As we come to rely more and more on the internet for everything from social interactions to the gathering and storing of information and data, webpages—those hypertext documents that give meaning to the chaos—are growing steadily more complex. This progress comes at a price, though, as pages are required to fetch multiple dependent support objects as they load. This can result in serious performance lag during the loading process.
Fortunately, researchers at MIT are hard at work on a solution. They’ve developed Polaris, a system that tracks all dependencies (which can number in the thousands for a single page) and logs them in a dependency graph that allows a page to fetch elements more efficiently. On average, sites running Polaris load 34% faster than those that don’t utilize it.
“It can take up to 100 milliseconds each time a browser has to cross a mobile network to fetch a piece of data,” PhD student Ravi Netravali said in a press release. As one of the researchers who worked on the project, Ravi recently presented his team’s findings at USENIX Symposium on Networked Systems Design and Implementation (NSDI ’16). “As pages increase in complexity, they often require multiple trips that create delays that really add up. Our approach minimizes the number of round trips so that we can substantially speed up a page’s load-time,” he said.
Although early tests are encouraging, it’s important to bear in mind that Polaris is not a panacea.
“The next evolution of websites is already well underway and mainstreamed, and good developers bundle nearly all of their dependencies,” says Chris Chapman, a Course Director in Full Sail’s Web Design & Development program. “On the surface, this MIT project seems to be an easy plug and play solution, but there are other bits to be concerned about. Their “client side bundling” technique may have its greatest uses in bundling ad tracker scripts in with your main site, making it harder to ad block or ghost the trackers.”
Still, researchers are optimistic about Polaris’ applications. ““We are hopeful that the system will eventually be integrated into the browser,” says Netravali. “Doing so will enable additional optimizations that can further accelerate page loads.”