request, it is straightforward to set Pylot up to do unit tests. Because Pylot can be run in command line mode and is easy to extend its a good starting point for rolling your own automated regression analysis and system monitoring.

Links

Advertisements

4 thoughts on “Pylot for unit testing, monitoring and load testing

  1. Hi George, I’ll post some details late this week. The basic benchmarking I did showed ~100 req/sec with a median of 200msec response time while the machine was getting hammered. The Apache tweaks were minor and our OpenX server only has one client, our web site. The client site has an average of 3 ads per page with some of them being served from double click.

    We upgraded to 2.6 last week, so I’ll re-run the benchmarks and post some details in a few days. –Sean

  2. @Neetha: You don’t have to run a pylot “installer.” unzip and then run it from the command line (DOS prompt). “cd” into the directory / folder (ex. pylot_1.26) that contains run.py and then use python to start pylot up “python run.py -g”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s