Content Delivery Infrastructure Research Articles

User Content Delivery Service for Efficient POMA based LTE Channel Spectrum Scheduling Algorithm

Currently Long-Term Evolution (LTE) technologies are developing with an advanced network content delivery infrastructure sharing capabilities for different users and operators. Therefore, an efficient open-infrastructure of Position-orthogonal multiple access (POMA) based LTE networks with the content delivery capability is crucial to design and implement at this stage of research in LTE area. In this paper, the main contribution is the design and implementation of such LTE networks users open-infrastructure to enable the content delivery and sharing capabilities to be performed in virtual iterative precoding architectures. The proposed user content delivery infrastructure enables this service and sharing to the components of LTE architecture. A channel spectrum scheduling algorithm has been implemented through validation with test level experiments conducted to estimate the compatibility of LTE infrastructure virtualization in proposed open-infrastructure. The impact and scope of the proposed algorithm service can make the operational cost of the LTE mobile user to reduce

Deep Reinforcement Learning for Adaptive Caching in Hierarchical Content Delivery Networks

Caching is envisioned to play a critical role in next-generation content delivery infrastructure, cellular networks, and Internet architectures. By smartly storing the most popular contents at the storage-enabled network entities during off-peak demand instances, caching can benefit both network infrastructure as well as end users, during on-peak periods. In this context, distributing the limited storage capacity across network entities calls for decentralized caching schemes. Many practical caching systems involve a parent caching node connected to multiple leaf nodes to serve user file requests. To model the two-way interactive influence between caching decisions at the parent and leaf nodes, a reinforcement learning (RL) framework is put forth. To handle the large continuous state space, a scalable deep RL approach is pursued. The novel approach relies on a hyper-deep Q-network to learn the Q-function, and thus the optimal caching policy, in an online fashion. Reinforcing the parent node with ability to learn-and-adapt to unknown policies of leaf nodes as well as spatio-temporal dynamic evolution of file requests, results in remarkable caching performance, as corroborated through numerical tests.

Real-Time Video Content Popularity Detection Based on Mean Change Point Analysis

Video content is responsible for more than 70% of the global IP traffic. Consequently, it is important for content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. In this paper, we propose the employment of on-line change point (CP) analysis to implement real-time, autonomous and low-complexity video content popularity detection. Our proposal, denoted as real-time change point detector (RCPD), estimates the existence, the number and the direction of changes on the average number of video visits by combining: (i) off-line and on-line CP detection algorithms; (ii) an improved time-series segmentation heuristic for the reliable detection of multiple CPs; and (iii) two algorithms for the identification of the direction of changes. The proposed detector is validated against synthetic data, as well as a large database of real YouTube video visits. It is demonstrated that the RCPD can accurately identify changes in the average content popularity and the direction of change. In particular, the success rate of the RCPD over synthetic data is shown to exceed 94% for medium and large changes in content popularity. Additionally, the dynamic time warping distance, between the actual and the estimated changes, has been found to range between 20 samples on average, over synthetic data, to 52 samples, in real data. The rapid responsiveness of the RCPD is instrumental in the deployment of real-time, lightweight load balancing solutions, as shown in a real example.

A Generic Approach to Video Buffer Modeling Using Discrete-Time Analysis

The large share of traffic in the Internet generated by video streaming services puts high loads on access and aggregation networks, resulting in high costs for the content delivery infrastructure. To reduce the bandwidth consumed while maintaining a high playback quality, video players use policies that control and limit the buffer level by using thresholds for pausing and continuing the video download. This allows shaping the bandwidth consumed by video streams and limiting the traffic wasted in case of playback abortion. Especially in mobile scenarios, where the throughput can be highly variant, the buffer policy can have a high impact on the probability of interruptions during video playback. To find the optimal setting for the buffer policy in each network condition, the relationship between the parameters of the buffer policy, the network throughput dynamics, and the corresponding video playback behavior needs to be understood. To this end, we model the video buffer as GI/GI/1 queue with pq -policy using discrete-time analysis. By studying the stochastic properties of the buffer-level distribution, we are able to accurately evaluate the impact of network and video bitrate dynamics on the video playback quality based on the buffer policy. We find a fundamental relationship between the bandwidth variation and the expected interarrival time of segments, meaning that overproportionately more bandwidth is necessary to prevent stalling events for high bandwidth variation. The proposed model further allows to optimize the trade-off between the traffic wasted in case of video abortion and video streaming quality experienced by the user.

Quaestor

Today, web performance is primarily governed by round-trip latencies between end devices and cloud services. To improve performance, services need to minimize the delay of accessing data. In this paper, we propose a novel approach to low latency that relies on existing content delivery and web caching infrastructure. The main idea is to enable application-independent caching of query results and records with tunable consistency guarantees, in particular bounded staleness. Q uaestor (Query Store) employs two key concepts to incorporate both expiration-based and invalidation-based web caches: (1) an Expiring Bloom Filter data structure to indicate potentially stale data, and (2) statistically derived cache expiration times to maximize cache hit rates. Through a distributed query invalidation pipeline, changes to cached query results are detected in real-time. The proposed caching algorithms offer a new means for data-centric cloud services to trade latency against staleness bounds, e.g. in a database-as-a-service. Q uaestor is the core technology of the backend-as-a-service platform Baqend, a cloud service for low-latency websites. We provide empirical evidence for Q uaestor 's scalability and performance through both simulation and experiments. The results indicate that for read-heavy workloads, up to tenfold speed-ups can be achieved through Q uaestor 's caching.

Understanding Performance of Edge Content Caching for Mobile Video Streaming

Today’s Internet has witnessed an increase in the popularity of mobile video streaming, which is expected to exceed 3/4 of the global mobile data traffic by 2019. To satisfy the considerable amount of mobile video requests, video service providers have been pushing their content delivery infrastructure to edge networks—from regional content delivery network (CDN) servers to peer CDN servers (e.g., smartrouters in users’ homes)—to cache content and serve users with storage and network resources nearby. Among the edge network content caching paradigms, Wi-Fi access point caching and cellular base station caching have become two mainstream solutions. Thus, understanding the effectiveness and performance of these solutions for large-scale mobile video delivery is important. However, the characteristics and request patterns of mobile video streaming are unclear in practical wireless network. In this paper, we use real-world data sets containing 50 million trace items of nearly 2 million users viewing more than 0.3 million unique videos using mobile devices in a metropolis in China over two weeks, not only to understand the request patterns and user behaviors in mobile video streaming, but also to evaluate the effectiveness of Wi-Fi and cellular-based edge content caching solutions. To understand the performance of edge content caching for mobile video streaming, we first present temporal and spatial video request patterns, and we analyze their impacts on caching performance using frequency-domain and entropy analysis approaches. We then study the behaviors of mobile video users, including their mobility and geographical migration behaviors, which determine the request patterns. Using trace-driven experiments, we compare strategies for edge content caching, including least recently used (LRU) and least frequently used (LFU), in terms of supporting mobile video requests. We reveal that content, location, and mobility factors all affect edge content caching performance. Moreover, we design an efficient caching strategy based on the measurement insights and experimentally evaluate its performance. The results show that our design significantly improves the cache hit rate by up to 30% compared with LRU/LFU.

Performance evaluation for new web caching strategies combining LRU with score based object selection

The topic of Internet content caching regained relevance over the last years due to the increasing and widespread use of content delivery infrastructures to meet capacity and delay demands of multimedia services. In this study, we evaluate the performance of web caching strategies in terms of the achievable hit rate for realistic scenarios of large user populations. We focus on a class of score gated least recently used (SG-LRU) strategies which combine the simple update effort of the LRU policy with the flexibility to keep the most important content in the cache according to an arbitrarily predefined score function.Caching efficiency is evaluated via simulations assuming Zipf distributed requests, which have been confirmed manifold in access pattern of popular web platforms for video streaming and various other content types. In this paper, we analyze the hit rate gain of alternative web caching strategies for the standard independent request model (IRM) within the complete relevant range of three basic system parameters. The results confirm absolute hit rate gains of 10%–20% over pure LRU as realistic estimation in general, as experienced in other trace-driven case studies for special caching strategies.Moreover, we compare the performance for the independent request model with results for correlated dynamic request pattern over time, based on Wikipedia statistics that are available as daily top-1000 page requests. In this way, we show that the IRM analysis is valid for caches with a large user population, although high dynamics tends to reduce the achievable hit rate below the IRM result for smaller user communities.

The Collaboration for Content Delivery and Network Infrastructures: A Survey

With the explosive growth of Internet traffic and the rapid development of network technology, content delivery has become a significant service in the current Internet. However, the existing content delivery solutions, such as peer-to-peer and content delivery network, have many insufficient aspects. Meanwhile, the collaboration between content delivery and network infrastructures has been considered as a promising technique in network field. From the perspective of collaboration, content delivery systems can make full use of the network characteristics and the effective information provided by the network operators, so as to improve the efficiency of the content distribution and optimize the overall performance of the network. In this paper, we present a comprehensive survey on the collaboration for content delivery and network infrastructures. First, we provide some of the works, which have been done on collaboration solutions from two perspectives: evolutionary and revolutionary. And then, the advantages and disadvantages of these solutions are compared and analyzed from three aspects of technology, business, and standardization. Finally, we outline some challenges and research directions in the future.

Designing mobile content delivery networks for the internet of vehicles

Content delivery is a key functionality for developing the Internet of Vehicles. In such networks, vehicles act as sensors of the urban mobility by constantly exchanging messages with another vehicles, the cellular network, and also the infrastructure (roadside units). However, the task of delivering content in such dynamic network is far from trivial. In this work, we investigate the development of Content Delivery Networks (CDN) in the context of vehicular networks. Roadside units support the communication by replicating and delivering contents to vehicles within their range of coverage. Initially, we devise a strategy for measuring the performance of the content delivery in vehicular networks. Then, we use the proposed metric for designing a deployment strategy allowing us to identify the better locations for deploying the roadside units in order to properly support the dissemination of a variety of contents, each content requiring specific levels of performance. We compare our deployment strategy to the intuitive strategy of allocating roadside units at the densest locations of the road network. The results demonstrate that our strategy requires less roadside units than the baseline for non-massive deployments in order to achieve similar levels of performance, incurring in less costs when setting up the content-delivery infrastructure.

Uniform load sharing on a hierarchical content delivery network interconnection model

Content management system (CMS) is an infrastructure for efficient distribution, organization, and delivery of digital content. It is desirable that the content must be successfully delivered regardless of the end users location or attachment network. For the end to end delivery of content, a virtual open content delivery infrastructure is formed by interconnecting several CDNs. In this paper, we focus on content delivery network interconnection. An efficient and suitable to implement hierarchical CDNI architecture, named as HCDNI, is proposed to reduce the limitations of CDNIs. Next, a content distribution and redistribution scheme is proposed so that the searching time and the round trip time for the content delivery can be minimized. Next, we find a reliable and fault tolerant scheme for web server replica placement and content caching. Finally, analysis and simulation studies show that proposed algorithm results in a significant improvement in terms of data routing, path selection, content distribution and redistribution, load balancing and network scalability.

High-Performance Centralized Content Delivery Infrastructure: Models and Asymptotics

We consider a centralized content delivery infrastructure where a large number of storage-intensive files are replicated across several collocated servers. To achieve scalable mean delays in file downloads under stochastic loads, we allow multiple servers to work together as a pooled resource to meet individual download requests. In such systems, basic questions include: How and where to replicate files? What is the impact of dynamic service allocation across request types, and whether such allocations can provide substantial gains over simpler load balancing policies? What are tradeoffs among performance, reliability and recovery costs, and energy? This paper provides a simple performance model for large systems towards addressing these basic questions.

Recent Changes in the Internet Landscape

Abstract The Internet is in a state of constant flux. This property of the Internet, that stems from the pressure from its continuously reinvented usage, makes it a lively environment, which must be re-assessed fundamentally from time to time. In this article, we review recent trends that have radically changed our perception of what the Internet is. Popular applications such as Youtube and Facebook have demonstrated the massive scale at which services can be delivered across the Internet. This has been possible only thanks to the massive deployment in content delivery infrastructures, leading to the rise of the Cloud paradigm. Recent studies from the research community have observed the consequences of this growth in the Internet infrastructure, through structural changes in the Internet topology and traffic patterns. The increasingly dynamic nature of Internet traffic patterns asks for improved capabilities to engineer traffic, that require a completely different level of programmability from the network. Software-defined networking is expected to fill this gap. Overall, the picture of the Internet that we draw in this article is the one of a very dynamic, challenging, and exciting playground.

Understanding internet video viewing behavior in the wild

Over the past few years video viewership over the Internet has risen dramatically and market predictions suggest that video will account for more than 50% of the traffic over the Internet in the next few years. Unfortunately, there has been signs that the Content Delivery Network (CDN) infrastructure is being stressed with the increasing video viewership load. Our goal in this paper is to provide a first step towards a principled understanding of how the content delivery infrastructure must be designed and provisioned to handle the increasing workload by analyzing video viewing behaviors and patterns in the wild. We analyze various viewing behaviors using a dataset consisting of over 30 million video sessions spanning two months of viewership from two large Internet video providers. In these preliminary results, we observe viewing patterns that have significant impact on the design of the video delivery infrastructure.

Towards a predictive cache replacement strategy for multimedia content

In recent years, telecom operators have been moving away from traditional broadcast-driven television, towards IP-based interactive and on-demand multimedia services. Consequently, multicast is no longer sufficient to limit the amount of generated traffic in the network. In order to prevent an explosive growth in traffic, caches can be strategically placed throughout the content delivery infrastructure. As the size of caches is usually limited to only a small fraction of the total size of all content items, it is important to accurately predict future content popularity. Traditional caching strategies only take into account the past when deciding what content to cache. Recently, a trend towards novel strategies that actually try to predict future content popularity has arisen. In this paper, we ascertain the viability of using popularity prediction in realistic multimedia content caching scenarios. The proposed generic popularity prediction algorithm is capable of predicting future content popularity, independent of specific content and service characteristics. Additionally, a novel cache replacement strategy, which employs the popularity prediction algorithm when making its decisions, is introduced. A detailed evaluation, based on simulation results using trace files from an actual deployed Video on Demand service, was performed. The evaluation results are used to determine the merits of popularity-based caching compared to traditional strategies. Additionally, the synergy between several parameters, such as cache size and prediction window, is investigated. Results show that the proposed prediction-based caching strategy has the potential to significantly outperform state-of-the-art traditional strategies. Specifically, the evaluated Video on Demand scenario showed a performance increase of up to 20% in terms of cache hit rate.

A case for a coordinated internet video control plane

Video traffic already represents a significant fraction of today's traffic and is projected to exceed 90% in the next five years. In parallel, user expectations for a high quality viewing experience (e.g., low startup delays, low buffering, and high bitrates) are continuously increasing. Unlike traditional workloads that either require low latency (e.g., short web transfers) or high average throughput (e.g., large file transfers), a high quality video viewing experience requires sustained performance over extended periods of time (e.g., tens of minutes). This imposes fundamentally different demands on content delivery infrastructures than those envisioned for traditional traffic patterns. Our large-scale measurements over 200 million video sessions show that today's delivery infrastructure fails to meet these requirements: more than 20% of sessions have a rebuffering ratio ≥ 10% and more than 14% of sessions have a video startup delay ≥ 10s. Using measurement-driven insights, we make a case for a video control plane that can use a global view of client and network conditions to dynamically optimize the video delivery in order to provide a high quality viewing experience despite an unreliable delivery infrastructure. Our analysis shows that such a control plane can potentially improve the rebuffering ratio by up to 2× in the average case and by more than one order of magnitude under stress.

UTILIZATION OF METADATA FOR QUALITY ASSURANCE IN THE INTERNATIONAL SHARING AND REUSE OF OPEN EDUCATIONAL RESOURCES

In order to sustain the development and utilization of learning content under the limited financial and human resources, the sharing and reuse of open educational resources (OERs) have been promoted in various school levels and disciplines. In many countries, central organization(s) collect both learning content and its metadata, and provide them to learners and teachers with distinctive value‐added services from their portals. Collaborations among the national portals have also emerged beyond borders.Global Learning Object Brokered Exchange (GLOBE) is an international consortium of the hub organizations, which manage a federated repository and/or a meta‐referatory in each country and region. GLOBE was established in September 2004 by five founding members, and now twelve organizations in the world participate in. By adopting IEEE LOM version 1.0 for the metadata standards, SQI for the query language, and OAI‐PMH for harvesting, GLOBE realized a global search and delivery infrastructure for lifelong learning (LLL). At present, while the repositories all over the world store huge number of learning content and metadata, the variance of the quality becomes much bigger. In addition, because the effectiveness of learning content is relative and dependent on contextual factors, both optimizations specialized in learner characteristics and localizations to each language/culture are indispensable. The users need some supports on the quality and pedagogical guidance when they find their right content. GLOBE, based on multilingualism, multiculturalism and pluralism, collaborates to build up standardized services on the quality assurance of the content utilizing the “Educational” items of LOM. In AAOU framework also, the organizations can promote “openness”, and share the quality learning content and educational information by using a standardized content search and delivery infrastructure.

A content delivery infrastructure to support advanced services for business travellers

A content delivery network (CDN) is a collection of network elements and procedures designed to efficiently connect information consumers with information providers, accelerating end-users access to contents and reducing network traffic and hardware requirements. The advent of nomadic users implies that CDNs should deal with routing and forwarding of requests for contents in a mobile environment too. The scenario we consider in this paper is that of the IST FIFTH project, where a local network on board a fast train is connected either via a satellite link or a Wireless-LAN access to the fixed content network. In this framework, we define a hierarchical structure of content servers for data management and delivery to business travellers on board trains. In addition, we propose a users' request routing approach to perform smart server selection, taking into account both network and server congestion.

Multitiered Cache Management and Acceleration for Database-Driven Websites

Response time is a key differentiation point among electronic commerce (e-commerce) applications. For many e-commerce applications, web pages are created dynamically based on the current state of a business stored in database systems. To improve the response time, many e-commerce websites deploy caching solutions for acceleration of content delivery. There are multiple tiers in the content delivery infrastructure where cache servers can be deployed, including (1) data caching (in data centers), (2) content page caching (in edge or front end caches), and (3) database query result set caching (between application servers and DBMS). The architecture of database-driven e-commerce websites is more complex than that of typical websites. It requires the integration of web servers, application servers, and back end database systems as well as dynamic content caching solutions. In this paper, the issues associated with the management of content page caching and database query result set caching tiers are studied. It is observed that caching management for these two tiers have their unique characteristics. It is because cached object types and information available for the caching management in the two tiers are different. Solutions are proposed for effective caching management for each tier and extensive evaluations are conducted. The experimental results show the usefulness of the technology in improving overall system performance.