This new code makes
websites load 34% faster in any browser
Just what we’ve always wanted.
If website loading times can’t keep up with
your browsing speed like your typing can’t keep up with your thoughts, hope is
on the horizon, thanks to a new code called Polaris that reportedly makes
websites load 34 percent faster in any browser.
How, I hear you ask? Developed by a team at
MIT, the code has been programmed to gather up and evaluate all the background
files of a particular website, including image files, HTML files, and
Javascript source code, way more efficiently than current systems.
"As pages increase in complexity, they
often require multiple trips that create delays that really add up," says one
of the team, Ravi Netravali. "Our approach minimises the number
of round trips so that we can substantially speed up a page’s load-time."
You’re not the only one who gets frustrated by
a site that doesn’t instantly load - pretty much everyone feels your pain, and
are quite happy to drop a page like a sack of oranges if it’s going to keep
them waiting. According to Amazon, every 100-millisecond
delay in website loading time cuts their profits by 1 percent.
If you’re not familiar with the nuts and bolts
of getting a website to display on a browser, it goes like this. To load a
page, your browser has to connect to the network and fetch 'objects' such as
HTML files, image files, and JavaScript source code. Each object has to be
evaluated before it can be added to the display.
The problem is that each of these objects is
likely to be dependent on another object - called
dependencies - so every time the browser wants to evaluate one
object, it has to go and fetch its dependencies, which will also have their own
dependencies, and so on. "As an example, a browser might have to execute a
file’s JavaScript code in order to discover more images to fetch and
render," the team explains.
That’s not so bad, but where things get slowed
down is when browsers can’t actually see all the dependencies, so end up making
multiple trips through the network to fetch and evaluate everything.
What the Polaris code does to address this
problem is automatically track every single interaction between objects and
their dependencies on a particular site, to creates a 'dependency graph' that
allows the browser to make more efficient trips through the network.
While this isn’t the first dependency-tracker
to be developed, the team says this is the only one that’s able to capture the
most subtle dependencies that exist on a page, which allows it to achieve such
impressive speeds. Another benefit is that it can be written in JavaScript,
which means it can work with any website.
The team hopes that the system will be
integrated with browsers in the future, so every page you load on a particular
browser will appear that much faster.
According to Jamie Condliffe at Gizmodo,
Polaris has so far been tested on 200 different websites, including ESPN,
Weather.com, and Wikipedia, and was able to load web pages 34 percent faster,
on average.
Their research has not yet been published or
peer-reviewed, so keep that in mind, but the results of their work will be
presented later this week at the USENIX Symposium on Networked Systems Design
and Implementation in California, which means they’re ready for
it to be scrutinised for publication.
"Tracking fine-grained dependencies has
the potential to greatly reduce page-load times, especially for low-bandwidth
or high-latency connections," Mark Marron, a senior research software
development engineer at Microsoft, who was not involved in the research, told Gizmodo.
"On top of that, the availability of
detailed dependence information has a wide range of possible applications, such
as tracking the source statement of an unexpected value that led to a crash at
runtime."
We are so ready for this. And if they can figure out how to
speed up gif loading times on mobile, our lives would be complete.