A case for faster mobile web in cellular IPv6 networks

Goel, Utkarsh; Steiner, Moritz; Wittie, Mike P.; Flack, Martin; Ludin, Stephen

doi:10.1145/2973750.2973771

Cited by 13 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Few comparative studies have examined different aspects and functionality of IPv4 and IPv6 in different underlying networks. In [12], the authors provide an experimental setup to compare Web performance on IPv4 and IPv6 networks in four leading cellular carriers, including T-Mobile, Verizon, AT&T, and Sprint. Several webpages are loaded over IPv4 and IPv6 to compare round trip time, latency, and page load time parameters.…”

Section: Related Workmentioning

confidence: 99%

IPv6 Transition Measurements in LTE and VHT Wi-Fi Mobile Networks

Mina

2019

IEEE Access

View full text Add to dashboard Cite

Full digital connectivity of the Internet of Things (IoT) devices demands several requirements including high-speed networks and a large number of IP addresses. The long term evolution (LTE) and very high throughput (VHT) 802.11ac networks are among the alternatives that can fulfill the speed requirements. To provide a large number of IP addresses, in addition to IPv4, LTE and 802.11ac also support IPv6. However, while the full digital connectivity cannot be fulfilled by IPv4 due to its limited address space and failure to support the scalability of the IoT applications, another major problem is that the potential benefits of IPv6 for LTE and 802.11ac mobile networks are completely ambiguous. The issue is further increased along with the design complexities inherent in LTE and 802.11ac infrastructures. Therefore, there are increasing concerns for cellular carriers and mobile service providers regarding migration to IPv6-only and whether the users in LTE-IPv6-only and 802.11ac-IPv6-only networks can achieve better performance than IPv4. To address the challenges associated with deploying IPv6-only in LTE and 802.11ac networks and quantify the performance, this work proposes a model. The model consists of a simulation environment with four distinct networks: LTE-IPv6-only, LTE-IPv4-only, 802.11ac-IPv6-only, and 802.11ac-IPv4-only. The model is further extended by setting up a real-world testbed environment to include four networks for replication of those simulations. To assure the most comprehensive environmental evaluation of the model, 128 distinct scenarios are developed and implemented, and the results are obtained in terms of quality of service parameters. The testbed results are compared to those of simulations to precisely assess the model. INDEX TERMS LTE, VHT, IPv6, QoS, Internet-of-Things (IoT).

show abstract

Section: Related Workmentioning

confidence: 99%

IPv6 Transition Measurements in LTE and VHT Wi-Fi Mobile Networks

Mina

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To understand the performance of Web content delivery on production traffic, Akamai utilizes Real User Measurement (RUM) [8] system that uses the Web browser exposed Navigation Timing API [9] to capture several performance metrics pertaining to webpage load. Specifically, when clients request the base page HTML, Akamai's RUM system injects custom JavaScript that runs on the client's browser and uses the Navigation Timing API to record the time taken to perform DNS lookup of the base page domain name, time to establish TCP connection, and the overall PLT, among many other metrics [29], [30]. While RUM allows for large scale measurements across many mobile networks and client devices, it does not record transport layer metrics, such as packet loss rates and timestamps of loss occurrence during the lifetime of a TCP connection.…”

Section: Why a New Cellular Emulation Testbed?mentioning

confidence: 99%

“…In this section, we emulate the five network qualities described in Section VI-A to compare PLTs over h2 and h1. Since there is no standard definition of PLT, similarly to many other studies [25], [29], [34], [41], [50], [53], [55], we estimate the PLT as the time from when the user enters the URL in the Web browser until the browser fires the JavaScript's OnLoad event. For measuring the PLT, we use the client to load several webpages synthesized from HTTP Archive [15].…”

Section: Comparing Web Performance Of H2 and H1mentioning

confidence: 99%

“…The use of domain-sharding incurs additional DNS lookups, which may take several hundred milliseconds in cellular networks [29], [30], [47], and can potentially increase the overall PLT. However, several Web optimization techniques, such as DNS Pre-Resolve [24] and DNS PiggyBacking [48], can eliminate this additional lookup latency by providing hints in the base page HTML, such that the Web browser resolves the domain names much before the resolution is needed.…”

Section: Challenges In Validating the Emulatormentioning

confidence: 99%

See 1 more Smart Citation

Domain-Sharding for Faster HTTP/2 in Lossy Cellular Networks

Goel,

Steiner,

Wittie

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

HTTP/2 (h2) is a new standard for Web communications that already delivers a large share of Web traffic. Unlike HTTP/1, h2 uses only one underlying TCP connection. In a cellular network with high loss and sudden spikes in latency, which the TCP stack might interpret as loss, using a single TCP connection can negatively impact Web performance. In this paper, we perform an extensive analysis of real world cellular network traffic and design a testbed to emulate loss characteristics in cellular networks. We use the emulated cellular network to measure h2 performance in comparison to HTTP/1.1, for webpages synthesized from HTTP Archive repository data.Our results show that, in lossy conditions, h2 achieves faster page load times (PLTs) for webpages with small objects. For webpages with large objects, h2 degrades the PLT. We devise a new domain-sharding technique that isolates large and small object downloads on separate connections. Using sharding, we show that under lossy cellular conditions, h2 over multiple connections improves the PLT compared to h2 with one connection and HTTP/1.1 with six connections. Finally, we recommend content providers and content delivery networks to apply h2-aware domain-sharding on webpages currently served over h2 for improved mobile Web performance.

show abstract

“…As websites continue to grow in complexity, the key to improve website responsiveness is to build faster networks, optimize website content, and produce mobile devices with faster CPUs. While many networks have already begun to deploy new infrastructure and support faster communication protocols [28,34,38,43], and most websites already employ a suite of website optimization techniques [35], the mobile device CPU remains a limiting factor to mobile Web performance [50]. Unlike desktop and laptop CPUs, Mobile CPUs are designed for power efficiency.…”

Section: Introductionmentioning

confidence: 99%

Web Performance with Android's Battery-Saver Mode

Goel,

Ludin,

Steiner

2020

Preprint

Self Cite

View full text Add to dashboard Cite

A Web browser utilizes a device's CPU to parse HTML, build a Document Object Model, a Cascading Style Sheets Object Model, and render trees, and parse, compile, and execute computationally-heavy JavaScript. A powerful CPU is required to perform these tasks as quickly as possible and provide the user with a great experience. However, increased CPU performance comes with increased power consumption and reduced battery life on mobile devices. As an option to extend battery life, Android offers a battery-saver mode that when activated, turns off the power-hungry and faster processor cores and turns on the battery-conserving and slower processor cores on the device. The transition from using faster processor cores to using slower processor cores throttles the CPU clock speed on the device, and therefore impacts the webpage load process.We utilize a large-scale data-set collected by a real user monitoring system of a major content delivery network to investigate the impact of Android's battery-saver mode on various mobile Web performance metrics. Our analysis suggests that users of select smartphones of Huawei and Sony experience a sudden or gradual degradation in Web performance when battery-saver mode is active. Batterysaver mode on newer flagship smartphones, however, does not impact the mobile Web performance. Finally, we encourage for new website design goals that treat slow (and throttled-CPU) devices kindly in favor of improving end-user experience and suggest that Web performance measurements should be aware of user device battery charge levels to correctly associate Web performance.

show abstract

A case for faster mobile web in cellular IPv6 networks

Cited by 13 publications

References 23 publications

IPv6 Transition Measurements in LTE and VHT Wi-Fi Mobile Networks

IPv6 Transition Measurements in LTE and VHT Wi-Fi Mobile Networks

Domain-Sharding for Faster HTTP/2 in Lossy Cellular Networks

Web Performance with Android's Battery-Saver Mode

Contact Info

Product

Resources

About