Web Page loading pattern of Web Page Speed (Browser) Monitor

Here's a question about the Web Page Speed (Browser) that came to our support team recently. Just thought of sharing this here in the blog so that it will be useful for other users as well. The way Web Page Speed (Browser) works, it seems to show that some images and javascript get downloaded in parallel. Is this way of loading a web page the standard way followed by Internet Explorer and Firefox?

When a page is requested from the server, we open an HTTP connection for the main HTML file for that page. For every single other element on that page, images, CSS files, scripts, we will open another HTTP connection to the server and pick up objects in the order in which they are present in the web page. Many of the images and javascript in the page will get downloaded in parallel, but not all. When downloading a web page, the WPA gets the HTML output of that web page, pick up all the objects in the order in which they are present in the web page and finally execute one object after the other. Some objects are downloaded in parallel, but not all. Let's take an example here to explain this better. Consider 'test.jsp' which has 11 objects present in an order shown below:

JS, CSS, IMG, CSS, IMG, IMG, IMG, CSS, JS, JS , IMG

After collecting all these elements present in order, the WPA will execute them one after the other. The logic here is that all the images that come in a stretch and the next element after those will be done in parallel. Step 1 - Request sent to JS. Step 2 - Request sent for CSS. Step 3 - Next up is IMG and CSS. So, for IMG and CSS objects, WPA will span requests in parallel. Step 4 - Similarly requests sent to IMG, IMG, IMG, and CSS in parallel. These objects will be downloaded simultaneously. Step 5 - Separate requests sent for JS, JS, and IMG and they will be downloaded separately. Of course, this approach is the best understanding of what browsers do and it's not according to the HTTP standard.

Comments (0)