HTTP/3 marks a significant advancement in protocol development, utilizing QUIC as its underlying transport layer to exploit multiplexing capabilities and minimize head-of-line blocking. The introduction of the Extensible Prioritization Scheme (EPS) offers a signaling mechanism for controlling the order of resource delivery. In this study, we introduce two mappings: Resource Type Aware Mapping (RTAM) and Direct Mapping (DM) from Chromium priority hints to EPS urgency levels with the goal of enhancing the key web performance metrics. The mappings are evaluated using EPS’s urgency-based, non-incremental resource delivery method on eight popular websites. We observe that First Contentful Paint (FCP), Largest Contentful Paint (LCP), Time To Interactive (TTI), and Speed Index (SI) show prominent improvements among the websites. Furthermore, RTAM outperforms DM in conjunction with non-incremental resource delivery method, achieving up to 100% improvements in Total Blocking Time (TBT) for some websites. An overall enhancement in Quality of Experience (QoE) metrics is observed across the range of websites.

This paper contains the research proposal of Sindhu Chellappa that was presented at the MMSys 2022 doctoral symposium. With the ever growing demand for media consumption, HTTP Adaptive Streaming (HAS) is the de-facto standard for adaptive bit rate streaming. Several efforts have been made to increase the Quality of Experience (QoE) for the users by proposing Adaptive Bit Rate (ABR) streaming algorithms to choose the best bitrate for the current network condition. With HAS, the segments are downloaded through Hyper Text Transfer Protocol (HTTP) which operates over Transmission Control Protocol (TCP). HTTP/3 is the recent standardized application protocol which operates over Quick UDP Internet Connection (QUIC). When the protocols changes, it is necessary to revisit the QoE of the ABR algorithms to select the suitable ABR algorithm best suited for the protocol. We present four research questions to address the adaptability between the ABR algorithms and the underlying protocols.

Opportunistic Mobile Social Networks (MSNs) are a type of delay-tolerant networks that are composed of mobile nodes with social characteristics. Most of current research on forwarding algorithms for MSNs leverages the properties of social networks, such as community and centrality. However, the fundamental property, interaction, has not received sufficient attention. The effects of interactions consist of two parts: direct effects and indirect effects. Most existing work based on interactions takes only direct effects into consideration. We propose a distributed Interaction Based Routing (IBR) algorithm. Our main contribution is a distributed method to compute the popularity of nodes by using the direct and indirect effects of interactions. We compare IBR with Epidemic, Prophet, and BubbleRap under three different scenarios. The results show that IBR improves the forwarding efficiency significantly and independent of the network scenarios.

In pre-planned large-scale mobile networks, nodes cooperate to achieve the missions assigned to the networks. Due to the mobility of nodes, the common assumption that the contemporaneous path between any two nodes always exists cannot be made. Local algorithms, with the characteristics locality and simplicity, can make the disconnected systems robust and reliable. The recently proposed work [1] provides simple, local algorithms to let nodes perform local tasks while achieving global information delivery. In this paper, we study these algorithms by investigating the properties of node trajectories, such as their distribution and length. We also examine the performance of these algorithms under different networks and failure scenarios. The results reveal that the properties of the trajectories are critical indicators of the performance of the algorithms.

Hop-by-hop transport and in-network storage have been widely used in intermittently connected networks. As a way to provide hop-by-hop congestion control, backpressure has been used in hop-by-hop transport and in-network storage protocols to provide quick congestion alleviation. However, relying on backpressure to direct traffic can introduce large packet delays, especially when traffic load is light, which in turn can degrade the energy efficiency of the system. In this paper, we propose a cross-layer protocol, Reactive Store-and-Forward (ReSaF). This protocol adopts a conditional-forwarding routing algorithm to route packets on a per-block basis. The conditional-forwarding algorithm takes advantage of estimated delivery probability, available buffer space and energy level of nodes to provide an energy efficient way to delivery packets with a high rate of success.

Robustness and reliability in large-scale sensor networks present a triple challenge. First, the large scale (possibly compounded by the use of unreliable communication) makes failures -- transient and permanent -- unavoidable. Second, the difficulty of deploying such systems, especially in challenging environments, makes it next to impossible to replace failed nodes or rearrange surviving nodes after each failure. Third, the limited resources and capabilities of most sensor nodes prohibit elaborate, global/centralized approaches, especially in view of the potentially large number of nodes in the network. In this paper, we propose simple mechanisms to alleviate the effect of node failure in large-scale disconnected networks and show that beneficial global network properties can emerge from local rules achievable by simple nodes.

HTTP-based segmented delivery has become a popular choice for video distribution. The ability for clients to explicitly perform dynamic bitrate adaptation on segment boundaries has allowed clients to optimize the end user viewing experience. Client-based rate adaptation schemes, though, are typically greedy in nature and within the context of the network as a whole, these schemes may be sub-optimal. In this paper we propose a congestion-aware rate adaptation scheme that works on top of existing segmented delivery protocols to provide network bandwidth management and fairness. We examine the HTTP Live Streaming approach and compare our community-based approach with the native Apple® implementation, though the framework is extendable to other protocols. We show the improved user experience with our scheme through higher quality video delivery and lower bitrate selection thrashing.

Digital rights management (DRM) and video quality are two important issues for content providers. DRM plays a critical role in enforcing monetization policies while video quality is key for protecting brand integrity. For stitched media file-based (SMF) rate adaptation schemes which rely on low latency seek operations, a seek friendly DRM encryption scheme is required. This paper describes our stream cipher-based DRM encryption scheme which enables secure SMF rate adaptation. We provide a model for and discuss the trade-offs involved in optimizing rate adaptation latency within our scheme. Our SMF scheme is one of a number of HTTP video delivery implementations that we have commercially deployed on the AndroidTM and Blackberry®mobile platforms and that use our DRM. Though the DRM scheme works with both SMF and segment-based approaches, this paper focuses on SMF considerations. We include cipher performance results from actual device tests, as well as rate adaptation latency results from our production implementation.

Monetization of premium video content requires two key components: quality and security. In the mobile space, quality issues have been largely address in recent years through faster CPUs and segment-based HTTP adaptive bitrate delivery (e.g., Apple®HTTP Live Streaming, Adobe®HTTP Dynamic Streaming, and Microsoft®Silverlight™ Smooth Streaming). This has enabled plethora of new services to deliver high definition video to smart phone and tablet devices. This in turn has brought mobile digital rights management (DRM) under increased scrutiny. Existing DRM guidelines are geared toward monolithic file distribution to fixed end-points. The emergence of segment-based HTTP distribution protocols has produced unforeseen issues in existing content delivery workflows. Support for over-the-top (OTT) delivery has further complicated DRM enforcement. In this paper we analyze the impact of segmented OTT delivery on traditional content preparation and distribution workflows. We discuss the use of pipelining for optimizing transcoding, segmentation, and encryption workflows. We also address security and scalability concerns related to key generation, device registration, and key distribution workflows. Together, these workflows provide the necessary components for an efficient and secure OTT content delivery platform.

The ubiquity and simplicity of HTTP makes it a popular choice for Web-based video retrieval. However, HTTP was not designed for retrieving data with just-in-time tolerances; HTTP servers have always taken an as-fast-as-possible approach to data delivery. For media with known bandwidth constraints (e.g., audio/video files), HTTP servers can be enhanced and optimized by taking these constraints into account. For these data types, we present our architecture for an HTTP streaming server using paced output with intelligent bursting. We also discuss the scalability advantages of our HTTP streaming server architecture and compare it with traditional HTTP server response times and bandwidth usage.

We consider the case of sparse mobile sensors deployed to implement missions in challenging environments. This paper explores a notion of tour networks that is well suited to circumstances in which autonomous sensing agents cannot rely on standard networking abstractions and must create their own opportunities for communication and interaction. Tours are high-level building blocks that combine motion, communication and sensing and can be assembled to implement a broad class of autonomous sensing missions. They are supported by an architecture designed to deliver performance and robustness without compromising design abstraction.

Abstract The solar‐powered autonomous underwater vehicle (SAUV) platform has been developed over the past several years as a means to achieve long‐endurance cooperative missions, and the design of mobile routing protocols has become one of the main areas of focus during this process. This paper reports on the lessons learned from the design, simulation‐based evaluation, and field testing of two protocols. Two areas for possible improvements are proposed and evaluated using simulations. Finally, we present suggestions drawn from field experience for improving the design, simulation, and evaluation of protocols for use in a harsh underwater environment. Copyright © 2008 John Wiley & Sons, Ltd.

Dynamic Bandwidth allocation in Ethernet Passive Optical Networks (EPONs) has been an area of intense research in recent years. Most of the proposed solutions offer clever methods for fair grant sizing, traffic prediction, and prioritized, differentiated services. Barring some work by Kamal et al. and some elements in the scheme proposed by Ma et al., no work has been done on exploring the order of granting (i.e., ONU sequencing) in an EPON. In this paper, we propose an unexplored heuristic for improving the performance of the IPACT scheme with respect to the most important metric: packet delay. In this heuristic, the OLT always grants that ONU which has the Smallest (Available) Reported queue length, First (SARF). Our simulations indicate that our heuristic can improve the delay performance of IPACT by 10-20% (when tested under the gated allocation policy).

Interleaved Polling with Adaptive Cycle Time (IPACT) is one of the earliest proposed polling schemes for dynamic bandwidth allocation in Ethernet Passive Optical Networks (EPONs) and has been extensively used as a benchmark by many subsequent allocation schemes. In this paper, we attempt to construct a mathematical model of the IPACT scheme under the gated service discipline. For N = 1 ONU, we derive closed-form expression for the steady state grant size. For N > 1 ONUs, we need to consider separately a small and a large load-distance ratio. For the former case, the N = 1 ONU model holds even for N > 1. For the latter case, we find a closed form expression for the grant size. Our model shows a reasonable match with the values obtained from simulation for the steady state queue size and hence the throughput and delay.

The popularity of Web-based transactions and the need for more sophisticated content distribution methods has helped to fuel the rapid growth of Web Service adoption, specifically, HTTP-bound Web Services. Secure and efficient content delivery has long been a requirement of traditional Web-based distribution schemes, and existing the Web infrastructure provides numerous options for securing and optimizing HTTP. Two exemplary technologies are SSL/TLS and HTTP compression. While efforts to solidify the more granular WSSecurity standards are ongoing, and methods for XML message compression schemes continue to be investigated, HTTP provides an interim solution, supporting transactional security and message compression. The SSL/TLS and HTTP compression technologies have become commoditized and pervasive. And with the trend in content delivery toward hardware offload for these functions, modern data centers have begun to raise the bar for performance. In this paper, we examine three different paradigms for implementing SSL/TLS and HTTP compression: softwarebased functionality, server-resident hardware accelerators, and centralized network-resident hardware accelerators. We discuss the trade-offs between the two different offload techniques (i.e., PCI accelerator vs. network proxy) and explore their relationship to the current performance activities, in the field of Web Services. In analyzing the results for SSL/TLS offload, and the effects of compression, in conjunction with SSL/TLS, we draw parallels with the efforts of WS-Security and XML message compression. Although optimizations for software-based cryptography will continue to advance, the potential for hardware-based acceleration should not be overlooked. We discuss our results and address deployment scenarios for optimizingWeb-based transactions, and the future optimization of Web Service transactions.

We derive a closed-form expression for the message collision probability in the IEEE 802.3ah Ethernet passive optical network (EPON) registration scheme. The expression obtained, although based on an approximation, shows a good match with simulation results. We use the results of our analysis to compute the size of the most efficient contention window and the most efficient number of nodes serviced by a given window size.

The benefits provided by Web service protocols are well recognized. Deployments to date, however, have concentrated on new applications, and existing Web-based applications. A host of legacy applications and protocols continue to exist in their native forms, outdated, yet entrenched due to large installed bases. This paper details our observations in integrating a Web service infrastructure into the simple network management protocol (SNMP). SNMP has been in use for over a decade and a half, predominantly in network equipment embedded systems. Our Web service-based approach allows us to enhance our existing application with XML/SOAP interoperability, SSL/TLS security, and the potential to migrate both application and protocol layers to encompass future extensions and Web browser accessibility. The difficulty with SNMP, and many other legacy networking protocols, is that much of the extensive installed base is hosted on limited capability, legacy hardware. While the benefits of our scheme are quite tangible, the performance impact of adding these features is not well known. We examine two approaches, an integrated solution using a light-weight HTTP/SOAP stack, as well as a standard Java Web server implementation. Our tests reveal unanticipated performance results through both the integrated and proxy methods. We discuss the impact of these anomalies on the viability of our approach and address the broad issue of migrating Web services to legacy embedded architectures.

This paper proposes a new approach to providing fault tolerance in MPLS networks based on the concept of "domain protection" where protection paths for all working paths that terminate in an egress router are calculated simultaneously. The proposed scheme guarantees that every protected node is connected to two protection paths placed in a way that no single link failure would cause simultaneous loss of connectivity between a node and the egress router on both protection paths. The use of dual protection paths permits decoupling the protection path placement from the working path placement thus allowing much greater flexibility than other previously proposed schemes. Several heuristics to improve the quality and reduce the cost of the protection path placement are proposed and evaluated. The simulation results show that the algorithm together with the heuristic extensions achieves protection which is less costly or comparable to two other proposed MPLS protection schemes-RSVP backup tunnels and fast reroute-while exhibiting comparatively lower algorithmic complexity.

We propose STWnet, a self-routing high bandwidth optical network architecture for interconnecting users, grouped together as g groups with w users per group. STWnet uses the three dimensions of space, time, and wavelength by combining the advantages of space and temporal switching with the benefits of wavelength parallel data transmissions. Technologically difficult switching of individual wavelengths is avoided by prearranging transmissions in a way that they can be switched in a wavelength insensitive manner. Wavelengths are reused within the network thus allowing for a larger switching fabric. The proposed architecture can be internally expanded either in the spatial or temporal dimension to allow for multiple packets to be delivered to the same destination group. The expansion factor is determined based on the group knockout principle and given typical traffic patterns is a small number. STWnet allows easy group to group multicasting and broadcasting while system-wide multicasts and broadcasts can be achieved through repetitive group-to-group transmissions. The network uses readily available components such as opto-electronic directional couplers, fixed wavelength transmitters, and diffraction based parallel receivers while avoiding the use of relatively slow and expensive tunable components.

We propose SXmin: a self-routing, group-knockout principle based asynchronous transfer mode (ATM) packet switch which provides comparable delay-throughput performance and packet loss probabilities at significantly reduced hardware requirements compared to earlier switches. The M/spl times/N SXmin consists of an N/spl times/N Batcher sorter followed by log/sub 2/N-1 stages of sort-expander (SX) modules arranged in the form of a complete binary tree. Each SX module consists of a column of 2/spl times/2 switches with a wraparound-unshuffle input-output interconnection. This enables the hierarchical utilization of the group-knockout principle to expand the number of inputs by a small factor at each stage, resulting in a significant reduction in overall hardware complexity. Routing at each switch is controlled by a single bit. However, in case of contention, a dual bit resolution algorithm is used locally which drops excess packets in a predetermined manner while ensuring global randomness of packet loss over the entire switching network. There are no internal buffers at the individual stages and therefore the internal delay is constant and proportional to the number of stages. The use of simple hardware components and regular interconnections in the SX modules makes the network suitable for optical implementation.