Personalized web experiences can have a downside: end users are often left waiting while back end web servers generate and then deliver their individualized HTML pages from across the Internet. Historically it has been difficult or impossible to improve the performance of these pages without deploying large numbers of additional servers on the back end. This is because of the way static and dynamic content are combined and then delivered together as a data package to the end user.
How HTML Streaming Works
Instart Logic’s HTML Streaming solution relies on several of the underlying capabilities of the endpoint-aware application delivery platform to improve dynamic page performance. Using machine learning, our solution can identify the static and dynamic elements of these pages and then transparently cache them in our CDN caching layer. Thereafter, when an end user accesses one of the personalized web pages, our platform responds immediately with the cached static elements and the Instart Logic Nanovisor.
In parallel, the Instart Logic platform makes a request to your origin web server for the dynamic content required to build out the end user’s web page. Once your server responds with the content, the dynamic portions are delivered to the end user over our accelerated network and invisibly stitched into the page in the browser dynamically using the endpoint-aware capabilities of our Nanovisor. As a result, the end user is able to view and interact with the page much more quickly than is possible when using a traditional delivery service.