G-WAN Rules in Static Benchmark Test Battery

Over at his blog Spoot!, Nicolas Bonvin recently posted two summaries of the great work he's done benchmarking how well various open-source and free Web servers dish up static content under high loads. Bonvin, a PhD. student at the …cole polytechnique fÈdÈrale de Lausanne (EPFL in Switzerland, specializes in high-volume distributed data systems, and brings considerable expertise and real-world experience to bear in designing his tests.

First Round of Performance Testing

In his first post, Bonvin laid out the evidence he'd accumulated by running benchmark tests against six Web servers: Apache MPM-worker, Apache MPM-event, Nginx, Varnish, G-WAN, and Cherokee, all running on a 64-bit Ubuntu build. (All Web servers used, save for G-WAN, were 64-bit.) This first set of benchmarks was run without any server optimization; each server was deployed with its default settings. Bonvin measured minimum, maximum, and average requests per second (RPS) for each server. All tests were performed locally, eliminating network latency from the equation.

On this initial test battery, G-WAN was the clear winner on every conceivable benchmark, with Cherokee placing second, Nginx and Varnish close to tied, and both strains of Apache coming in dead last. As Bonvin notes, it wasn't even close. G-WAN, a small Web server built for high performance, completed 2.25 more requests per second than Cherokee (its closest competitor), and served a whopping 9 to 13.5 more requests per second than the two versions of Apache.

I read an article today on the E-commerce Times site Web Performance Metrics That Matter. It was the first result in a Google search on "web performance metrics", and the title sounded like a perfect hit. My intent was to see what other people think are the best performance metrics.

Here is a paragraph that really surprised me:

An article caught my attention because I follow the author on Twitter. Two days ago, Jeffery Way published 15+ Tips to Speed Up Your Website, and Optimize Your Code!

SQL Server performance tuning is usually left until problems arise. Most of the time developers and even database administrators only focus on the tuning elements of their MSSQL servers when performance starts to become slow and adversely affects users. Although the typical reason for slow issues is configuration or development issues (poor stored procedure design), occasionally hardware can be an issue.

This is quite a list of open source performance testing tools. While there are definitely some excellent products on this list, there are also many (perhaps most) which fall into the category of poor to mediocre. As is the case with any open source category, these were developed by programmers who generally develop something for their own use, without documentation and very specific application. The most popular according to this page are:

• Apache JMeter
• OpenSTA
• WebLOAD

If your web server is starting to lose performance and resources are lacking, it may be time to implement a web farm. A web farm will pool sources and allow your web server to share its resources with another high powered machine. It can greatly increase web form performance and rid some of the end user frustration of timeouts and slow responses.

LoadRunner (now called HP LoadRunner, since HP bought Mercury Interactive) is one of the most popular software automation and testing programs on the market. It is a comprehensive testing program that gives webmasters and developers the ability to test the functionality and load managing ability of their website. There are many reasons why LoadRunner is so popular and why it is a good choice as a software testing tool for your website.