Efficient Network Resource Management Research Articles

시그니쳐 매칭 유형 분류를 통한 트래픽 분석 시스템의 처리 속도 향상

응용 레벨 트래픽 분석은 네트워크의 효율적인 운영과 안정적인 서비스 제공을 위한 필수적인 요소이다. 응용 레벨 트래픽 분석을 위한 다양한 방법이 존재하지만 분류의 정확성, 분석률, 실용성을 고려했을 때 페이로드 시그니쳐 기반 분석 방법이 가장 높은 성능을 보인다. 하지만 페이로드 시그니쳐 기반 분석 방법은 다른 방법론에 비해 처리속도가 느리다는 단점이 있다. 본 논문에서는 각 시그니쳐가 페이로드에 매칭 되는 범위와 패킷의 순서 그리고 방향성과 같은 Offset value을 자동으로 추출하고 활용하여 시그니쳐를 매칭 유형별로 분류한다. 유형별로 분류된 시그니쳐에 최적화된 탐색범위를 지정하여 탐색범위를 최적화함으로써 페이로드 시그니쳐 기반 분석 방법의 처리 속도를 향상 시키는 방법을 제안한다. The traffic identification is a preliminary and essential step for stable network service provision and efficient network resource management. While a number of identification methods have been introduced in literature, the payload signature-based identification method shows the highest performance in terms of accuracy, completeness, and practicality. However, the payload signature-based method's processing speed is much slower than other identification method such as header-based and statistical methods. In this paper, we first classifies signatures by matching type based on range, order, and direction of packet in a flow which was automatically extracted. By using this classification, we suggest a novel method to improve processing speed of payload signature-based identification by reducing searching space.

Ultra-low-latency ubiquitous connections in heterogeneous cloud radio access networks

Although heterogeneous cloud radio access networks (H-CRAN) have emerged for efficient network management, resource management, and throughput enhancement, new technical challenges remain to support the urgent need to achieve full automation for ultra-low-latency connections. To simultaneously accommodate such diverse technical requirements in the HCRAN, in this article, a new paradigm of HCRAN design to significantly reduce latency is presented. From the refined radio access in the air interface, new methodology of radio resource optimization in the system architecture, and intelligent routing/paging in the backhaul, the provided foundation suggests open-loop communications and other new directions to facilitate state-of-the-art practices of the H-CRAN.

응용 트래픽의 지역성을 이용한 페이로드 시그니쳐 기반 트래픽 분석 시스템의 성능 향상

응용 레벨 트래픽 분류는 안정적인 네트워크 운영과 자원 관리를 위해서 필수적으로 요구된다. 트래픽분류에 있어서 페이로드 시그니쳐 기반 응용 레벨 트래픽 분류 방법은 고속 링크의 트래픽을 실시간으로 처리하는 과정에서 헤더 정보 및 통계 정보 이용 방법론에 비해 상대적으로 높은 부하를 발생시키며 처리 속도가 느린 단점을 갖는다. 본 논문에서는 페이로드 시그니쳐 기반 트래픽 분류 시스템의 처리 속도를 향상 위하여 응용 트래픽의 지역성을 이용한 서버 IP, Port캐쉬 기반 트래픽 분석 시스템을 제안한다. 제안하는 방법을 학내 망의 실제 트래픽에 적용하여 최대 10배 이상의 처리 속도 향상과 10% 이상의 플로우 분석률을 향상 시킬 수 있었다. The traffic classification is a preliminary and essential step for stable network service provision and efficient network resource management. However, the payload signature-based method has a significant drawback in high-speed network environment that the processing speed is much slower than other method such as header-based and statistical methods. In this paper, We propose the server IP, Port cache-based traffic classification method using application traffic locality to improve the processing speed of traffic classification. The suggested method achieved about 10 folds improvement in processing speed and 10% improvement in completeness over the payload-based classification system.

CPS: Operating system architecture for efficient network resource management with control-theoretic packet scheduler

The efficient network resource management is one of the important topics in a real-time system. In this paper, we present a practical network resource management framework, control-theoretic packet scheduler (CPS) system. Using our framework, an operating system can schedule both input and output streams accurately and efficiently. Our framework adopts very portable feedback control theory for efficiency and accuracy. The CPS system is able to operate independent of the internal network protocol state, and it is designed to schedule packet streams in fine-grained time intervals to meet the resource requirement. This approach simplifies the design of the CPS system, and leads us to obtain the intended output bandwidth. We implemented our prototype system in Linux, and measured the performance of the network resource management system under various network QoS constraints. The distinctive features of our principles are as follows: It is robust and accurate, and its operation is independent of internal network protocols.

응용 레벨 트래픽 분류를 위한 시그니쳐 생성 및 갱신 시스템 개발

네트워크 트래픽 모니터링과 분석은 엔터프라이즈 네트워크의 효율적인 운영과 안정적 서비스를 제공하기 위한 필수적인 요소이다. 응용 레벨 트래픽의 분석을 위한 다양한 방법이 존재하지만 분류의 정확성, 분석률, 실용성을 고려했을 때 페이로드 시그니쳐 기반의 분석 방법은 가장 높은 성능을 보인다. 하지만 페이로드 시그니쳐를 수작업으로 추출하는 과정은 응용프로그램 및 응용 프로토콜에 대한 선행적인 분석이 필요하기 때문에 많은 시간과 인력이 요구된다. 또한 응용프로그램의 통합, 변경, 출현은 시그니쳐의 유지 및 관리에 대한 복잡성을 증대시킨다. 따라서 본 논문에서는 응용프로그램의 페이로드 시그니쳐 생성 과정의 단점을 보완할 수 있는 시그니쳐 자동 생성 시스템을 제안하여 시그니쳐 생성 효율을 향상시키며, 응용프로그램의 변화, 출현에 유연하게 대처할 수 있는 페이로드 시그니쳐 갱신 시스템을 구축하여 지속적으로 시그니쳐 유지, 관리가 가능하도록 하였다. 또한 학내망에 적용하여 제안한 시스템의 실용성을 증명하였다. The traffic classification is a preliminary but essentialstep for stable network service provision and efficient network resource management. While various classification methods have been introduced in literature, the payload signature-based classification is accepted to give the highest performance in terms of accuracy, completeness, and practicality. However, the collection and maintenance of up-to-date signatures is very difficult and time consuming process to cope with the dynamics of Internet traffic over time. In this paper, We propose an automatic payload signature generation mechanism which reduces the time for signature generation and increases the granularity of signatures. Furthermore, We describe a signature update system to keep the latest signatures over time. By experiments with our campus network traffic we proved the feasibility of our mechanism.

Directed broadcast with overhearing for sensor networks

The efficient management of scarce network resources, including energy and bandwidth, represents a central challenge for wireless sensor networks. The current trend in resource management relies on the introduction of control mechanisms, such as control message exchanges, node-specific addressing, and storage of partial network state information. These mechanisms typically incur communication and processing overhead that does not scale well for larger or denser networks. Instead of introducing control mechanisms for network resource management, this article proposes and evaluates a Directed Broadcast with Overhearing (DBO) approach for sensor networks that combines directed broadcast at the network layer with CSMA and packet overhearing at the MAC layer. Through avoidance of control messaging and exchange of network state information, DBO trades off limited packet duplication overhead for control messaging overhead. This article introduces an analytical model that provides the basis for DBO evaluation and for analysis of the approach's transient packet retransmissions, route convergence, and energy consumption in the average and worst cases. We also present the model implementation details and the simulation experiments that explore the suitability of DBO for networks of different sizes with three different radio models that vary the width of grey regions, and we compare DBO's energy consumption against conventional unicast beacon-based and snooping-based routing protocols. The results indicate that that DBO's route convergence requires an average of five hops for ideal radio reception, seven hops for narrow grey regions, and twelve hops for wide grey regions. These results confirm that DBO shifts energy consumption from critical nodes near the base station to nodes near the source. The overall energy consumption of limited packet duplication overhead with DBO compared to unicast routing shrinks for medium- to large-size networks, rendering it more favorable than conventional communication approaches for large and dense sensor networks.

A Review of Policy-Based Resource and Admission Control Functions in Evolving Access and Next Generation Networks

The IP Multimedia Subsystem (IMS) adopted in the core of Next Generation Networks (NGNs) promises to make network management easier by separating the control and the transport planes. Therefore, an interface between applications and the underlying transport network has been defined that offers a dynamic and efficient management of network resources based on a policy-based resource control engine. The resulting resource management framework enables the delivery of both the existing carrier grade existing and the next generation Quality of Service (QoS) sensitive services across operator-controlled networks using heterogeneous transport technologies. This review sheds some light into the policy control layer concept and the extended nomenclature introduced by current standardization works. The approaches of international standards development organizations, such as the Third Generation Partnership Project (3GPP), the International Telecommunication Union (ITU), the European Telecommunications Standards Institute (ETSI), the WiMAX Forum, and CableLabs are reviewed and compared with each other revealing the common architectural trend. Challenges and works in progress of NGN resource management towards Fixed and Mobile Convergence (FMC) are discussed as well.

Virtual network approach to scalable IP service deployment and efficient resource management

As the Internet evolves into a global all-service communication infrastructure, a key consideration is providing quality of service guarantees over IP with efficient resource utilization in a scalable, flexible, and automatic way. In this article we present a virtual network (VN) based architecture for scalable IP service deployment and efficient network resource management. Particularly considering a DiffServ/MPLS III transport network supporting multiple VNs, we propose a dynamic approach for efficient bandwidth sharing among VNs. The bandwidth sharing is service-level-agreement-based; the spare capacity in underloaded VNs is adaptively and efficiently utilized, and SLA compliance for all the VNs involved is always guaranteed.

Enhanced path prediction for network resource management in wireless LANs

Path prediction is currently being considered for use in the context of mobile and wireless computing toward more efficient network resource management schemes. Path prediction allows the network and services to further enhance the quality of service levels the user enjoys. Such mechanisms are mostly meaningful in infrastructures like wireless LANs. In this article we present a path prediction algorithm that exploits the machine learning algorithm of learning automata. The decision of the learning automaton is driven by the movement patterns of a single user but is also affected by the aggregated patterns demonstrated by all users. Simulations of the algorithm, performed using the realistic mobility pattern generator, show increased prediction accuracy.

Performance issues of bandwidth management in ATM networks

AbstractIn our days, efficient management of the available network resources becomes a critical issue, both from a functional point of view (so that users can be provided with the bandwidth they need), and an economical point of view (so that carriers can satisfactorily and efficiently serve as many customers as possible and at the same time increase their revenue). In this paper we consider a bandwidth control scheme (i.e. managed bandwidth service) for an ATM network infrastructure which is applied to the Greek research and technology network (GRNET). We present some methods that we have tested (in a simulation setting) in order to increase the efficiency of the system and the utilization of the available bandwidth. More specifically, we consider a bandwidth‐resizing algorithm for virtual paths, in order to keep the allocated bandwidth very close to the bandwidth actually used. This leads to an increased number of accepted requests and better network utilization. We, also, use the simulation results in order to get an estimation of the effective bandwidth for VBR paths that can be used in call admission. Finally, we consider a semi‐offline scheme where requests are gathered and considered for acceptance in regular intervals. Simulation results show an increase in the utilization of resources. As a further improvement, we allow connections to be allocated a little before or after the time initially requested. This leads to further improvement in network utilization. All the improvement schemes were tested with the ATM‐TN simulator and the results look promising. Copyright © 2003 John Wiley & Sons, Ltd.

An Adaptive Fuzzy Clustering Technique for Traffic Prediction of Packet-switched Networks

Traffic prediction is significant to QoS design because it assists efficient management of network resources to improve the reliability and performance of the next generation Internet. The unavoidable traffic variation caused by diverse Internet services complicates traffic prediction, particularly in a multi-hop network. To simplify the complicated statistical analysis used in traditional approaches, an adaptive traffic prediction approach featuring robustness, high accuracy and high adaptability is proposed in this paper. The proposed approach bases on a novel fuzzy clustering algorithm to generalize and unveil the hidden structure of traffic patterns. The unveiled structure represents the characteristics of the target traffic. Therefore, it can be referenced to predict traffic in a limited time period by fuzzy matching. To track the variation of target traffic, the proposed approach adopts an incremental and dynamic on-line clustering procedure so that the prediction can maintain high accuracy under traffic variation. To verify the performance of the proposed approach and investigate its properties, the periodical, Poisson and real video traffic patterns have been used to experiment. The experimental results showed an excellent performance of the developed adaptive predictor. The prediction errors, in average, are near 2.2%, 13.6% and 7.62% for periodical, Poisson and real video traffics, respectively.

TeleMIP: telecommunications-enhanced mobile IP architecture for fast intradomain mobility

This article first surveys existing protocols for supporting IP mobility and then proposes an extension to the mobile IP architecture, called TeleMIP. Our architecture attempts to achieve smaller handoff latency by localizing the scope of most location update messages within an administrative domain or a geographical region. TeleMIP is intended for use in evolving third-generation wireless networks, and introduces a new logical entity, called the mobility agent, which provides a mobile node with a stable point of attachment in a foreign network. While the MA is functionally similar to conventional foreign agents, it is located at a higher level in the network hierarchy than the subnet-specific FAs. Location updates for intradomain mobility are localized only up to the MA; transmission of global location updates are necessary only when the mobile changes administrative domains and/or geographical regions. By permitting the use of private or locally scoped addresses for handling intradomain mobility, TeleMIP allows efficient use of public address space. Also, by reducing the frequency of global update messages, our architecture overcomes several drawbacks of existing protocols, such as large latencies in location updates, higher likelihood of loss of binding update messages, and loss of inflight packets, and thus provides better mobility support for real-time services and applications. The dynamic creation of mobility agents (in TeleMLP) permits the use of load balancing schemes for the efficient management of network resources.

Realizing the MPEG-4 multimedia delivery framework

MPEG-4 is an emerging international standard based on the encoding of audiovisual data using object description techniques. This novel approach to encoding allows MPEG-4-based multimedia applications to dynamically compose complex scenes from one or more elementary stream objects. These stream objects are carried within potentially hundreds of channels. As a result, the underlying transport system must support the rapid creation and release of such channels with QoS guarantees. In addition to defining the format, structure and rules of composition of these objects, MPEG-4 also specifies a general application and transport delivery framework called the delivery multimedia integration framework (DMIF). DMIF's main purpose is to hide the details of the transport network from the user, as well as to ensure signaling and transport interoperability between end-systems. We describe the first reference implementation of the MPEG-4 DMIF standard for the ISO, including a full specification of the DMIF application interface. The DAI offers the required functionality for realizing multimedia applications without any concerns regarding the selected communication protocol. The realization of DMIF is greatly facilitated by xbind, a broadband kernel that provides connection management, routing, QoS mapping, and intelligent selection of suitable transport protocol stacks. The resulting system, called XDMIF, supports different QoS requirements, and the rapid creation and release of numerous transport channels, along with efficient management of network resources.