Can SPDY Really Make the Web Faster?
📜 Abstract
HTTP is a successful Internet technology on top of which a lot of the web resides. However, limitations with its current specification have encouraged some to look for the next generation of HTTP. In SPDY, Google has come up with such a proposal that has growing community acceptance, especially after being adopted by the IETF HTTPbis-WG as the basis for HTTP/2.0. SPDY has the potential to greatly improve web experience with little deployment overhead, but we still lack an understanding of its true potential in different environments. This paper offers a comprehensive evaluation of SPDY’s performance using extensive experiments. We identify the impact of network characteristics and website infrastructure on SPDY’s potential page loading benefits, finding that these factors are decisive for an optimal SPDY deployment strategy. Through exploring such key aspects that affect SPDY, and accordingly HTTP/2.0, we feed into the wider debate regarding the impact of future protocols.
✨ Summary
Paper summary
The paper evaluates SPDY, an application-layer protocol designed to improve web page loading through multiplexing, header compression, persistent encrypted connections, server push or hints, and content prioritization. The authors combine measurements of live websites with controlled experiments using an emulated network testbed. Their live crawl found SPDY support on 208 of the Alexa top 10,000 websites in October 2012 and 271 in April 2013. Results varied substantially: SPDY improved page load times for some sites but degraded them for others. In the authors’ measurements, average time-on-wire reductions included 7% for Facebook, 4.7% for YouTube, 9.7% for Youm7, and 10.6% for Twitter, while increases were observed for Blogspot, WordPress, and Google’s search homepage. (yelkhatib.github.io)
The controlled experiments attribute this variation primarily to network and website-infrastructure characteristics. SPDY generally performs better on high-latency, low-bandwidth paths because multiplexing reduces connection-establishment overhead and avoids some congestion caused by parallel HTTPS connections. Its benefits diminish as bandwidth increases, and packet loss can be especially harmful because SPDY relies on a long-lived multiplexed TCP connection. Website sharding also reduces SPDY’s advantage by distributing resources across multiple connections and domains, thereby limiting multiplexing. The paper reports that the degree of multiplexing required to outperform HTTPS grows approximately with page complexity, although the authors explicitly describe this relationship as non-robust. (yelkhatib.github.io)
Influence
The paper became a cited empirical reference in subsequent evaluations of SPDY and HTTP/2. Later comparative studies, including work explicitly asking whether HTTP/2 is faster than HTTP/1.1, cite it alongside other early protocol-performance measurements. (tma.ifip.org) The authors’ central finding—that protocol performance depends strongly on latency, loss, bandwidth, page complexity, and deployment architecture—was therefore carried into the broader research literature on HTTP/2 and next-generation web transport. The available sources establish academic citation and reuse, but do not document a specific industrial deployment directly attributable to this paper. (research.lancaster-university.uk)