Content Delivery Network Service Research Articles

TrafAda: Cost-Aware Traffic Adaptation for Maximizing Bitrates in Live Streaming

The business growth of live streaming causes expensive bandwidth costs from the Content Delivery Network service. It necessitates traffic adaptation, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , adapting video bitrates for cost-efficient bandwidth utilization, especially under the 95 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <i>th</i> </sup> percentile pricing. However, our data-driven investigations indicate the existing methods are hard to achieve bitrate-cost balance in a long month-level billing cycle due to dynamic traffic patterns. We propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TrafAda</i> , a learning-based cost-aware traffic adaptation method consisting of i) an ultra-long-term bandwidth demand forecasting model to learn complex bandwidth usage patterns, and ii) an imitation learning-based bitrate decision mechanism to optimize the ultra-long-term objective. We have implemented and deployed <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TrafAda</i> on a large-scale live streaming system in China serving over one billion viewers from 388 cities. The results show that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TrafAda</i> improves peak-hour bitrate, quality of experience (QoE), and watching time by 34.75%, 44.56%, and 10.68%, respectively, without extra bandwidth cost, which can be converted to a considerable value for a commercial system.

Pemanfaatan layanan cloud computing dan docker container untuk meningkatkan kinerja aplikasi web

This research aims to analyze and implement the utilization of Content Delivery Networks (CDN) and cloud computing services in web applications using Nginx Proxy Manager and Docker containers. This research uses experimental methods and case studies to analyze the performance of web applications before and after using CDN services and implementing the application in a cloud computing environment using Docker containers. Next, Nginx Proxy Manager is used to provide a user-friendly interface for configuring and managing Nginx proxy servers. Docker containers are used to manage web applications and proxy servers. In the CDN and cloud computing environment, Cloudflare Google Cloud Platform will be used as service providers. The results of this research indicate a significant improvement in the security and reliability of web applications after implementing CDN services and adopting a cloud computing environment. Additionally, this research provides insight into how to use Nginx Proxy Manager and Docker container to manage web applications efficiently and effectively in a cloud computing environment.

AMAZON S3 STATIC WEBSITE HOSTING

Nowadays, especially in period of energetic crisis, it is very important to get reliable data storage and uninterrupted access to websites. Considering that the world’s mainstream is using the clouds services for educational, scientific, business purposes it is important to explore features of hosting websites in major cloud service providers. One of the leader’s vendors in cloud servicing is Amazon with its Web Services (AWS). AWS offers hosting solutions such as Lightsail, Amplify Console, Simple Storage Service (S3) and Elastic Cloud Computing, from simple static websites to complicated dynamic, The article deals with the method of hosting static websites. Simple Storage Service was used for this purpose. The method has many benefits: the service is simple, cheap, scales well, manages everything and has great integration with other AWS services. This method is the best for websites that do not contain server-side scripting, like PHP or ASP.NET. It is ideal for websites that change infrequently, such as personal, promo, startup websites, websites of small businesses and organizations. The article focuses on how set permissions, properties and policies of AWS S3 bucket for hosting a website. But usage of S3 bucket alone provides only http connection. That’s why to deploy secure connection and speed up delivery of the static content was proposed to use AWS CloudFront service. CloudFront is content delivery network service, that uses the latest version of Transport Layer Security Protocol and saves files in the cache for 24 hours. The article shows how to configure CloudFront distribution to serve https requests for an Amazon S3 bucket. Configuration S3 bucket REST API endpoint was used as the most secure case. As a result, latency was reduced, security was improved due to traffic encryption and the content of S3 bucket was kept private.

A Review on Software Defined Content Delivery Network: A Novel Combination of CDN and SDN

With the rapid growth of multimedia services, the burden of infrastructure is becoming heavier, which may not guarantee a high-quality experience for users. Whether the multimedia category is video on demand (VOD), live broadcast, or social media is a hot issue. Traditionally, content delivery network (CDN) service providers deploy servers as close to clients as possible to reduce latency. CDN is an overlay network mainly responsible for routing requests, distribution, delivery, and audit. As a supplement to computing and storage capacity, CDN service providers have begun migrating some of their services to the cloud to focus on the delivery process. At the same time, the collaboration between CDN service providers promotes scalability to meet the growing number of requests. However, the underlying physical layer needs to be improved. The cost of the traditional underlying network is very high and cannot be easily expanded. The concept of software definition network (SDN) is proposed to solve these problems. SDN functions are divided into control layer and data layer. In the SDN, the global view is achieved to receive information by centralization, and then the deployment is dynamically adjusted to control the network. In this paper, we summarize the advantage structure and problems of a newly developed system SDCDN (software defined content distributed network). We introduce similar research progress. Then we summarize the current situation of SDN and CDN, compare and analyze their advantages and disadvantages, and finally put forward the directions of open issues and the future research.

Software‐defined content delivery network at the edge for adaptive video streaming

SummaryIn DASH (Dynamic Adaptive Streaming over HTTP), it is required to store multiple encoding for same video segments which increases the storage requirement of CDN (content delivery network) servers. Further, to keep network delay in check, it is not feasible to continuously increase the number of CDN servers in different geographical locations as per the ever‐increasing demand. Peer‐to‐peer (P2P) CDN acts as an elegant alternative to supplement existing CDN servers to harness the benefits of both P2P and CDN technologies. We present the concept of “CDN at the edge,” where end hosts can collaborate for video streaming in pure P2P manner to provide better QoE (Quality of Experience). However, dynamically changing behavior of underlying network, uncertain leaving, joining pattern of clients, and frequent change in source node during video streaming session make the selection of the desired bitrate (at any given time) extremely difficult. In this paper, we have designed and implemented a software‐defined network (SDN)‐based P2P CDN system that supports browser‐based adaptive video streaming. SDN provides global view of the topology and thus assists in providing timely information about leaving or joining peers in the proposed “CDN at the edge.” Further, this SDN‐based P2P CDN is compared with non‐SDN‐based P2P CDN using both virtual switch‐based emulations and physical switch‐based actual implementation. The comparison is performed under different churn rates with respect to QoE parameters such as start‐up delay, bitrate, stall count, and its duration. The experimental results prove the applicability of our approach.

Access protocol aware controller design for eMBB traffic in SD-CDN

Under a tremendous increase in 5th Generation (5G) traffic, the conventional Domain Name Server(DNS)-based routing in Content Delivery Network (CDN) cannot meet enhanced Mobile BroadBand (eMBB) communication requirements. It routes the traffic flow to the local CDN server which is closest to the end-user geographically using Border Gateway Protocol (BGP). Therefore; the overloading in the local CDN server is inevitable. According to literature, the access types of eMBB content achieve a higher hit-ratio in CDN caching and the anycast-based routing can redirect some of the traffic flows to the original server despite having a geographical disadvantage. Therefore, we consider access protocol types of a user such as TCP-based Hyper-Text Transfer Protocol 2 (HTTP2) and Quick User Datagram Protocol Internet Connections (QUIC) which are commonly used today in iphone Operating System (ioS) and Android-based applications. However, they do not have a central view to select optimal server according to access type in CDN and they are unaware of network dynamics such as protocol-based load and heterogeneity; i.e. the number of total matches packets and the ratio of the packets according to access protocols (HTTP2/QUIC). Therefore, we propose access protocol aware Software Defined Content Delivery Networks (SD-CDN) for eMBB traffic which decouples Data and Control planes and enables easy implementation of anycast routing on this infrastructure. It dynamically takes statistics from OpenFlow switches in the Data plane to calculate a novel load and the heterogeneity metrics per device periodically. It embeds the OpenFlow rules by using a novel access protocol-aware routing algorithm in SD-CDN. According to performance evaluation in a real test-bed environment; the Quality of Server (QoS) of received eMBB content by end-user from either the local or the original CDN servers, the proposed SD-CDN outputs 60% better PSNR than the conventional DNS-based routing with an acceptable rate for 5G requirements.

A Semantic virtualized network functions description and discovery model

Network Function Virtualization (NFV) has increasingly gained importance to address some emerging networking challenges like agility and cost-effectiveness. NFV enables to run Virtualized Network Functions (VNF) on top of any generic, Commercial-Off-The-Shelf (COTS) hardware, anytime and anywhere in the network. Specific service providers offer VNFs to prospective network providers. Service providers publish VNFs in dedicated marketplaces where network providers search VNFs and instantiate them according to a pre-established service-level agreement. On top of being proprietary and specific to the service providers, the existing VNF description models include details on VNF deployment but fail to fit VNF functional and non-functional specifications. This description alters an efficient selection of the most relevant VNFs and prevents full automation of the VNFs provisioning. This paper introduces a novel domain-independent VIrtualized networK functIoN ontoloGy (VIKING for short) for VNF description and publication in federated repositories. It also proposes a semantic-based matchmaking algorithm to discover and select the most relevant VNFs that satisfy prospective VNF consumers’ requests. As for validation, a prototype called Mastermyr Chest, including VIKING’s instantiation along with the matchmaker in Content Delivery Networks (CDN) domain was implemented. This prototype illustrates a new way to contribute to the redesign of the CDN’s traditional architecture by enabling value-added CDN service provisioning in an agile and dynamic manner. A set of experiments was run to (i) evaluate the matchmaker performances and (ii) demonstrate its accuracy and precision.

STYX: A Hierarchical Key Management System for Elastic Content Delivery Networks on Public Clouds

Hosting content delivery networks (CDNs) on clouds has the potential to improve the performance as resources and caches can be placed closer to subscribers. However, avoiding data leakage over an untrusted public cloud is critical, especially for sensitive data such as the SSL private key. The popular Keyless SSL solution allows content owners to retain on-premise custody of SSL private keys on their own key servers, but this solution likely causes performance bottlenecks and impedes the elasticity of CDNs. This paper describes a novel key management system, named STYX, for transmitting trusted data over untrusted channels and storing them on untrusted platforms. STYX accomplishes secure key provisioning for CDN scale-out and the key is securely protected with full revocation rights for CDN scale-in. STYX is implemented as a three-phase hierarchical key management scheme by leveraging Intel Software Guard Extensions (SGX) and QuickAssist Technology (QAT). Furthermore, STYX supports CDN services by integrating Nginx as the SSL termination proxy and the popular Redis/Memcached/Apache as backend caching engines. The performance evaluation shows that STYX significantly outperforms the native HTTPS servers on the CDN node due to QAT acceleration, providing up to a 5x enhancement in throughput and a 50 percent reduction in latency.

Elastic FemtoCaching: Scale, Cache, and Route

The advent of elastic Content Delivery Networks (CDNs) enable Content Providers (CPs) to lease cache capacity on demand and at different cloud and edge locations in order to enhance the quality of their services. This article addresses key challenges in this context, namely how to invest an available budget in cache space in order to match spatio-temporal fluctuations of demand, wireless environment and storage prices. Specifically, we jointly consider dynamic cache rental, content placement, and request-cache association in wireless scenarios in order to provide just-in-time CDN services. The goal is to maximize the an aggregate utility metric for the CP that captures both service benefits due to caching and fairness in servicing different end users. We leverage the Lyapunov drift-minus-benefit technique and Jensen's inequality to transform our infinite horizon problem into hour-by-hour subproblems which can be solved without knowledge of future file popularity and transmission rates. For the case of non-overlapping small cells, we provide an optimal subproblem solution. However, in the general overlapping case, the subproblem becomes a mixed integer non-linear program (MINLP). In this case, we employ a randomized cache lease method to derive a scalable solution. We show that the proposed algorithm guarantees the theoretical performance bound by exploiting the submodularity property of the objective function and pick-and-compare property of the randomized cache lease method. Finally, via real dataset driven simulations, we find that the proposed algorithm achieves 154% utility compared to similar static cache storage-based algorithms in a representative urban topology.

CDN Slicing over a Multi-Domain Edge Cloud

We present an architecture for the provision of video Content Delivery Network (CDN) functionality as a service over a multi-domain cloud. We introduce the concept of a CDN slice, that is, a CDN service instance which is created upon a content provider's request, is autonomously managed, and spans multiple, potentially heterogeneous, edge cloud infrastructures. Our design is tailored to a 5G mobile network context, building on its inherent programmability, management flexibility, and the availability of cloud resources at the mobile edge level, thus close to end users. We exploit Network Functions Virtualization (NFV) and Multi-access Edge Computing (MEC) technologies, proposing a system which is aligned with the recent NFV and MEC standards. To deliver a Quality-of-Experience (QoE) optimized video service, we derive empirical models of video QoE as a function of service workload, which, coupled with multi-level service monitoring, drive our slice resource allocation and elastic management mechanisms. These management schemes feature autonomic compute resource scaling, and on-the-fly transcoding to adapt video bit-rate to the current network conditions. Their effectiveness is demonstrated via testbed experiments.

RL-Cache: Learning-Based Cache Admission for Content Delivery

Content delivery networks (CDNs) distribute much of the Internet content by caching and serving the objects requested by users. A major goal of a CDN is to maximize the hit rates of its caches, thereby enabling faster content downloads to the users. Content caching involves two components: an admission algorithm to decide whether to cache an object and an eviction algorithm to determine which object to evict from the cache when it is full. In this paper, we focus on cache admission and propose a novel algorithm called RL-Cache that uses model-free reinforcement learning (RL) to decide whether or not to admit a requested object into the CDN's cache. Unlike prior approaches that use a small set of criteria for decision making, RL-Cache weights a large set of features that include the object size, recency, and frequency of access. We develop a publicly available implementation of RL-Cache and perform an evaluation using production traces for the image, video, and web traffic classes from Akamai's CDN. The evaluation shows that RL-Cache improves the hit rate in comparison with the state of the art and imposes only a modest resource overhead on the CDN servers. Further, RL-Cache is robust enough that it can be trained in one location and executed on request traces of the same or different traffic classes in other locations of the same geographic region. The paper also reports extensive analyses of the RL-Cache sensitivity to its features and hyperparameter values. The analyses validate the made design choices and reveal interesting insights into the RL-Cache behavior.

Reducing Service Creation Time Leveraging on Network Function Virtualization

Fifth-generation (5G) networks are envisioned to simultaneously support several services with different connectivity requirements. In this respect, service creation time is a key performance indicator (KPI) for service providers when planning the migration to 5G. For example, the European 5G infrastructure public private partnership (5G-PPP) suggests to reduce this time from 90 hours to 90 minutes, in the different phases of the service creation time KPI identified by this organization. This reduction can be achieved by leveraging on 5G state-of-the-art technologies: network function virtualization, network slicing, software-defined networking, and cloud computing, among others. Although some authors and projects have already studied the service creation time KPI in 5G, there is no literature that comprehensively analyzes and presents results related to each phase of this KPI. In this article, we explore the potential of network function virtualization technologies to reduce service creation time. To this end, we investigate the various phases of the service creation time KPI by designing and implementing, a realistic as well as complex network service that leverages on network function virtualization and related technologies. For our use case, we chose a content delivery network service specifically designed to distribute video. This decision was based on an analysis where we considered several parameters, like the complexity in the phases of design, fulfillment, and service assurance. We dissected all phases of the service creation time KPI required to turn our service blueprint into a deployment by utilizing network function virtualization tools. Henceforth, we defined and conducted several experiments, which were oriented to analyzing the different phases of the service creation time KPI. After analyzing the obtained results, we can conclude that using these new tools permits a substantial reduction in the time taken by each phase of the service creation time KPI.

Joint Location-Dependent Pricing and Request Mapping in ICN-Based Telco CDNs For 5G

Telco content delivery networks (CDNs) have envisioned building highly distributed and cloudified sites to provide a high-quality CDN service in the 5G era. However, there are still two open problems to be addressed. First, telco CDNs are operated upon the underlay network evolving towards information-centric networking (ICN). Different from CDNs that perform on the application layer, ICN enables information-centric forwarding to the network layer. Thus, it is challenging to take advantage of the benefits of both ICN and CDN to provide a high-quality content delivery service in the context of ICN-based telco CDNs. Second, bandwidth pricing and request mapping issues in ICN-based telco CDNs have not been thoroughly studied. In this paper, we first propose an ICN-based telco CDN framework that integrates the information-centric forwarding enabled by ICN and the powerful edge caching enabled by telco CDNs. Then, we propose a location-dependent pricing (LDP) strategy, taking into consideration the congestion level of different sites. Furthermore, on the basis of LDP, we formulate a price-aware request mapping (PARM) problem, which can be solved by existing linear programming solvers. Finally, we conduct extensive simulations to evaluate the effectiveness of our design.

Peer‐assisted content delivery network by vehicular clouds: Algorithm and evaluation

In this paper, we study how to improve the performance of content delivery networks (CDNs) using the novel concept of vehicular clouds. Our main idea is to use a cluster of connected vehicles (ie, vehicular clouds) to act as virtual servers in CDNs in order to relieve the workload of physical CDN servers. The key benefits include guaranteed content availability/reliability in terms of mean time between failures and optimized data storage workload at vehicular clouds. Workload is an important metric as it is directly related to the latency of content delivery. Specifically, we present a content placement algorithm that has an optimal workload at vehicular clouds. Finally, we verify the system efficacy using our own simulator based on simulation of urban mobility (SUMO) and Python.

A control and management architecture supporting autonomic NFV services

The proposed control, orchestration and management (COM) architecture is presented from a high-level point of view; it enables the dynamic provisioning of services such as network data connectivity or generic network slicing instances based on virtual network functions (VNF). The COM is based on Software Defined Networking (SDN) principles and is hierarchical, with a dedicated controller per technology domain. Along with the SDN control plane for the provisioning of connectivity, an ETSI NFV management and orchestration system is responsible for the instantiation of Network Services, understood in this context as interconnected VNFs. A key, novel component of the COM architecture is the monitoring and data analytics (MDA) system, able to collect monitoring data from the network, datacenters and applications which outputs can be used to proactively reconfigure resources thus adapting to future conditions, like load or degradations. To illustrate the COM architecture, a use case of a Content Delivery Network service taking advantage of the MDA ability to collect and deliver monitoring data is experimentally demonstrated.

Exploring and Analysing the African Web Ecosystem

It is well known that internet infrastructure deployment is progressing at a rapid pace in the African continent. A flurry of recent research has quantified this, highlighting the expansion of its underlying connectivity network. However, improving the infrastructure is not useful without appropriately provisioned services to exploit it. This article measures the availability and utilisation of web infrastructure in Africa. Whereas others have explored web infrastructure in developed regions, we shed light on practices in developing regions. To achieve this, we apply a comprehensive measurement methodology to collect data from a variety of sources. We first focus on Google to reveal that its content infrastructure in Africa is, indeed, expanding. That said, we find that much of its web content is still served from the US and Europe, despite being the most popular website in many African countries. We repeat the same analysis across a number of other regionally popular websites to find that even top African websites prefer to host their content abroad. To explore the reasons for this, we evaluate some of the major bottlenecks facing content delivery networks (CDNs) in Africa. Amongst other factors, we find a lack of peering between the networks hosting our probes, preventing the sharing of CDN servers, as well as poorly configured DNS resolvers. Finally, our mapping of middleboxes in the region reveals that there is a greater presence of transparent proxies in Africa than in Europe or the US. We conclude the work with a number of suggestions for alleviating the issues observed.

A data-driven and the deep learning based CDN recommendation framework for ICPs

It is a significant trend that the Internet Content Providers (ICPs) improve the quality of service and reduce the cost of content distribution by the Content Delivery Networks (CDNs). In order to spend the less money to get better services, ICPs need to find a lot of information about CDNs, such as server deployment, performance, price and so on, to determine whether CDN services satisfy their requirements. Unfortunately, these information can’t be obtained by third party due to business secret. ICPs still choose CDNs on the basis of one-sided viewpoint. For this reason, we have proposed a data-driven and the deep learning based CDN recommendation framework for ICPs. The contributions lie in: 1) A three-tier CDN recommendation framework is presented to achieve data-driven and the deep learning based recommendation service. 2) A CDN recommendation model is built based on the deep neural network, which improves the efficiency of the recommendation service and satisfies the personalized demand. 3) A prototype system is developed and deployed on the real-world large-scale Internet in China. Experimental results demonstrate that the correctness of the recommendation results is up to 91%, and degree of satisfaction reached 80%.

Introducing Development Features for Virtualized Network Services

Network virtualization and softwarizing network functions are trends aiming at higher network efficiency, cost reduction and agility. They are driven by the evolution in Software Defined Networking (SDN) and Network Function Virtualization (NFV). This shows that software will play an increasingly important role within telecommunication services, which were previously dominated by hardware appliances. Service providers can benefit from this, as it enables faster introduction of new telecom services, combined with an agile set of possibilities to optimize and fine-tune their operations. However, the provided telecom services can only evolve if the adequate software tools are available. In this article, we explain how the development, deployment and maintenance of such an SDN/NFV-based telecom service puts specific requirements on the platform providing it. A Software Development Kit (SDK) is introduced, allowing service providers to adequately design, test and evaluate services before they are deployed in production and also update them during their lifetime. This continuous cycle between development and operations, a concept known as DevOps, is a well known strategy in software development. To extend its context further to SDN/NFV-based services, the functionalities provided by traditional cloud platforms are not yet sufficient. By giving an overview of the currently available tools and their limitations, the gaps in DevOps for SDN/NFV services are highlighted. The benefit of such an SDK is illustrated by a secure content delivery network service (enhanced with deep packet inspection and elastic routing capabilities). With this use-case, the dynamics between developing and deploying a service are further illustrated.

Adaptive TTL-Based Caching for Content Delivery

Content delivery networks (CDNs) cache and serve a majority of the user-requested content on the Internet. Designing caching algorithms that automatically adapt to the heterogeneity, burstiness, and non-stationary nature of real-world content requests is a major challenge and is the focus of our work. While there is much work on caching algorithms for stationary request traffic, the work on non-stationary request traffic is very limited. Consequently, most prior models are inaccurate for non-stationary production CDN traffic. We propose two TTL-based caching algorithms that provide provable performance guarantees for request traffic that is bursty and non-stationary. The first algorithm called d-TTL dynamically adapts a TTL parameter using stochastic approximation. Given a feasible target hit rate, we show that d-TTL converges to its target value for a general class of bursty traffic that allows Markov dependence over time and non-stationary arrivals. The second algorithm called f-TTL uses two caches, each with its own TTL. The first-level cache adaptively filters out non-stationary traffic, while the second-level cache stores frequently-accessed stationary traffic. Given feasible targets for both the hit rate and the expected cache size, f-TTL asymptotically achieves both targets. We evaluate both d-TTL and f-TTL using an extensive trace containing more than 500 million requests from a production CDN server. We show that both d-TTL and f-TTL converge to their hit rate targets with an error of about 1.3%. But, f-TTL requires a significantly smaller cache size than d-TTL to achieve the same hit rate, since it effectively filters out non-stationary content.

A data-driven approach of performance evaluation for cache server groups in content delivery network

In industry, Content Delivery Network (CDN) service providers are increasingly using data-driven mechanisms to build the performance models of the service-providing systems. Building a model to accurately describe the performance of the existing infrastructure is very crucial to make resource management decisions. Conventional approaches that use hand-tuned parameters or linear models have their drawbacks. Recently, data-driven paradigm has been shown to greatly outperform traditional methods in modeling complex systems. We design a data-driven approach to building a reasonable and feasible performance model for CDN cache server groups. We use deep LSTM auto-encoder to capture the temporal structures from the high-dimensional monitoring data, and use a deep neural network to predict the reach rate which is a client QoS measurement from the CDN service providers’ perspective. The experimental results have shown that our model is able to outperform state-of-the-art models.