All posts tagged ‘Amazon Silk’

File Under: Browsers

Amazon’s Silk Web Browser Adds New Twist to Old Idea

One of the headline features of Amazon’s new Kindle Fire tablet is a completely new Web browser called Silk that is designed with a “split” architecture, allowing it to offload much of the heavy lifting to Amazon’s cloud computing cluster for superior browsing performance.

When the user requests a webpage in Silk, the request will be routed to Amazon’s servers in the cloud. Amazon will load the webpage on the server side, downloading all of the necessary content elements in parallel. After downloading the content, Amazon will send the compiled page — including HTML, JavaSript, CSS, and images — back to the device as a single stream of data.
Amazon can can take advantage of its high-bandwidth connection to the Internet backbone to retrieve individual page elements faster than the user would be able to natively on the device. Web content that is already on EC2 or S3 will obviously be right at Amazon’s fingertips, further reducing the time it takes for Amazon to collect that content.

Amazon can also use its massive cloud storage infrastructure to cache enormous amounts of content that is commonly loaded by users, ensuring that it is instantly available to transmit. Amazon intends to put its machine learning expertise to use determining which pages users are likely to load so that the relevant content can be aggressively pre-cached and ready when needed.

The company’s engineers say that Silk’s robust server-side caching even obviates the need to cache anything locally on a device’s internal storage. The cached content can be pushed first while the cloud is loading all the other content elements, reducing overall page load time.

To further reduce network overhead, individual content elements like images can be compressed as appropriate for the target form factor — based on screen size and pixel depth — to further shave down their size. Amazon can use much more aggressive compression on text and other elements than has historically been possible with standard Web technologies.

The Silk browser maintains a single persistent connection to Amazon’s cloud (using Google’s fast SPDY protocol), through which requests are sent and content is received. This single connection to the Web is what lends Silk its name — as Amazon puts it, a single thread of silk is an “invisible and yet incredibly strong connection between two things.”

The ideas behind Silk are compelling, but they aren’t particularly novel. Opera has been using a similar approach for years to power the “turbo” mode of its desktop and mobile browsers. Amazon, of course, benefits from a much larger-scale cloud computing infrastructure with which to get the job done.

Amazon has also added a few unique twists of its own that will further improve the user experience. An Amazon engineer at the New York launch event told us that the split browsing infrastructure can even compile JavaScript to ARM machine code on the server side in situations where it will provide a speed boost. He also told us that Amazon will track whether users prefer the full or mobile versions of various websites so that they can predict which one is better to send to users.

We asked Amazon a few questions about the privacy implications of the split browsing model. We were told that collected usage data is anonymous and stored in aggregate, thus protecting user privacy. It’s also possible to completely turn off the split browsing mode and use Silk like a conventional Web browser.

Silk’s split design is a good fit for the Kindle Fire, a content-focused device with tight cloud integration. The feature will likely bring completely transparent performance improvements to mobile browsing. For more details, you can watch Amazon’s explanatory video.