Load Testing Blog
Showing posts tagged “page performance”
Measuring Pagespeed and Customer Engagement
Recently we started using Google Pagespeed here at webperformance.com, and one of the cool features you can do with it is split traffic, so half of your customers get no pagespeed, and the other half get pagespeed turned on. With this approach you can actually measure how pagespeed does across your entire site, as well as drill down and look at its effects on individual pages, or even divide performance by country or continent.
As you can see, the effects on our top four most popular pages were mixed. The most improved page was the list of product … Continue reading »
IE Performance Tip: avoid repeating <img> tags
This tip is more for those users who have developed a sophisticated data-crunching interface to their app. On one screen, for simplicity, the app gives a long list of data, without pagination controls being created by the user (such as a “View All” feature). To keep it pretty, each line has a visual icon or some surrounding images. Sound familiar? For example, let’s say the list is created by a testing system, indicating which tests were passed and which failed. The list may look something like:
For this example, we’ve created a few lengthy files to measure IE’s rendering performance. They … Continue reading »
Persistent Connections & Performance
Reaping performance benefits from using persistent connections on your HTTP web server is nothing new. In a recent test of a SSL site, one of our clients enabled persistent connections on their servers as part of a new deployment. The results were quite impressive: for a single user & negligible server utilization, page durations dropped from 12 seconds to only 3.5 seconds. Persistent connections can contribute to much of this duration difference by reducing the amount of time needed establish new connections. The use of SSL makes this feature even more important, as the SSL layer requires additional session negotiation. … Continue reading »
Why is my Page being Hit more than Once per View?
During testing of some sites, we found an strange case where pages seem to download themselves. What does this look like?
Note that the page gets requested, and seems to request itself while it is rendering. This has an obvious problem: the dynamic content is being requested twice as many times as necessary, causing the server to have to render it twice as often as necessary. If users turn off Load Tester’s “streaming” feature while recording, this looks a little different: the 2nd instance of the request is still made, but instead we get an error message: “Connection to server abandoned … Continue reading »
Load Testing with IE 8.0, Firefox 3.5
The performance improvements in the latest browsers (Internet Explorer 8.0 and Firefox 3.5) have been eagerly awaited for many months…though they come at a price.
New Load Tester feature – web page performance goals
Load Tester has supported the concept of a performance goal for quite a while, but the implementation has been rather limited. We have completed work on an improved version for our next release (3.6). Load Tester now provides the ability to set a performance goal for each web page and/or transaction in a testcase individually. Of course, you can still specify a global page performance goal, as you could in the past. In addition, users can specify a default page (or transaction) duration goal for an entire testcase and then override that value for each page or transaction. The performance … Continue reading »