Advancements in technology have allowed us to load websites at breakneck speeds compared to what was possible just a couple of decades ago. As these technologies have evolved, so too have the complexity of the websites we call upon. It's sort of like leveling up your character in a video game only to be met by tougher enemies.
Researchers at MIT's Computer Science and Artificial Intelligence Lab (CSAIL) and Harvard have been working on the matter since 2014 and have come up with a solution they say can improve page load times by an average of 34 percent. The framework is called Polaris and here's how it works (in Layman's terms).
When you navigate to a website, a lot of back-and-forth communication is happening in the background to fetch the site. This travel takes time and the more trips that are needed, the more time is wasted which ultimately results in a slower loading website for the end-user. Polaris aims to minimize the number of round trips and ultimately speed up page load times.
Harvard professor James Mickens likens it to a travelling businessperson. When the person visits one city, they may sometimes discover more cities they need to visit before returning home. If the businessperson had the entire list of cities ahead of time, they could then plan the fastest route possible. Without the list, they're left discovering new cities as they go, resulting in unnecessary zigzagging between far-away cities.
For a web browser, he says, loading all of a page's objects is like visiting all of the cities. Polaris effectively provides a list of all the cities before the trip begins which is what allows it to speed up page load times.
Because it is written in JavaScript, Polaris can be used with virtually any browser and deployed on a site-by-site basis. The end game, however, is to eventually have the functionality baked into web browsers as that would allow for additional optimizations for even more speed.
Image via Zurijeta, Shutterstock