2
BrowserMob Reduces Costs Even More with Virtual Users
No comments · Posted by Patrick Lightbody in Announcements
As of today, BrowserMob now supports both concurrent browsers and simulated virtual users. The term virtual user was adopted by early load test providers as a way to describe the technique of simulating the HTTP traffic that a real user would cause when visiting a website. While BrowserMob’s core technology sidesteps the concept of virtual users and actually drives real browsers, we felt it was important to also provide a second option for our customers.
As a service that prides itself on delivering load tests that closely emulate reality, it may at first seem odd for us to embrace the traditional concept of virtual users (VUs) that other load testing tools use. But we think it makes perfect sense when viewed in the larger context of our goal: to offer on-demand, low-cost load testing to everyone, including companies previously unable to afford it.
Reducing Costs Even More
We still think that using real browsers is the best way to generate load, as it gives the most realistic playback of load, responding to JavaScript, AJAX, and Flash in realtime. And we’re proud of the fact that we can offer real browsers at a fraction of the price of our competitors offer virtual users.
In fact, the only way to reduce our pricing further was to drop the overhead a real browser introduces. And that’s just what we did. For as little as 10 cents per VU/hour, you can schedule a test that uses VUs and generates massive amounts of load at an extremely low price. For example, 2,000 VUs can now run for an hour, generating over a 750mbps in data throughput, with a cost of only $200.
Visit our pricing page to learn more about this new offering and see how little it costs compared to the alternatives out there. Alternatively, if you have any questions, such as when to use virtual users and when to use real browsers, don’t hesitate to contact us.
So Which is Better?
This is a complex topic and there is no good answer. In a few days we’ll post a follow-up that discusses when to use real browsers vs. virtual users. In general, it comes down to a trade-off between price and fidelity, but usually at least some use cases can be simulated with virtual users and still achieve your requirements for load. Look out for our follow-up discussion for more on this.
No tags
