Modeling HTTP/2 Speed from HTTP/1 Traces

@inproceedings{Zarifis2016ModelingHS,
  title={Modeling HTTP/2 Speed from HTTP/1 Traces},
  author={Kyriakos Zarifis and Mark Holland and Manish Jain and Ethan Katz-Bassett and Ramesh Govindan},
  booktitle={PAM},
  year={2016}
}
With the standardization of HTTP/2, content providerswant to understand the benefits and pitfalls of transitioning to the new standard. Using a large dataset of HTTP/1.1 resource timing data from production traffic on Akamai’s CDN, and a model of HTTP/2 behavior, we obtain the distribution of performance differences between the protocol versions for nearly 280,000 downloads. We find that HTTP/2 provides significant performanceimprovements in the tail, and, for websites for which HTTP/2 does not… 

SMig: Stream Migration Extension for HTTP/2

It is demonstrated through real implementation that SMig can bring substantial performance improvement under certain common usage scenarios (e.g., up to 99% of download time reduction for small delay-sensitive objects when a concurrent large download is present).

Are HTTP/2 Servers Ready Yet?

This study conducts the first systematic investigation by inspecting six popular implementations of HTTP/2 servers and measuring the top 1 million Alexa web sites, proposing new methods and developing a tool named H2Scope to assess the new features in those servers.

Making Effective Use of HTTP / 2 Server Push in Content Delivery Networks

This work investigates the factors that should be considered when implementing a Push policy and combines them to propose an efficient Push policy geared for Content Delivery Networks, and indicates that recommended Push configurations yield performance benefits of up to 27% for the pages tested.

Is There a Case for Parallel Connections with Modern Web Protocols?

By opening only two connections to load a page with H2-Parallel, the page load time can be reduced substantially in adverse network conditions, and the use of a single connection sometimes impairs application performance.

Impact of Implementing HTTP/2 in Web Services

The research try to figure out the pros and cons of the this new protocol version from different aspect specially security issues.

Cloud autoscaling for HTTP/2 workloads

A state-of-the-art reactive autoscaling policy is applied to assess the effects of deploying the HTTP/2 server push mechanism in cloud environments and it is shown that the autoscaled mechanism is beneficial for web servers even though pushing a large number of objects might lead to server overload.

Push or Request: An Investigation of HTTP/2 Server Push for Improving Mobile Performance

This paper first examines whether pushing all content or just the CSS and Javascript files performs better, and finds the former leads to much better web performance, and there are a number of challenges in achieving the full benefits of Server Push.

Mind theDelay: TheAdverse Effects of Delay-Based TCP onHTTP

Igor is developed, 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, and a Chromium patch which allows fine-grained control on the HTTPversion to beused per domain.

Towards faster web page loads over multiple network paths

This work determines whether using MPTCP can reduce web page load times, and presents enhancements to multipath web browsing that allow it to perform as well as or better than regular TCP on the best path.

Influence of HTTP/2 on User Experience

Findings show that users report marginal differences, with 22%, 52%, 26% of HTTP/2 MOS being better, identical, or worse than HTTP/1, respectively, which shows the complexity of understanding the web experience and the need to involve actual users in the quality assessment process.

References

SHOWING 1-10 OF 14 REFERENCES

How Speedy is SPDY?

A systematic study of Web page load time (PLT) under SPDY and compare it to HTTP finds that SPDY provides a significant improvement over HTTP when the authors ignore dependencies in the page load process and the effects of browser computation.

Towards a SPDY'ier mobile web?

A detailed analysis of the performance of HTTP and SPDY, how they interact with the various layers, and their implications on web design concludes that a viable solution has to account for these unique cross-layer dependencies to achieve improved performance over cellular networks.

To HTTP/2, or Not To HTTP/2, That Is The Question

The initial findings from a 6 month measurement campaign find 13,000 websites reporting HTTP/2 support, but only 600, mostly hosted by Google and Twitter, actually serving content, and no significant benefits are found under stable network conditions.

DASH fast start using HTTP/2

This paper focuses on using the HTTP/2 protocol and the server push feature to reduce the start-up delay in a DASH streaming session and proposes a new approach for video adaptation, which consists in estimating the bandwidth, using WebSocket (WS) overHTTP/2, and in making partial adaptation on the server side.

Can SPDY really make the web faster?

The impact of network characteristics and website infrastructure on SPDY's potential page loading benefits are identified, finding that these factors are decisive for an optimal SPDY deployment strategy and feed into the wider debate regarding the impact of future protocols.

Demystifying Page Load Performance with WProf

This work presents WProf, a lightweight in-browser profiler that produces a detailed dependency graph of the activities that make up a page load, and finds that computation is a significant factor that makes up as much as 35% of the critical path, and that synchronous JavaScript plays a significant role in page load time by blocking HTML parsing.

Reducing web latency: the virtue of gentle aggression

This paper presents the design of novel loss recovery mechanisms for TCP that judiciously use redundant transmissions to minimize timeout-driven recovery and are compatible both with middleboxes and with TCP's existing congestion control and loss recovery.

CUBIC: a new TCP-friendly high-speed TCP variant

The CUBIC protocol modifies the linear window growth function of existing TCP standards to be a cubic function in order to improve the scalability of TCP over fast and long distance networks.

How Fast is Your Web Site?

The overwhelming evidence indicates that a Web site’s performance (speed) correlates directly to its success, across industries and business metrics. With such a clear correlation (and even proven