Researchers at MIT have found a way to make webpages load up to a third faster, promising an end to frustrating long waits.
Given low retention times of webpages, organizations lose prospective customers and revenue if their websites take time to load. Additionally, the costs for mobile user are higher as delayed loading increases data usage.
"It can take up to 100 milliseconds each time a browser has to cross a mobile network to fetch a piece of data," said the study's first author Ravi Netravali.
Netravali, with others from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) and Harvard, detailed in a paper the functioning of Polaris, the idea that exploits real dependencies between objects on a webpage to make it load faster.
It highlights the drawbacks of traditional loading including mapping lexical dependencies, which are not actual dependencies between objects.
"The objects in a web page can interact in complex and subtle ways; however, those subtle interactions are only partially captured by lexical relationships between HTML tags," the paper stated.
What that means is when a browser loads a page, the HTML file on the server lists objects in a specific order and tells the browser to fetch objects in that order. This way of functioning can lead to delay due to ways in which complex websites, for instance, those with media files and JavaScript, can function.
Polaris relies on building a graph based on actual dependencies. It provides a browser a list of all objects and the quickest way to get them. Co-author Harvard professor James Mickens drew parallels with an analogy of a businessman seeking to visit a certain number of cities for work.
"For a Web browser, loading all of a page's objects is like visiting all of the cities," Mickens said. "Polaris effectively gives you a list of all the cities before your trip actually begins. It's what allows the browser to load a webpage more quickly."
To test its efficiency, researchers used Polaris to load 200 top websites ranked from Alexia's top-500 list. They found it hastened median loading times by 34 percent.
