Abstract
The last three decades have seen much evolution in web and network protocols: amongst them, a transition from HTTP/1.1 to HTTP/2 and a shift from loss-based to delay-based TCP congestion control algorithms. This paper argues that these two trends come at odds with one another, ultimately hurting web performance. Using a controlled synthetic study, we show how delay-based congestion control protocols (e.g., BBR and CUBIC + Hybrid Slow Start) result in the underestimation of the available congestion window in mobile networks, and how that dramatically hampers the effectiveness of HTTP/2. To quantify the impact of such finding in the current web, we evolve the web performance toolbox in two ways. First, we develop Igor, a client-side TCP congestion control detection tool that can differentiate between loss-based and delay-based algorithms by focusing on their behavior during slow start. Second, we develop a Chromium patch which allows fine-grained control on the HTTP version to be used per domain. Using these new web performance tools, we analyze over 300 real websites and find that 67% of sites relying solely on delay-based congestion control algorithms have better performance with HTTP/1.1.
Highlights
HTTP has underpinned web page loads for the past three decades, defining both message formats as well as transmission patterns for web transfers
We find that delay-based variants (BBR, YeAH, and CUBIC + Hybrid Slow Start) favor H1 for large page sizes and cellular-like network conditions
In a series of deep-dive case studies, we deconstruct page loads to uncover how exactly this behavior plays out for real websites. We verify this behavior on several webpages, and find that, despite the negative interplay between H2 and delay-based TCP variants, there are still cases where H2 outperforms H1 due to the overheads introduced by Head of Line (HOL) blocking and setting up multiple TLS connections
Summary
HTTP has underpinned web page loads for the past three decades, defining both message formats as well as transmission patterns for web transfers. In a series of deep-dive case studies, we deconstruct page loads to uncover how exactly this behavior plays out for real websites We verify this behavior on several webpages, and find that, despite the negative interplay between H2 and delay-based TCP variants, there are still cases where H2 outperforms H1 due to the overheads introduced by Head of Line (HOL) blocking and setting up multiple TLS connections. A majority of web flows are often short-lived (page loads) and behavior during slow start dramatically affects page load performance This observation becomes even more important when we consider recent studies that show that delay-based congestion control algorithms fail to accurately estimate the actual congestion window in mobile, jittery network conditions [15, 16]. The overhead introduced by establishing TLS for each of H1’s multiple connections along with any HOL blocking that occurs can still exceed the benefits H1 achieves for delay-based variants
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
