As webpages are increasingly being designed to show more content, slower loading times have frustrated Internet users. A new system, however, is looking to end this worldwide problem.
Researchers from MIT's Computer Science and Artificial Intelligence Laboratory, or CSAIL, and Harvard University have been able to develop a technology that will load webpages faster by 34 percent.
The system, named Polaris, is able to figure out how best to overlap downloading the objects of a page, which would lead to less time taken to load webpages.
"It can take up to 100 milliseconds each time a browser has to cross a mobile network to fetch a piece of data," said Ravi Netravali, an MIT Ph.D., student who will be presenting a paper on the technology this week at the USENIX Symposium on Networked Systems Design and Implementation.
Netravali explained that more complex webpages often needs more trips to download needed data, slowing down the webpage significantly. Polaris, simply out, reduces the number of trips needed to speed up the loading times.
The co-authors of the paper are MIT graduate student Ameesh Goyal, MIT professor Hari Balakrishnan and Harvard professor James Mickens.
Mickens, who began working on the project as a visiting professor at MIT in 2014, offered the analogy of Internet users as travelling businessmen. When visiting a city, businessmen could then find out that there are other cities that they can visit as well and just then begin to plan where to go next. However, if businessmen were given all the possible cities that they can visit before even starting their trip, then they would be able to plan the most efficient itinerary possible and follow it throughout their travel.
Polaris, in essence, functions as the provider of the list of all the possible cities in the above analogy. When loading webpages, certain objects being fetched sometimes lead to other objects that need to be downloaded. Polaris tracks all these dependencies at once, which leads to faster loading times.
According to researchers, in addition to larger and more complex webpages, the Polaris system is also geared for mobile use due to the higher risk of delays on these networks compared to wired ones.
