Machine learning speeds up web application load times
The increase in web application use means that we're loading more code than ever into our browsers. According to httparchive, the amount of JavaScript used by the top 100 websites has almost tripled in the last three years whilst HTML has almost doubled. This has performance implications because larger page-sizes mean end users have to wait significantly longer for web pages to load.
California-based cloud delivery service Instart Logic is hoping to improve things with the launch of SmartSequence, a new technology that employs machine learning to optimize how HTML and JavaScript code is loaded in browsers.
SmartSequence uses a cloud-client architecture and machine learning technology to accelerate JavaScript and HTML, enabling browsers to load only the code that's needed, thus reducing download size and speeding up performance.
It learns which code is most frequently used by watching user behavior. For JavaScript Streaming, it then downloads only what a web page or web application needs, while ignoring unused code. For HTML Streaming, it determines which portions of dynamic HTML are used across all users, and sends that in advance. SmartSequence can also learn which sections of the page <HEAD> are dynamic as opposed to static and patch dynamic data in as needed.
"With SmartSequence, we are taking a revolutionary approach to a critical web performance challenge that is created by dynamic HTML and heavy use of JavaScript," says Peter Blum, vice president of product management at Instart Logic. "Our team of computer scientists and algorithmic experts have developed new machine learning techniques that enable us to intelligently stream JavaScript and HTML code, greatly improving page load times and customer experience".
SmartSequence is available now as part of Instart Logic's software-defined application delivery platform and you can read more about it on the company's blog.
Image Credit: isak55 / Shutterstock