The State of Web Readiness 2012: overconfidence behind world wide performance problems.
The report analyzes data from 8,522 load tests executed in 132 different countries. The tested sites range from government organizations that have to be prepared for emergencies and e-retailers who need to cope with seasonal peaks, to media companies and smaller sites that may need to handle a "Slashdot effect" when a larger website links to them.
"Site owners tend to be overly confident that their site can handle expected traffic levels, which means the risk of crashes is probably higher than most of us think. In light of this, most organizations have had a web site for the past 15 years or so, and since the web is a truly vital part of modern business, it is remarkable that website performance is not given higher priority", said Ragnar Lonn, CEO and founder of Load Impact.
E-commerce sites distinguish themselves from other types of sites in the survey. Among the top 100 sites surveyed, the largest e-retailers are on average investing three times as much on load testing compared to media sites, event sites and corporate/organization sites.
"The report shows that e-commerce sites are most likely better prepared to handle heavy traffic during visitor peaks. They lose money immediately when the site becomes slower due to an increase in the number of visitors, so it is natural that they see a direct benefit of staying informed regarding the performance of their site. It allows them to optimize and plan ahead so they can cope with Christmas sales or other sales peaks", said Ragnar Lonn. "It is a bit surprising that ad-funded sites, like many media sites, don't seem to be equally concerned."
When comparing different regions, North American companies invest the most in load testing, followed by South America and Europe. Asia/Pacific spends the least.
Facts about the report
The report "The State of Web Readiness 2012" measures load tests performed by companies and organizations all over the world. The overconfidence factor in the report indicates the number of concurrent users that effectively cause the server response-time to double from its initial level, compared to the maximum number of concurrent users the load test was configured to simulate. So an overconfidence factor of 3.4, as in this case, means the average load test was configured to simulate 3.4 times as great a load on the site as was required to achieve a 100% increase in load time.
Number of tests: 8,522, Number of websites: 3,968
Nationalities: 132 Download the report from:
Email The Editor With All Your Latest News, New Product Releases, Reports and Technical White Papers For Worldwide Coverage. email@example.com
|Printer friendly Cite/link Email Feedback|
|Publication:||Database and Network Journal|
|Date:||Aug 1, 2012|
|Previous Article:||IPv6: a primer: an introduction to IPv6.|
|Next Article:||The Big Data Refinery: distilling intelligence from Big Data.: In a world where data volumes are literally exploding in size, the winners will not...|