In quantum networks, long-distance quantum communication relies on Einstein–Podolsky–Rosen (EPR) pairs. However, EPR pairs undergo decoherence over time, while their on-demand generation incurs latency. Pre-Storing EPR pairs in quantum memories can mitigate this latency, but such memories are costly and scarce resources, making their efficient placement an important design problem. While prior studies have mainly focused on routing or entanglement distribution, the placement of quantum memories under fidelity and purification constraints has received limited attention. In this paper, we propose an optimization framework for quantum memory placement that minimizes the average entanglement generation latency under a limited budget of deployable memories. Simulations across multiple real-world topologies show that our method consistently achieves lower latency than random and centrality-based placements, suggesting its potential to improve the scalability of large-scale quantum communication networks.

In recent years, with the spread of services that handle large amounts of content, the importance of content recommendation systems has increased. At the same time, many content providers utilize CDNs (Content Delivery Networks) because delays in content delivery and low throughput lead to an increase in service churn rates. However, when using CDNs in combination, issues arise because the content that should be recommended differs from the user's perspective and the network's perspective. Therefore, if content recommendation does not simultaneously consider both of these factors, there is a risk of degraded delivery quality. While previous studies have considered both factors, most of them assumed static environments. In this paper, we propose a recommendation method that aims to improve both user satisfaction and cache hit rate by utilizing reinforcement learning in an environment where user content preferences change dynamically.

In recent years, the low earth orbit (LEO) satellite network has attracted much attention. LEO satellites enable low latency communications, however the communication range of individual satellites is narrow, requiring a constellation of many satellites. The limited number of satellites in the initial phase of service provisioning limits the service area and communication time, making it difficult to secure users. In this paper, we propose a model in which multiple LEO satellite operators cooperate to provide services, and the revenue is reasonably allocated to each service provider based on the Shapley value of cooperative game theory. The possibility of extending communication services through cooperation among operators is verified through numerical evaluations.

With the rise of IoT and 5G, there is increasing demand for linked processing of various tasks, such as IoT data processing. In-network computing combined with information-centric networking (ICN) offers a generic solution, but autonomous distributed computing face challenges like task duplication and inefficient resource use in DAG. This paper proposes a scheduling algorithm using step-by-step task ordering and one-stroke requests to prevent duplication.

Content Delivery Network (CDN) enhances Internet scalability and resilience against traditional Distributed Denial-of-Service (DDoS) attacks by deploying cached edge servers close to end-users. However, CDNs remain vulnerable to unique attack vectors, such as cache pollution attack (CPA). While existing research focuses on attack detection and mitigation, this paper investigates optimized CPA strategies from an adversarial perspective, specifically targeting CDN origin servers to bypass edge server defenses and amplify DDoS impacts. We analyze the attack mechanisms, evaluate performance degradation under varying CPA parameters, and quantify the scalability of this DDoS vector. Experimental results demonstrate that optimized CPA can exhaust origin server resources with fewer malicious requests, highlighting a pressing need for robust CDN architectural countermeasures.

With the widespread adoption of social networking services (SNS) and streaming, the volume of traffic on the Internet has seen a rapid increase. Traditionally, to ensure the delivery of high-quality network services, mitigation strategies such as Content Delivery Networks (CDNs) have been proposed. How-ever, the underlying TCP/IP focuses on packet transfer based on the content's delivery source, the host, which may not be crucial content for users. In response to this, Named Data Networking (NDN) has been introduced as a new network architecture to efficiently deliver content. NDN incorporates features to deliver content from caches within the network. Nevertheless, a challenge persists when the content cache is not available nearby, requiring content retrieval from a distant location. Therefore, this paper proposes a method aiming to reduce inter-autonomous system (AS) traffic by dynamically generating copies, i.e., mirrors, of content at its origin. Additionally, to reflect a more realistic scenario, the proposed method is evaluated through computer simulations considering fluctuations in the popularity of real-world content. The numerical results confirm the effectiveness of the proposed approach.

The HTTP/2 and HTTP/3 protocols are commonly used as techniques for minimizing webpage load time by enabling the simultaneous retrieval of multiple webpage components. Despite their effectiveness in retrieving components from a single server, their performance degrades when interfacing with multiple servers due to the increased number of connections required. To address this challenge, our team introduced the notion of “co-occurrence”, which reflects the repeated appearance of certain components on different webpages, quantified as the “degree of co-occurrence”. Previous research has demonstrated the potential application of this concept in cache control schemes. Consequently, this article presents a new cache control methodology that utilizes the concept of co-occurrence, particularly when dealing with multiple cache servers supporting HTTP/2 and HTTP/3. The proposed method aims to consolidate co-occurring components on a single cache server as much as possible. The proposed method was evaluated by computer simulation and confirmed to improve the effectiveness of object aggregation while suppressing the degradation of hit rate.

The demand for content such as music and videos has increased, and it drives up the Internet traffic. Content recommendation systems that suggest content based on user preferences have become a crucial component of content delivery services like Netflix, and it accounts for a significant portion of content requests. Moreover, cache delivery using CDN (content delivery network) has been widely used to reduce the delivery latency and network traffic. However, existing recommendation systems did not consider the status of caches, and they focused on just recommending contents which fitted the user interest. From the user perspective, recommending content tailored to individual preferences is desirable. On the other hand, from the network-providers perspective, recommending highly popular content to reduce delivery costs is preferred. Therefore, in this paper, we propose a content recommendation method aimed at improving user satisfaction as well as enhancing cache efficiency. The proposed method limits the targets of collaborative filtering to contents cached within cache servers. It recommends content that is physically closer to users and cached in nearby cache servers to reduce the network load and shorten access times. Additionally, to address the cold-start problem in collaborative filtering, we combine cache-limited collaborative filtering with a recommendation method based on reinforcement learning called $\epsilon$-greedy. We evaluate the effectiveness of the proposed method through computer simulations.

Information-centric networking (ICN) has received a wide attention as a next-generation network. Unlike conventional IP networks, which forward packets based on IP addresses, ICNs packets are sent based on the name of the contents which are cached to the routers involved, then delivered to the consumer. Since any participating party can upload content to the ICN, the risk of content poisoning attack (CPA) is ever-present. In CPA, an attacker degrades the cache efficiency by uploading fake content under a real name posting as a legitimate publisher. As a countermeasure to CPA, many existing methods determine the legitimacy of content using digital signatures with public keys, and alerts routers of unjustified content upon detection. However, it is difficult for them to detect fake-CPA attacks which use public keys of fabricated content to generate digital signatures. Most methods also lack counter-measures to spoofed fake-CPA attacks, in which the certification authority (CA) that manages the public key or its staff member colludes with the attacker to rewrite the legitimate publisher’s public key to the attacker’s and inject fake contents that pretends to be authentic contents that are high in popularity into the cache. In this paper, we propose a method to prevent the spoofed fake-CPA by managing content names with IOTA, a distributed ledger technology that blocks tampering of contents registered on the system. We also numerically compare the search time and memory requirement of four search methods that search content names managed in the ledger in the proposed method. As a result, we confirm the trade-off between the search time and memory requirement.

In recent years, there has been growing interest in the effective operation of any cast Content Delivery Networks (CDN). Existing anycast CDNs answer the same set of cache servers (CSes) for all content requests without considering the geographical distribution of popularity of contents. It has also been reported that the appropriateness of the selection of delivery servers decreases as the number of servers assigned to the same IP address increases in anycast delivery. To solve these problems, we propose a new anycast distribution method that incorporates the spatial locality of the popularity of content. Specifically, we propose an algorithm that creates multiple CS sets consisting of a small number of CSs and assigns the optimal CS set to each content by minimizing the number of CSes in the CS set. The algorithm aims to reduce the dispersion of delivery delay by minimizing the number of CSes in the CS set, while covering a wide range of countries. Using a genetic algorithm, we construct CS sets that incorporates spatially biased ASes and verify its effectiveness through computer simulation. Numerical evaluation results show that the proposed method can effectively cover about 80% of requests of highly popular content with about 15 ASes. It is expected that the proposed method can effectively cover the requirements for most of the contents with a small number of CS sets and improve the quality of user experience.

Information-Centric Networking (ICN) is gaining attention as the next generation network for efficient content delivery. ICN is deployed by each Internet Service Provider (ISP) based on its management decision. Since ISPs pay each other transit fees based on the amount of traffic they receive, the deployment of ICN has a significant impact on the profits of ISPs. Therefore, it is necessary to analyze the process of ICN adoption by ISPs in order to clarify the potential of ICN adoption. In this study, we use publicly available measured Autonomous System (AS) connection topology data and Multi-Agent Simulation (MAS) to clarify the feasibility of ICN deployment for ISPs, taking into account economic efficiency. We will also clarify the mechanisms necessary for the diffusion of ICN.

The amount of content delivery traffic on the Internet has been increasing. To make Content Delivery Network (CDN) effective, it is necessary to clarify the attack methods targeting CDN cache servers and prevent them effectively. Therefore, in this paper, we evaluate and investigate the influence of attacks against CDN cache servers on the effect of increasing the response time. Many providers deploy multiple cache servers to prevent DDoS attacks and provide better services. In this paper, we investigate the impact of the two types of attacks, the Distributed Denial of Service (DDoS) attack and the Cache Pollution Attack (CPA), against the CDN cache servers. With the CDN model consisting of multiple cache servers, we evaluate the effect of the DDoS and CPA targeting CDN caches under different attacking scenarios, and we analyze the factors that affect the attack effect. Finally, we find several factors that affect the effect of the attacks, and CDN provider can focus on these factors to prevent the threat of cyber attacks on CDNs.

In mobile crowdsensing (MCS), sensing data measured by mobile devices equipped with sensing capabilities are collected from various workers to estimate true values. Since the widespread use of smartphones, MCS has attracted particular attention because of its excellent sensing capabilities with low cost without sensing infrastructure. However, values measured by smartphones have errors, so MCS service providers need to control errors. Moreover, MCS cannot totally avoid attackers who try to maximize the errors in estimated values of MCS, e.g., data poisoning attack (DPA). Although existing studies related these issues assumed DPA of a single area, many MCSs seem to consist of multiple areas. In addition, although existing studies have assumed that sensing data is collected from all workers, it is necessary to provide incentives to workers under the budget constraints of the MCS, and it is expected that sensing data is actually collected only from workers sampled with a certain probability. Therefore, this paper proposes a method for setting the optimal number of workers to sample in each area when collecting sensing data from workers in multiple areas under the condition that the total number of sampled workers is fixed. Through the computer simulation, we confirm the effectiveness of the proposed method in improving the estimation accuracy in MCS with multiple areas.

In the last decade, ICN (Information-Centric Networking) has attracted our attention as a replacement of the current Internet. ICN is a network architecture focusing on the content itself rather than the host. This poses a challenge to design ICN to be scalable in terms of the content space. One solution to this scalability concern is to reduce the size of FIB (Forwarding Information Base) of ICN routers. Several studies have already tried to reduce the FIB size, but it has not been fully understood how potentially the routing strategy contributes to reducing the FIB size. Therefore, the present paper reports how much the FIB size at routers can be reduced by our proposed routing strategy, Constrained Shortest-Path Tree (CSPT) routing. The CSPT routing intentionally adopts a detour path passing through a specific node so as to aggregate more FIB entries. As a consequence of the preliminary evaluation, we reveal that our CSPT routing can significantly reduce the FIB size while keeping the path length.

Information-centric networking (ICN) has been widely studied as a new network architecture for efficient content delivery. However, the threat of content poisoning attack (CPA) has been pointed out. In the CPA, a malicious user degrades the cache hit ratio of router caches by inserting fake contents into router caches. To investigate effective countermeasures against the CPA, we need to clarify the impact of the CPA on reducing the cache hit ratio. However, existing studies assumed limited CPA schemes, and they were based on small network topologies without considering caching methods and locations of fake contents and bots. In this paper, we focus on the original-fake type CPA which is the easiest CPA for attackers and most difficult to be detected, and we investigate the potential threat of the original-fake CPA for various caching methods on large-scale networks when optimally assigning locations of fake contents and bots.

Information-centric networking (ICN) has been at-tracting attention as a network that efficiently transfers content such as internet of things (IoT). Named data networking (NDN) is one of the architectures in ICN. In NDN, access control by Publisher is difficult, and privacy leakage for content names is an issue. Although an existing access control method of NDN called name-based access control (NAC) solves the access control issue in the NDN, the problem of privacy leakage still remains. Moreover, although we can avoid privacy leakage by encrypting content names, the encrypted content name can be still identified by a frequency attack. Therefore, we propose an access control method at Publisher which always allows the first Interest to reach Publisher, protects privacy by encrypting content names, and reduces the impact of frequency attack by dynamically changing the encrypted content names. In the proposed method, however, the amount of control traffic will increase. Therefore, we compare the amount of control traffic of the proposed method with that of the NAC method. Moreover, we confirm the effectiveness of the proposed method by evaluating its defense ability against the frequency attack.

In recent years, with the rapid growth of Social Networking Service (SNS), e.g., Twitter and Facebook, the traffic of user-generated contents has dramatically increased. One of promising network architectures to realize efficient content delivery in SNS is cache network, e.g., Content Delivery Network (CDN) and Information-Centric Networking (ICN). In order for the cache network to effectively operate, the key is to appropriately design the caching strategy and the cache-replacement policy at caching nodes, e.g., cache server in CDNs and router in ICNs. However, to design a cache network for distributing SNS contents, we have to carefully consider tremendous factors – features of social network representing the social relationship among SNS users and those of cache network. Therefore, in this paper, we assume that ICN is introduced as a content distribution infrastructure for SNS, and extensively analyze the characteristics of the socially-aware ICN. In particular, we focus on the influential user called influencer, which is one of typical features of social networks, and investigate the effect of the selection of influential users on the content caching in ICNs. Consequently, we reveal that the cache hit ratio differs according to centrality measures that determine influential users.

Wildlife has been damaging crops in mountainous areas where it is difficult to access commercial power sources. Therefore, a power-saving system for monitoring harmful wildlife has been proposed that combines multiple sensor nodes with different levels of power consumption. A sensor with higher power consumption capable of obtaining more detailed information is activated step by step after a sensor with low power consumption detects approaching harmful wildlife. However, such a method does not consider the coordination among multiple sensor nodes deployed in a region. An autonomous recording unit (ARU) uses sound to monitor wildlife intermittently for long-term operation to reduce power consumption. However, events are not captured when the ARU is in standby mode. Therefore, we propose a wildlife monitoring system that improves the probability of event acquisition while reducing power consumption by only activating sensor nodes in a target area. Other sensors are activated in standby mode by receiving wake-up calls sent by those sensor nodes when they detect events. We also evaluate the ability of event detection and power consumption of the proposed method.

A new type of Distributed denial of service (DDoS) attack, called a crossfire attack (CFA), has appeared, whereby attackers prevent packets from arriving at servers in the target area by sending many packets on some links connecting the target area and other areas and overloading those links. In the CFA, bot hosts repeatedly send traceroute packets to multiple servers in the target area within a short time frame to select the target links prior to the attack. Therefore, to prevent the CFA, it might be effective to filter all the traceroute packets sent within a time interval less than a threshold. However, legitimate hosts which send traceroute packets within a short time interval will be also detected as bot hosts. In this paper, we propose a method using two thresholds to detect the bot hosts of a CFA: one threshold is used for detecting the target servers of traceroutes; the other threshold is used for detecting the bot hosts. By using two thresholds, we can expect to detect bot hosts while avoiding false identification of legitimate hosts. We also propose a method for optimally designing detection thresholds that maximize the detection accuracy against bot hosts from the given upper limit of false identification probability of legitimate hosts.

Mobile crowdsensing (MCS) using mobile devices has been gaining attention as a low-cost way to estimate various environmental data. In one type of MCS service, the service platform estimates environmental data, such as temperature or carbon dioxide concentration, from data reported by the mobile devices of crowdsensing workers. However, due to the nature of collecting data from such varied and unknown users, there is a risk of data poisoning attacks, in which malicious workers intentionally send data with large errors to increase the estimation error of data. While existing studies have investigated data poisoning attacks and methods of preventing them assuming a single area, data poisoning attacks on multiple areas have not been studied. In this paper, we propose strategies for data poisoning attacks in MCS across multiple areas in which the environmental value is estimated at each area independently. The proposed methods optimize the number of malicious workers allocated to each area to maximize the effectiveness of the attacker. Using computer simulations, we show that the proposed methods generate errors in many areas and increase the total error among all areas.

Information-centric networking (ICN), which forwards request packet on the basis of the name of the content without using the Domain Name System (DNS) and caches data packets at routers, is attracting attention as a network architecture that efficiently delivers content such as Internet of Things (IoT) data. Certain content, e.g., paid content, can be obtained by limited users, and an access-control method for determining whether the requesting user can access is required when requesting this type of content. On the current Internet, request packets reach the servers of content providers using the authoritative DNS server of content providers, so access control is easily achieved at this server. However, in ICN, access control is difficult because data packets can return from routers, and request packets cannot always reach the hosts of content providers. We propose an access-control method with which request packets can always reach the host of content providers. The overhead traffic of key distribution was derived using simple analysis models and numerically compared with an existing access-control method. We confirm that the proposed method creates approximately 7 Gbytes of control overhead traffic per day, while the existing method with which a key is periodically distributed to all users creates at least 250 Gbytes of unauthorized traffic per day.

In recent years, crowdsourcing, in which various workers provide computing and human resources for requested jobs through networks, has attracted wide attention as an efficient matching system between service users and workers. In crowdsourcing, a service provider needs to evaluate each worker on the basis of their contribution in order to fairly reward workers, so the provider is required to manage the private data of workers, e.g., trajectories and behaviors of workers. However, in crowdsourcing, a single service provider manages all the private data of workers, so there is a possibility that this data might be leaked or manipulated when the provider is attacked by malicious users. Moreover, employees of a service provider might abuse the private data of workers. To solve these problems, we propose managing the private data of workers by using the blockchain for crowdsourcing. In particular, we focus on a type of crowdsourcing in which workers have a direct contract with the service provider providing a service to users, e.g., Uber Eats. Moreover, we also propose a method for evaluating worker scores that uses an entropy weight to weight each evaluation criterion of workers on the basis of the distribution of each criterion. Through a computer simulation using data of cycling conditions obtained by the motion sensor of smartphones attached to bicycles, we evaluate the effectiveness of the proposed system when applied to worker evaluation for a food delivery service, e.g., Uber Eats.

Network softwarization is one of the key enablers of the future Internet evolution, also supporting the road from the fifth generation (5G) to the next-generation communication systems, namely 6G, with their main objective of bringing hyper-connected experience to every corner of society.

In large-scale disasters such as earthquakes, a quick and safe evacuation is necessary. For this purpose, it is important to smoothly share information about damaged roads among evacuees. However, in large-scale disasters, the communication infrastructure might become damaged to the point of being unusable or subject to a loss of power. As alternatives means of communications, various methods using the delay tolerant network (DTN) concept, in which mobile terminals directly communicate without using a communication infrastructure, have been proposed for disaster communication. However, when using the DTN, nearby mobile terminals need to exchange road information many times, so their power consumption will increase. Moreover, when a blackout occurs in a disaster, it may become impossible to charge the mobile battery, so it is important to reduce its power consumption. In this paper, we propose to place information boxes at multiple locations that store the information on road locations that have sustained damage. By exchanging such road information through the information boxes, i.e., uploading and downloading the information at the information boxes, one can expect to dramatically reduce the power consumption of mobile terminals. We also propose a method of determining the placement of the information boxes that considers the estimated number of evacuees passing through and their separation. A numerical evaluation using a multi-agent simulation shows the effectiveness of the proposed method.

A low-cost migration from IP-to-NDN is one of the most critical challenges that should be answered before the NDN protocol replacing the IP protocol on the Internet. A translation approach is considered as the practical and economical approach for IP to NDN migration. This paper proposes a lightweight translation method between IP and NDN by utilizing the data payload for bridging the semantic protocol gap in some possible producer and consumer scenarios. We evaluate the packet throughput, latency, and processing load to clarify the gateway effectiveness using a computer emulator.

To provide web browsing and video streaming services with desirable quality, cache servers have been widely used to deliver digital data to users from locations close to users. For example, in the MEC (mobile edge computing), cache memories are provided at base stations of 5G cellular networks to reduce the traffic load in the backhaul networks. Cache servers are also connected to many edge routers in the CDN (content delivery network), and they are provided at routers in the ICN (information-centric networking). However, the cache pollution attack (CPA) which degrades the cache hit ratio by intentionally sending many requests to non-popular contents will be a serious threat in the cache networks. Quickly detecting the CPA hosts and protecting the cache servers is important to effectively utilize the cache resources. Therefore, in this paper, we propose a method of accurately detecting the CPA hosts using a limited amount of memory resources. The proposed method is based on a Bloom filter using the combination of identifiers of host and content as keys. We also propose to use two Bloom filters in parallel to continuously detect CPA hosts. Through numerical evaluations, we show that the proposed method suppresses the degradation of the cache hit ratio caused by the CPA while avoiding the false identification of legitimate hosts.

The co-existence between Internet Protocol (IP) and Named-Data Networking (NDN) protocol is inevitable during the transition period. We propose a privacy-preserving translation method between IP and NDN called the dual-channel translation gateway. The gateway provides two different channels dedicated to the interest and the data packet to translate the IP to the NDN protocol and vice versa. Additionally, the name resolution table is provided at the gateway that binds an IP packet securely with a prefix name. Moreover, we compare the dual-channel gateway performance with the encapsulation gateway.

Information on evacuation shelters and routes is important for evacuees to evacuate safely and promptly when a disaster occurs. To take appropriate evacuation routes, evacuees need to gather information on roads that are damaged. However, in large-scale disasters, network infrastructure can be damaged and become unusable in many cases. To solve this problem, some existing studies propose providing evacuation routes by collecting road information from mobile terminals of evacuees by using a delay tolerant network (DTN) without using network infrastructure. However, in the existing works, congestion caused by evacuees concentrating on a specific route is a problem. This congestion is caused by identical evacuation routes being presented to all evacuees. Therefore, in this paper, we propose recommending different routes to evacuees on the basis of evacuee attributes (e.g. age) to alleviate congestion on roads. We numerically evaluate the proposed method by multi-agent simulation (MAS) and show that it achieves a 22% or greater reduction in the average evacuation distance and 44% or greater reduction in the average evacuation time where the number of evacuees was 30% of the daytime population.

The Softwarization of networks is enabled by the SDN (Software-Defined Networking), NV (Network Virtualization), and NFV (Network Function Virtualization) paradigms, and offers many advantages for network operators, service providers and data-center providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special issues was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on the management of softwarized networks.

In recent years, ICN (Information-Centric Networking) has been under the spotlight as a network that mainly focuses on transmitted and received data rather than on the hosts that transmit and receive data. Generally, the communication networks such as ICNs are required to be robust against network failures caused by attacks and disasters. One of the metrics for the robustness of conventional host-centric networks, e.g., TCP/IP network, isreachabilitybetween nodes in the network after network failures, whereas the key metric for the robustness of ICNs iscontent availability. In this paper, we analytically obtain the content availability against the node removal in the case where just a single path from a node to a repository, i.e., contents server, storing contents is available and where multiple paths to the repository are available, respectively. Furthermore, through several numerical evaluations, we investigate the effect of the structure of network topology as well as the pattern and scale of the network failures on the content availability in ICN. Our findings include that, regardless of patterns of network failures, the content availability is significantly improved by caching contents at routers and using multiple paths.

In recent years, video traffic has rapidly increased, and reducing video traffic is an important issue for network providers. By caching video content at cache servers close to users, network providers can expect to reduce the video traffic in the networks. However, the storage capacity of cache servers is limited, so it is necessary to carefully select contents to be cached to effectively utilize the limited cache resources. In order to make effective use of cache resources, it is important to cache content based on the popularity dynamics of video contents. It is known that video contents have different popularity dynamics in each video category. For example, videos of movie and music categories tend to maintain view counts over long time, whereas the view counts of videos of news and sports categories rapidly decrease. In this paper, we propose a caching method that selects video content to be cached based on the popularity dynamics of video content in each category. To clarify the effectiveness of the proposed caching method, we evaluate the cache hit ratio of the proposed method by a trace-driven simulator using a measured request pattern of YouTube videos. We show that the proposed method improves the cache hit ratio compared with the LRU.

In recent years, ICN (Information-Centric Networking) has been under the spotlight as a network that mainly focuses on transmitted and received data rather than on the hosts that transmit and receive data. Generally, the communication networks such as ICNs are required to be robust against network failures caused by attacks and disasters. One of the metrics for the robustness of conventional host-centric networks, e.g., TCP/IP network, is reachability between nodes in the network after network failures, whereas the key metric for the robustness of ICNs is content availability. In this paper, we focus on an arbitrary ICN network and derive the content availability for a given probability of node removal. Especially, we analytically obtain the average content availability over an entire network in the case where just a single path from a node to a repository, i.e., contents server, storing contents is available and where multiple paths to the repository are available, respectively. Furthermore, through several numerical evaluations, we investigate the effect of the structure of network topology as well as the pattern and scale of the network failures on the content availability in ICN. Our findings include that, regardless of patterns of network failures, the content availability is significantly improved by caching contents at routers and using multiple paths, and that the content availability is more degraded at cluster-based node removal compared with random node removal.

Information-centric networking (ICN), a new network architecture for efficiently delivering content, has been widely investigated recently. To be widely spread as a social infrastructure, ICN is required to sustain not only network availability, i.e., connectivity between operating routers, but also content availability, i.e., reachability to content, at network failures. In ICN, FIBs (forwarding information bases) at routers are configured so that content requests reach hosts of content providers having the originals of content. Therefore, requests for content whose connectivity to originals is lost cannot be transferred in networks, and the content availability of these content items is lost. However, copies of unavailable content are possibly cached at one or more operating routers in ICN, so content availability can be recovered by promoting one copy cached at operating routers to the original. Therefore, it is desirable to cache content at routers located far from its original to improve the recover probability of unavailable content items. In this paper, we propose a caching strategy of ICN to achieve this goal. Through numerical evaluation, we show that the proposed caching strategy can increase the maximum distance between the originals and cached copies by several percent to about 20% compared with the case simply caching content at all routers in ICN.

With the increase of smart devices, mobile crowdsensing, in which a crowdsensing Internet of Things (IoT) platform collects data from smart devices (such as smartphone) users, has become a popular paradigm. Various incentive mechanisms are widely employed for the IoT platform to incentivize smart device users to provide sensing data. Traditional works concentrated on rewarding smart device users for their short term effort to provide data, without considering smart device users’ long term factors and the quality of data. In this paper, smart device users’ quality of data is considered by incorporating smart device users’ long term factor reputation. A quality maximization problem with budget constraints is formulated for IoT platform, and the optimal pricing solution is obtained through theoretical analysis. Our proposed optimal pricing based incentive mechanism is validated by extensive numerical simulations.

By obtaining virtual machines (VMs) from infrastructure providers (InPs) according to the demand in public cloud services, service providers (SPs) can elastically provide network services to users. As the charging methods of VMs, reserved instance (RI) and on-demand instance (ODI) are widely used. For InPs, RI is more desirable than ODI thanks to easiness of estimating long-term revenue, risk aversion of occurring idle VM resources, and reduction of charging cost. In this paper, to improve the ratio of RI in VMs prepared by an InP, we propose VM trading methods in which idle RI of SPs with VM demand falling below the number of contracted RI are transferred to SPs with VM demand exceeding the number of contracted RI. As the VM trading mechanisms, we investigate two approaches: RI with self-help effort (RISE) and RI with mutual aid (RIMA). Through numerical evaluation using the demand pattern of commercial VoD service, we show that the proposed VM trading methods decrease the number of VMs required for ODI by about 50% to 100% and increase the ratio of RI by about 15% to 85%.

In recent years, the majority of internet traffic is video traffic such as YouTube and NetFlix, and reducing video traffic is an essential subject for network providers. Placing copy of original video contents in replication servers close to users has a chance to reduce the traffic in the networks. However, the storage capacity of a replication server is generally limited, so it is necessary to select video contents which will be requested in the near future. In the existing researches, temporarily or regionally popular contents are selected by focusing on the past popularity. In this paper, we would like to focus on another aspect of content request characteristics, user taste, and we propose to place contents which are not so popular but expected to be viewed by other users having similar tastes in the near future. Users with similar taste tend to request similar video contents, so we propose to select contents based on the measured tastes of users to place in the storage of replication servers. In the proposed method, cache storage is divided into two parts, a large static part and a small dynamic part. Cache-decision policy using user tastes is applied for the static part. In the dynamic part, LRU (Least Recently Used) cache-decision policy is applied to store contents having high popularity just after they are generated. The numerical results using MovieLens dataset show that the proposed method can make user groups with similar taste and improve the cache hit ratio compared with the LRU.

In this paper, we propose a high-precision zero-rating architecture based on our application-specific slicing technology [1] [2], where we tag traffic with application info at customized smart phones, which can be used to identify applications at MVNOs with 100% accuracy. We examine several zero-rating MVNO networks and observe that the existing zero-rating schemes of MVNOs are not accurate enough, and charge users on those applications that are advertised as count-free, which may result in disruption in the MVNO market. We also posit that the evaluation result highlights the necessity of our proposed architecture.

Recently, the virtualized CDN such as Amazon CloudFront has been widely used. In the virtualized CDN, cache servers are provided on virtual machines of cloud datacenters. Using the virtualized CDN, content providers can flexibly select the regions of using cache servers according to the geographical demand pattern of their users, so content providers can expect to reduce the cost of using CDN. However, to maximize the profit of content providers, they need to carefully select the locations of cache servers which affect the quality perceived by users as well as the total cost. In this paper, we propose a method of optimally selecting the geographical regions to use cache servers in the virtualized CDN maximizing the profit of the content providers.

In many countries, mobile virtual network operators (MVNOs) provide mobile network services to users by leasing the wireless bandwidth from mobile network operators (MNOs). To attract many users and increase the number of subscribers, some MVNOs introduce the strategy called zero rating (ZR) which exempts traffic of specific content providers (CPs) from usage-based charging. The ZR differentiates traffic of specific CPs from that of other CPs, so the ZR violates the principle of network neutrality, and the ZR is prohibited in some countries. However, to clarify the desirable rules against the ZR, we need to analyze its impact on end users. In this paper, we investigate the charging strategy of ZR MVNOs by analyzing the price plans of major MVNOs in Japan. Moreover, we model the dynamics of the MVNO market consisting of low-price (LP) MVNOs and ZR MVNOs by the evolutionary game which can model the dynamics of social environment described by strategic distribution. We show that the MVNO market will be monopolized by MVNOs using either strategy, and the monthly fee of users will increase at the steady state. Therefore, we conclude that ZR MVNOs are required to introduce a service plan for users who do not benefit from the ZR.

In recent years, the rapid development of cloud computing, networking, and mobile computing have substantially promoted mobile edge computing (MEC). Currently, most of the MEC services can be roughly divided into two categories: computation offloading to accelerate computation and save the energy of mobile devices and data services to shorten the latency between the content providers and the mobile users. Although emerging services such as user-specified transcoding and AR/VR systems require joint optimization of computing resource allocation and data placement, there is little research on it. In this work, we carry out an in-depth study on the interaction of computing resource allocation and data placement in mobile edge computing environments. Based on the analysis of the temporal and spatial characteristics of the two tasks, we propose a joint optimization framework that works with online manner. The proposed method employs hybrid timescales: a coarse-grained timescale to update the data placement and a fine-grained timescale to decide computing resource allocation. The proposed method achieves provable near-optimal performance without buffering users' requirements and does not assume that future trends in user requirements are predictable.

Information-centric networking (ICN), a new network architecture for efficiently delivering content, has been widely investigated recently. In ICN, cache memory is implemented at each router, and content items are routed in the network by using content name as the locator determining the destination. The caching strategy that determines the content to be cached at each router strongly affects the cache hit ratio and flow hop length, and it is important to efficiently utilize limited cache resources by avoiding duplicated caching of the same content among routers located closely. In this paper, we propose a spatially dispersed caching (SDC), which is a caching strategy dispersing content by assigning a binary ID to each router and limiting the cache targets at each router to content with names whose hash value coincides with the router ID. Through computer simulations using backbone networks of actual ISPs in the United States, we show that the SDC improves the cache hit ratio by about 30%-170% compared with the case when caching content at all routers on the default path, and the SDC reduces the average hop length at cache hit by about 50%-90% compared with the existing caching strategies. Moreover, we show that the SDC improves the sustainable ratio of content acquisition in large-scale failures of routers by about 25%-200% compared with the existing caching strategies.

To provide video streaming of user-generated contents (UGCs) with high quality and at low cost by maximizing the effect of content delivery network (CDN), CDN providers are required to adequately design CDN cache servers by accurately estimating the UGC view-count distribution. To achieve this goal in a practical time frame, we need to construct a simple time-series model that captures the transition of UGC popularity. Therefore, in this paper, we first analyze the daily view count (DVC) of YouTube videos over nine months and find that the DVC of YouTube videos obeys a lognormal distribution. As a simple time-series model of the DVC of each YouTube video, we propose the grouped MPP (gMPP), extending the multiplicative process (MPP) which is widely known as a simple time-series model generating a lognormal distribution. We also propose reproducing the DVC distribution of YouTube videos by using a superposed gMPP (SgMPP) that aggregates multiple gMPPs. The SgMPP can accurately reproduce the DVC distribution of YouTube videos with a low computational overhead, so we can expect to use the SgMPP as the input for computer simulations for designing various network components that require the popularity distribution of UGC, e.g., cache capacities. Through numerical evaluation, we confirm that we can adequately design the storage capacity of a cache server with the average error rate of several percent against the target cache hit ratio.

The softwarization of networks is enabled by the SDN (Software-Defined Networking), NV (Network Virtualization), and NFV (Network Function Virtualization) paradigms, and offers many advantages for network operators, service providers and datacenter providers. Given the strong interest in both industry and academia in the softwarization of telecommunication networks and cloud computing infrastructures, a series of special issues was established in IEEE Transactions on Network and Service Management, which aims at the timely publication of recent innovative research results on management of softwarized networks.

Information-centric networking (ICN), a new network architecture for efficiently delivering content, has been widely investigated recently. In ICN, cache memory is implemented at each router, and content items are routed in the network by using the content name as the locator determining the destination. The caching strategy that determines the content to be cached at each router strongly affects the cache hit ratio and flow hop length, and it is important to efficiently utilize limited cache resources by avoiding duplicated caching of the same content among routers located closely. However, no caching strategy aiming at dispersing content over networks has been investigated. In this paper, we propose spatially dispersed caching (SDC), which is a caching strategy dispersing content by assigning a binary ID to each router and limiting the cache targets at each router to content with names whose hash value coincides with the router ID. Through computer simulations using backbone networks of actual ISPs in the USA, we show that SDC reduced the average hop length by about 5% to 20% compared with the existing caching strategies.

By obtaining virtual machines (VMs) from infrastructure providers (InPs) according to the demand in public cloud services, service providers (SPs) can elastically provide network services to users. As the charging methods of VMs, reserved instance (RI) and on-demand instance (ODI) are widely used. For InPs, RI is more desirable than ODI thanks to easiness of estimating long-term revenue, risk aversion of occurring idle VM resources, and reduction of charging cost. In this paper, to improve the ratio of RI in VMs prepared by an InP, we propose VM trading methods in which idle RI of SPs with VM demand falling below the amount of contracted RI are applied to SPs with VM demand exceeding the amount of contracted RI. As the VM trading mechanisms, we investigate two approaches: RI with self-help effort (RISE) and RI with mutual aid (RIMA). Through numerical evaluation using the demand pattern of commercial VoD service, we show that the proposed VM trading methods decrease the number of VMs required for ODI by about 50% to 100% and increase the ratio of RI by about 10% to 70%.

Modern webpages consist of many rich objects dynamically produced by servers and client terminals at diverse locations, so we face an increase in web response time. To reduce the time, edge computing, in which dynamic objects are generated and delivered from edge nodes, is effective. For ISPs and CDN providers, it is desirable to estimate the effect of reducing the web response time when introducing edge computing. Therefore, in this paper, we derive a simple formula that estimates the lower bound of the reduction of the response time by modeling flows obtaining objects of webpages. We investigate the effect of edge computing in each webpage category, e.g., News and Sports, using data measured by browsing about 1,000 popular webpages from 12 locations in the world on PlanetLab.

With network virtualization, users of network resources, i.e., service providers (SPs), can be separated from the owners of the network resources, i.e., infrastructure providers (InPs). In this case, SPs are not required to own the network infrastructure, and they can focus on providing network services without constructing and managing the network infrastructure. This is expected to encourage small and medium-sized enterprises to start various types of network services. However, to promote the separation between owners and users of network resources, it is important to suitably define the management model of network resources that is necessary to trade resources between InPs and SPs. In this paper, we summarize the requirements for a management model of network resources considering the aspects of both InPs and SPs and propose a management model consisting of allocation blocks (ABs) that satisfies these requirements. We also propose a method of optimally designing the ABs for a given time-series of resource demand by SPs. Through numerical evaluation using the actual log data of a VoD service, we clarify that the proposed method can reduce the total fee paid by SPs compared with a straightforward approach, i.e., uniform ABs.

For Internet service providers to efficiently use network resources, they need to conduct traffic engineering to dynamically control traffic routes to accommodate traffic with limited network resources. The performance of traffic engineering depends on the accuracy of traffic prediction. However, the volume of network traffic has been changing drastically in recent years due to the growth of various types of network services, making traffic prediction increasingly difficult. Our simple ideas to overcome this challenge are to separate traffic into predictable and unpredictable parts and to apply different control policies to predictable and unpredictable traffic. To promote these ideas, we use software-defined networking technology, particularly Open-Flow, that can control macroflows defined by any combination of L2-L4 packet header information such as 5-tuple. In this paper, we therefore propose the macroflow-generating method for separating traffic into predictable macroflows that have little traffic variation and unpredictable macroflows that have large traffic variation within a limited flow table size. We also propose a macroflow-based traffic engineering scheme that uses different routing policies in accordance with traffic predictability. Simulation evaluation results suggest that our proposed scheme can reduce the maximum link load in a network at the most congested time by 34% and the average link load in a network on average by 11% compared with the current traffic engineering schemes.

The number of users downloading video content on the Internet has dramatically increased recently, and obtaining huge rich content items, e.g., movie files, is becoming a popular application of the Internet. The transmission bandwidth consumed by delivering rich content is enormous. Therefore, it is urgent for Internet service providers to design an efficient delivery system that minimizes the amount of network resources consumed. Content delivery networks are widely used to reduce the flow hop length and the response time when obtaining content items. To improve the cache efficiency, various methods of replacing caches have been proposed, with the target of improving the cache hit ratio. The effect of reducing the traffic load on network links by delivering content items from caches depends on the distance from the origin server to the users. However, many existing methods do not consider the origin server distance. To maximize the effect of reducing the link load obtained by caches, we propose here a cache-replacement policy based on the hop distance to the origin servers. To simplify the caching algorithm, we divide the storage capacity of cache severs into multiple virtual caches and manage content items separately on the basis of the hop distance to origin servers. We also propose an optimal method for designing the capacities of virtual caches that maximizes the total expected reduction of flow hop length. Through numerical evaluation, we show that the proposed method can decrease the average link load by about 10% compared with the normal least recently used.

Web performance, the time from clicking a link on a web page to finishing displaying the web page of the link, is becoming increasingly important. Low web performance of web pages tends to result in the loss of customers. In our research, we measured the time for downloading files on popular web pages by running web browsers on four hosts worldwide using PlanetLab and detected the longest portion in download time. We found the longest portion in download time to be Blocked time, which is the waiting time for the start of downloading in web browsers. In this paper, we propose a method for accelerating web performance by reducing such Blocked time with a cache of rendering results. The proposed method uses an in-network rendering function which renders web pages instead of web browsers. The in-network rendering function also stores the rendering results in its cache and reuses them for other web browsers to reduce the Blocked time. To evaluate the proposed method, we calculated the web performance of web pages whose render results are cached by analyzing the measured download time of actual web pages. We found that the proposed method accelerates web performance of long round trip time (RTT) web pages or long RTT clients if the web pages' dynamic file percentages are within 80%.

Modern websites consist of many rich objects dynamically produced by servers and client terminals at diverse locations. Consequently, we face complications in understanding the communication structure generated when accessing websites. To reduce the response time at browsed websites, many website objects are delivered using content delivery networks (CDNs), in which data objects are delivered from cache servers located close to user terminals. Although the use of CDNs have been assumed to reduce web response time, the actual effect of CDNs on this reduction has not been clarified. To answer this fundamental question, we measured the communication structure of traffic generated when accessing the 1,000 most popular websites from 12 locations worldwide. We found, for example, that it will be desirable to give high priority to entertainment websites at night and to business-related websites during the day.

The transmission bandwidth consumed by delivering rich content is enormous, so it is urgent for Internet service providers to design an efficient delivery system that minimizes the amount of network resources consumed, i.e., minimizing the hop length of delivery flows. Content delivery networks are widely used to reduce the flow hop length and the response time when obtaining content items. To improve the cache efficiency, various methods of replacing caches have been proposed, with the target of improving the cache hit ratio. However, although the effect of delivering content items from caches depends on the distance from the origin server to the users, these methods do not consider the origin server distance. We propose here a cache-replacement policy based on the hop distance to the origin servers. We divide the storage capacity of cache severs into multiple virtual caches and manage content items separately based on the hop distance to origin servers. We also propose an optimal method for designing the capacities of virtual caches that maximizes the total expected reduction of flow hop length. Through numerical evaluation, we show that the proposed method can increase the average expected reduction of flow hop length by about 20% to 80% compared with normal LRU (least recently used).

Content centric networking (CCN) has attracted a great deal of attention as a network that can efficiently deliver content. In CCN, content is delivered using the content name, instead of the host IP address, from cache memory implemented at routers. The nodes sending content are not explicitly indicated, and content is delivered from routers that have copies of content on the routes where the Interest packets are transmitted. Therefore, as a result of introducing CCN in ISP networks, the pattern of traffic exchanges among ISPs will change considerably. Customer ISPs normally pay a transit fee to transit ISPs based on the traffic volume transmitted on the transit links. Therefore, the introduction of CCN by ISPs will affect the profit of ISPs. CCN is introduced and operated by ISPs based on their business judgment, so it is important to estimate how CCN affects ISP profit to investigate the likelihood of CCN spreading among many ISPs. In this paper, we formalize the profit of ISPs when implementing CCN, assuming a hierarchical topology of ISPs in three levels and show that introducing CCN increases the profit of layer 2 and 3 ISPs, whereas introducing CCN decreases the profit of layer 1 ISPs. We also clarify that the effect of introducing CCN for the profit of ISPs is more remarkable as the cache capacity or the bias of content popularity increases.

Although the use of software-defined networking (SDN) enables routes of packets to be controlled with finer granularity (down to the individual flow level) by using traffic engineering (TE) and thereby enables better balancing of the link loads, the corresponding increase in the number of states that need to be managed at routers and controller is problematic in large-scale networks. Aggregating flows into macro flows and assigning routes by macro flow should be an effective approach to solving this problem. However, when macro flows are constructed as TE targets, variations of traffic rates in each macro flow should be minimized to improve route stability. We propose two methods for generating macro flows: one is based on a greedy algorithm that minimizes the variation in rates, and the other clusters micro flows with similar traffic variation patterns into groups and optimizes the traffic ratio of extracted from each cluster to aggregate into each macro flow. Evaluation using traffic demand matrixes for 48 hours of Internet2 traffic demonstrated that the proposed methods can reduce the number of TE targets to about 1/50 ~ 1/400 without degrading the link-load balancing effect of TE.

The use of content delivery services in which users pay a fee for each content delivery to the content provider (CP) is dramatically increasing. For Internet service providers (ISPs), the increased investment cost required to maintain stable quality for delivering rich content is a serious problem, and ISPs need to recover this cost from CPs because it is difficult to do so by increasing fees to users. However, CPs usually pay a transit fee in which the increase ratio diminishes as the volume of transmitted data increases, so the revenue collected by the ISPs is not sufficient to cover their investment cost. To address this problem, a content charge in which ISPs charge a fee for each content delivery to CPs would seem to be effective. However, it is anticipated that CPs will switch to another ISP if an ISP introduces a content charge, so introducing a content charge may not always increase the revenue of ISPs. Assuming a competitive environment of two ISPs in which one ISP introduces a content charge and each ISP can freely set its charging parameter, this paper models the relationship among CPs and two ISPs using a three-stage Stackelberg game and investigates the effect of a content charge by ISPs on the revenue of each player.

Monitoring flows at routers for flow analysis or deep packet inspection requires the monitors to update monitored flow information at the transmission line rate and needs to use highspeed memory such as SRAM. Therefore, it is difficult to measure all flows, and the monitors need to limit the monitoring target to a part of the flows. However, if monitoring targets are randomly selected, an identical flow will be monitored at multiple routers on its route, or a flow will not be monitored at any routers on its route. To maximize the number of flows monitored in the entire network, the monitors are required to select the monitoring targets while maintaining a balanced load among the monitors. In this paper, we propose an autonomous load balancing method where monitors exchange monitor load information with only adjacent monitors.

Peer-assisted content distribution technologies have been attracting attention. By using not only server resources but also the resources of end hosts (i.e., peers), we can reduce the offered load on servers as well as utilization of the access bandwidth of the servers. However, offered traffi to the network may increase because the traffi exchanged between peers passes across the network. Specificall, if individual peers send traffi disregarding underlay network topology and traffi conditions, the peer-assisted content distribution method may cause excessive traffi offered to the network and poor application performance. We thus investigated the impact of traffi caused by peer-assisted content distribution on the underlay network. We found that although peer-assisted content distribution disregarding underlay network topology causes 80-120% additional traffi compared with the optimal case, i.e., content distribution using cache servers allocated optimally in the network, using underlay network topology enables us to achieve almost the same efficien network resource utilization as the optimal case. We also found that the peer-assisted approach can adaptively cope with change in the traffi demand matrix because uploaders in the network are generated according to the demand matrix in a self-organizing manner. This is because peers that have downloaded the content become uploaders so many uploaders are generated in the area where a large number of content requests exist according to the traffi condition; therefore, the content delivery traffi can be localized.

Due to integrated high-speed networks accommodating various types of services and applications, the quality of service (QoS) requirements for those networks have also become diverse. The network resources are shared by the individual service traffic in the integrated network. Thus, the QoS of all the services may be degraded indiscriminately when the network becomes congested due to a sudden increase in traffic for a particular service if there is no traffic engineering taking into account each service's QoS requirement. To resolve this problem, we present a method of controlling individual service traffic by using an overlay network, which makes it possible to flexibly add various functionalities. The overlay network provides functionalities to control individual service traffic, such as constructing an overlay network topology for each service, calculating the optimal route for the service's QoS, and caching the content to reduce traffic. Specifically, we present a method of overlay routing that is based on the Hedge algorithm, an online learning algorithm to guarantee an upper bound in the difference from the optimal performance. We show the effectiveness of our overlay routing through simulation analysis for various network topologies.

In Video on Demand (VoD) services, the demand for content items greatly changes daily, so reducing the server load at the peak time is an important issue for ISPs to reduce the server cost. To achieve this goal, we proposed to reduce the server load by multicasting popular content items to all users independently of actual requests as well as providing on-demand unicast delivery. In this solution, however, the hit ratio of pre-distributed content items is small, and a large-capacity storage is required at set-top box (STB). We might be able to cope with this problem by limiting the number of pre-distributed content items or clustering users based on the history of viewing. We evaluate the effect of these techniques using actual VoD access log data. We clarify that the required storage capacity at STB can be halved while keeping the effect of server load reduction to about 80% by limiting pre-distributed content items, and user clustering is effective only when the cluster count is about two.

When a link or node fails, flows are detoured around the failed portion, so the hop count of flows and the link load could change dramatically as a result of the failure. As real-time traffic such as video or voice increases on the Internet, ISPs are required to provide stable quality as well as connectivity at failures. For ISPs, how to effectively improve the stability of these qualities at failures with the minimum investment cost is an important issue, and they need to effectively select a limited number of locations to add link facilities. In this paper, efficient design algorithms to select the locations for adding link facilities are proposed and their effectiveness is evaluated using the actual backbone networks of 36 commercial ISPs.

Designing network topologies requires simultaneous consideration of multiple criteria, such as network cost and reliability. So, the author applied the analytic hierarchy process, a way to make a rational decision considering multiple criteria, to network topology evaluation. However, the time required to construct the candidate topology set greatly increases as the network scale grows. This paper proposes to use of a multiagent system, in which each agent autonomously behaves to maximize each criterion, for generating a candidate topology set with high diversity within a practical time frame for large-scale networks.

Traffic caused by P2P services dominates a large part of traffic on the Internet and imposes significant loads on the Internet, so reducing P2P traffic within networks is an important issue for ISPs. In particular, a huge amount of traffic is transferred within backbone networks; therefore reducing P2P traffic is important for transit ISPs to improve the efficiency of network resource usage and reduce network capital cost. To reduce P2P traffic, it is effective for ISPs to implement cache devices at some router ports and reduce the hop length of P2P flows by delivering the required content from caches. However, the design problem of cache locations and capacities has not been well investigated, although the effect of caches strongly depends on the cache locations and capacities. We propose an optimum design method of cache capacity and location for minimizing the total amount of P2P traffic based on dynamic programming, assuming that transit ISPs provide caches at transit links to access ISP networks. We apply the proposed design method to 31 actual ISP backbone networks.

The demand for delivery services for large-sized content such as video has increased dramatically, and the use of cloud computing services in which users can use IT services via networks has also increased. To provide these services with high quality and high reliability, ISPs need to carefully design network topology and the positions of data centers. However, network topology and data center location strongly affect various evaluation criteria, such as cost, path length, and reliability; therefore, these criteria with different respective units need to be considered simultaneously when designing a data center network. The analytic hierarchy process (AHP) is a way to make a rational decision considering multiple criteria. This paper proposes to design data center networks by evaluating both network topology and data center locations simultaneously using AHP and also shows the numerical results of applying the proposed design method to the three areas of Japan, USA, and Europe.

Although video-on-demand (VoD) services are provided by many ISPs, the amount of content provided by each VoD service is one order smaller than that provided by rental video services, so the limited content count is one of the obstacles to widespread VoD services. To solve this problem, ISPs can form a coalition with other ISPs and use content owned by other ISPs. However, to form a coalition among multiple ISPs, ISPs need to rationally allocate the profit obtained by the coalition to convince all ISPs participating in the coalition. We propose using the Shapley value of the coalitional game as the rational allocation of profit. Assuming that all but one ISP has the same number of users or (and) the same number of rare content, we derive the Shapley value in closed form and clarify the influence of the numbers of users and rare content on the coalition. We also compare the Shapley value with three general allocation models and show that the Shapley value agrees with the allocation when the profit obtained by each content delivery is equally shared by two ISPs, one that accommodates the receiving user and the other that owns the delivered content.

Video streaming with HDTV or UHDV quality will be provided and widely demanded in the future. However, the transmission bit-rate of high-quality video streaming is quite large, so generated traffic flows will cause link congestion. Therefore, when providing streaming services of rich content, it is important to flatten the link utilization, i.e., reduce the maximum link utilization. To achieve this goal, parallel video streaming in which ISPs use multiple servers to deliver rich content is effective. However, the effect of parallel video streaming depends on the network topology and link capacities. In this paper, we investigate the impact of network topologies on the effect of parallel video streaming using 23 actual commercial ISP networks, when optimally designing server locations and optimally selecting servers.

When designing network topology, we need to simultaneously consider multiple criteria, such as network cost and reliability. Therefore, the author applied the analytic hierarchy process (AHP) to network topology evaluation. However, the time required to construct a candidate topology set drastically increases as the network scale grows. So, the author proposed an efficient method for constructing candidate topologies based on the binary partition method. However, the author applied this construction method to only small-scale networks. This paper numerically shows the limitation of this method when applied to large-scale networks and presents a realistic method for constructing candidate topologies for large-scale networks.

When a link or node failure occurs in a network, flows are detoured and therefore the hop counts of the flows increase. This increase could drastically deteriorate the quality of a network. The flow hop length stability as well as network connectivity is important for network reliability. We investigate a network design method that improves stability and connectivity of a network during a failure with a limited investment cost. We formalize a network design problem, prove that this problem is NP-complete, and propose an approximation algorithm. In addition, we evaluate the performance of the algorithm by using publicly available data for 39 backbone networks of commercial ISPs and networks generated by two well-known models. The results show that the proposed algorithm provides effective solutions from a practical viewpoint in sufficiently small computation time.

It is known that there exist Triangle Inequality Violations (TIVs) with respect to network Quality of Service (QoS) metrics such as latency between nodes in the Internet. This motivates the exploitation of QoS-aware routing overlays. To find an optimal overlay route, we would usually need to examine all the possible overlay routes. However, this requires both measuring QoS between all node pairs and investigating all the routes in the full-mesh overlay topology, which poses scalability problem in terms of both measurement cost and route calculation and dissemination cost. We thus propose a method of constructing a QoS overlay network that enables us to find a near optimal route in a cost-effective manner. Our idea is based on the finding that a small number of overlay nodes can provide the optimal routes for a large number of node pairs, which is obtained through measurement data analysis between PlanetLab nodes. Our overlay network has two layers where the upper-layer consists of such small number of overlay nodes that can provide the optimal routes while the lower-layer consists of the other overlay nodes. By allocating such overlay nodes at the upper-layer, we can provide better QoS routes for each node pair with high probability. We construct the overlay network topology where the upper-layer overlay nodes are connected in full-mesh manner while the lower-layer overlay nodes are not connected in full-mesh but only to upper-layer nodes. Through this structure, we can reduce measurement and route calculation costs. Using PlanetLab data, we show that our method can achieve almost the same performance as the optimal solution.

As a promising solution to manage the huge workload of large-scale VoD services, managed peer-assisted CDN systems, such as P4P has attracted attention. Although the approach works well in theory or in a controlled environment, to our best knowledge, there have been no general studies that address how actual peers can be incentivized in the wild Internet; thus, deployablity of the system with respect to incentives to users has been an open issue. With this background in mind, we propose a new business model that aims to make peer-assisted approaches more feasible. The key idea of the model is that users sell their idle resources back to ISPs. In other words, ISPs can leverage resources of cooperative users by giving them explicit incentives, e.g., virtual currency. We show the high-level framework of designing optimal incentive amount to users. We also analyze how incentives and other external factors affect the efficiency of the system through simulation. Finally, we discuss other fundamental factors that are essential for the deployability of managed peer-assisted model. We believe that the new business model and the insights obtained through this work are useful for assessing the practical design and deployment of managed peer-assisted CDNs.

Network topology seriously affects network cost, path length, distribution of link load, and reliability, for example, so we need to consider these multiple criteria with different units simultaneously when designing network topology. The analytic hierarchy process (AHP) is a way to make a rational decision considering multiple criteria. Using AHP, we can reflect the relative importance of each criterion on the evaluation result, so we have applied it to network topology evaluation. When evaluating network topologies using AHP, we need to construct the set of topology candidates prior to the evaluation. However, the time required to construct this set greatly increases as the network size grows. In this paper, we propose to apply a binary partition approach for constructing a topology candidate set with dramatically reduced calculation time. To reduce the calculation time, we introduce an upper limit for the total link length. Although the results of AHP are affected by introducing the upper limit of total link length, we show that desirable topologies are still selected in AHP.

We describe a method of adaptively controlling bandwidth allocation to flows for reducing the file transfer time of short flows without decreasing throughput of long-duration large flows. According to the rapid increase in Internet traffic volume, effective traffic engineering is increasingly required. Specifically, the traffic of long-duration large flows due to the use of peer-to-peer applications, for example, is a problem. Most conventional QoS controls allocate a fair-share bandwidth to each flow regardless of its duration. Thus, a long-duration large flow (such as a P2P flow) is allocated the same bandwidth as a short- duration flow (such as data from a Web page) in which the user is more sensitive to response time, i.e., file transfer time. As a result, long-duration large flows consume bandwidth over a long period and increase response times of short-duration flows, and conventional QoS methods do nothing to prevent this. In this paper, we therefore investigate a different approach, that is, a new form of bandwidth control that enables us to achieve better performance when handling short-duration flows while maintaining performance when handling long-duration flows. The basic idea is to tag packets of long-duration large flows according to traffic conditions and to give temporarily higher priority to non-tagged packets during network congestion. We also show the effectiveness of our method through simulation.

We present a method of detecting network anomalies, such as DDoS (distributed denial of service) attacks and flash crowds, automatically in real time. We evaluated this method using measured traffic data and found that it successfully differentiated suspicious traffic. In this paper, we focus on cyclic traffic, which has a daily and/or weekly cycle, and show that the differentiation accuracy is improved by utilizing such a cyclic tendency in anomaly detection. Our method differentiates suspicious traffic that has different statistical characteristics from normal traffic. At the same time, it learns about cyclic large- volume traffic, such as traffic for network operations, and finally considers it to be legitimate.

When a link or node failure occurs, flows are detoured around the failed portion, so the hop count of flows could change dramatically as a result of a failure. It is therefore important to evaluate the reliability of backbone networks from the viewpoint of flow hop length stability as well as network connectivity. In this paper, we analyze the possible primary causes of the degradation of reliability of the backbone networks of 36 commercial ISPs and present guidelines for making effective investments in link addition.

We propose a method of identifying anomalous traffic sources using flow statistics. We have investigated a way of detecting whether or not anomalies occur by observing the behavior of several time-series of flow statistics such as the number of flows. After detecting the occurrences of network anomalies, we need to identify the source of the anomalies. In this paper, we describe a method of identifying anomalous traffic sources. For this purpose, we apply data mining approaches such as the K-nearest neighbor method, naive Bayesian classifier, neural network, and support vector machine. We show how to use such approaches to identify anomalous traffic sources by using flow statistics. We also show evaluation results for the effectiveness of our approach using two measurement data sets.

The topology of a network seriously affects its cost, reliability, throughput, and traffic pattern, etc, so we need to simultaneously consider these multiple criteria, which have different units, when evaluating network topologies. Moreover, we need to reflect the relative importance of each criterion when evaluating the network topology. However, ordinary methods of network topology design consider only a single criterion. In this paper, we apply AHP (analytic hierarchy process), which has been widely used when making a rational decision with multiple criteria, to network topology design. AHP enables us to reflect the relative importance of each criterion on the results. In network topology design, we need to consider a huge number of candidates, and the differences in the weights among candidates are small. Therefore, the optimum topologies will be determined by just the weighted criteria, and other criteria will not be considered. Therefore, to emphasize the difference of the weights, we propose to use a linear-transformed value of each criterion when constructing weights in AHP.

Although an AWG-based single-hop WDM network is a desirable architecture to achieve an all-optical packet network in MANs, scalability is a problem. This problem can be solved by providing optical couplers between the AWG and other nodes and collecting multiple nodes. In this case, packet collisions at couplers will significantly increase the packet network delay. Therefore, the author proposed to install an autonomic packet collision avoidance mechanism in couplers. The installation of active devices at couplers degrades the network reliability. Therefore, we need to improve the reliability by providing a backup plane when applying this method to MANs. However, simply doubling all the couplers will largely increase the CAPEX. The probability that multiple collision-avoidance mechanisms breakdown simultaneously is very small, so this paper proposes to construct the backup plane by sharing couplers among nodes accommodated in different couplers on the primary plane.

The topology of a network seriously affects its cost, reliability, throughput, and traffic pattern, etc, so we need to simultaneously consider these multiple criteria, which have different units, when evaluating network topologies. However, ordinary methods of network topology design considered only a single criterion. DEA (data envelopment analysis) enables us to simultaneously evaluate multiple criteria, and it has been widely used when evaluating the efficiency of a business or a project. DEA derives the optimality of each candidate by emphasizing strong criteria in its evaluation. In this paper, we apply DEA to network topology design, and we numerically show that it enables us to effectively focus on a small number of desirable topology candidates.

We investigate the detection accuracy of network anomalies when we use flow statistics obtained through packet sampling. We have already shown, through a case study based on measurement data, that network anomalies generating a huge number of small flows, such as network scans or SYN flooding, become hard to detect when we perform packet sampling. In this paper, we first develop an analytical model that enables us to quantitatively evaluate the effect of packet sampling on the detection accuracy and then investigate why detection accuracy worsens when the packet sampling rate decreases. In addition, we show that, even with a low sampling rate, spatially partitioning the monitored traffic into groups makes it possible to increase the detection accuracy. We also develop a method of determining an appropriate number of partitioned groups and show its effectiveness.

Traffic flow measurement is essential to implement QoS control in the Internet. Flow monitoring system collects and stores sampled flow states in a flow table (FT) and the entries are renewed at every packet sampling. Entries in the FT are checked and removed when no packets are sampled within a predetermined timeout. We propose an efficient timeout checking mechanism based on checking a small number of entries selected randomly from the FT. Our proposed method aims to reduce the number of memory accesses dramatically and keep the memory size small. We evaluate our method and compare with the conventional method that checks all flow entries of the FT periodically. Our simulation and comparison results show that our method is able to reduce the number of memory access at a factor of 1000 with a small increase in memory size of approximately 10 percent.

Abusive traffic caused by worms is increasing severely in the Internet. In many cases, worm-infected hosts generate a huge number of flows of small size during a short time. To suppress the abusive traffic and prevent worms from spreading, identifying these "superspreaders" as soon as possible and coping with them, e.g, disconnecting them from the network, is important. This paper proposes a simple and adaptive method of identifying superspreaders by flow sampling. By satisfying the given memory size and the requirement for the processing time, the proposed method can adaptively optimize parameters according to changes in traffic patterns.

Load balancing among multiple mirror servers located at distributed positions is a key function in content distribution networks. For bandwidth allocated services, we consider how to select a suitable server from several candidates containing the same content at the time of a request. We propose limiting the candidates in advance and selecting a server from the limited set of servers in a round-robin fashion. The server sets that minimize the variance of the link load are derived using a greedy method for a given network topology and service demand. Through numerical evaluation, we demonstrate that the proposed method is superior to previous methods.

A method of estimating TCP flow-rates of sampled flows through packet sampling is described in this paper. We use sequence numbers of sampled packets, which make it possible to improve markedly the accuracy of estimating the flow rates. Using an analytical model, we investigate how to set parameters such as packet sampling probability used in this method of estimation. As a remarkable result, we show that the estimation accuracy improves as the sampling probability decreases. Using measured data, we also show that this method gives accurate estimations. We also show that this estimation method enables us to detect performance degradation at the TCP flow level.

Broadcast data delivery is attractive for large-size data distribution where a large user community is connected to a server through a network. Because the number of channels in a broadcast network is limited and a large number of users share these channels, an immediate delivery is difficult to achieve. Thus, it is expected that users request the service in advance, and it is important to guarantee the expected waiting time at the time of request. In this paper, we propose a method dividing all contents into several groups according to the contents popularity and assigns FIFO to each group. The proposed method can guarantee the response time for each user at his request. The average guaranteed response time in the proposed method is lower than those in FIFO (in high load) and a fixed scheduling method (in low load).

Abstract — Unfairness among best-effort flows is a serious problem on the Internet. In particular, UDP flows or unresponsive flows that do not obey the TCP flow control mechanism can consume a large share of the available bandwidth. High-rate flows seriously affect other flows, so it is important to identify them and limit their throughput by selectively dropping their packets. As link transmission capacity increases and the number of active flows increases, however, capturing all packet information becomes more difficult. In this paper, we propose a novel method of identifying high-rate flows by using sampled packets. The proposed method simply identifies flows from which Y packets are sampled without timeout. The identification principle is very simple and the implementation is easy. We derive the identification probability for flows with arbitrary flow rates and obtain an identification curve that clearly demonstrates the accuracy of identification. The characteristics of this method are determined by three parameters: the identification threshold Y, the timeout coefficient K, and the sampling interval N. To match the experimental identification probability to the theoretical one and to simplify the identification mechanism, we should set K to the maximum allowable value. Although increasing Y improves the identification accuracy, both the required memory size and the processing power grow as Y increases. Numerical evaluation using an actual packet trace demonstrated that the proposed method achieves very high identification accuracy with a much simpler mechanism than that of previously proposed methods. I.

Unfairness among best-effort flows is a serious problem in the Internet. In particular, UDP flows or unresponsive flows that do not obey the TCP flow control mechanism or normal TCP flows with small RTT can consume a large share of the available bandwidth. Since they will seriously affect other flows, it is important to identify these high-rate flows and limit their throughput by selectively dropping their packets. In this paper, we propose a novel method that identifies high-rate flows using sampled packets. By reducing the timeout length for holding the flow state, the proposed method accurately identifies high-rate flows while using only a small amount of memory. We derive the identification probability for flows with arbitrary rates and obtain an identification curve that clearly reveals the identification accuracy.