Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488139
N. Markovich, A. Undheim, P. Emstad
Estimation of bit loss is essential to assess the Quality of Service of a video stream. This paper is devoted to the estimation of the bit loss when the bitrate exceeds the capacity of the bottleneck link. The video stream under study is encoded using the explicit slice-based H.264/AVC encoding scheme which is recently developed. This scheme reduces the burstiness of regular H.264/AVC encoded video by removing the traditional GoP structure. Instead, a repetitive combination of intracoded and predicted slices is employed, thereby introducing a specific dependence structure in the video data. Due to the high variability in the underlying video data, different types of classification for the data are considered, and used for the analysis. We consider a bufferless model of the communication system. The frame sizes are measured and the losses that arise due to the exceedances of the frame sizes over a threshold derived from the channel capacity are evaluated. The high quantiles of the losses are estimated.
{"title":"Slice-based VBR video traffic-estimation of link loss by exceedance","authors":"N. Markovich, A. Undheim, P. Emstad","doi":"10.1109/ITNEWS.2008.4488139","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488139","url":null,"abstract":"Estimation of bit loss is essential to assess the Quality of Service of a video stream. This paper is devoted to the estimation of the bit loss when the bitrate exceeds the capacity of the bottleneck link. The video stream under study is encoded using the explicit slice-based H.264/AVC encoding scheme which is recently developed. This scheme reduces the burstiness of regular H.264/AVC encoded video by removing the traditional GoP structure. Instead, a repetitive combination of intracoded and predicted slices is employed, thereby introducing a specific dependence structure in the video data. Due to the high variability in the underlying video data, different types of classification for the data are considered, and used for the analysis. We consider a bufferless model of the communication system. The frame sizes are measured and the losses that arise due to the exceedances of the frame sizes over a threshold derived from the channel capacity are evaluated. The high quantiles of the losses are estimated.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121042772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488153
V. Eramo, M. Listanti, A. Cianfrani
In this paper we analyze intra-domain routing protocols improvements to support new features required by real time services. In particular we introduce OSPF Fast Convergence and highlights the advantage of using an incremental algorithm instead of Djikstra one to compute the shortest paths. Then we propose a new multi-path incremental algorithm that we have implemented in OSPF code of Quagga open-source routing software. Analyzing an index characterizing OSPF performance we compare our algorithm with an incremental algorithm not supporting multi-path and demonstrate that, even if multi-path support, the reconfiguration times are really similar; moreover,in some cases, our algorithm performs better, especially in a link failure scenario.
{"title":"Multi-path OSPF performance of a software router in a link failure scenario","authors":"V. Eramo, M. Listanti, A. Cianfrani","doi":"10.1109/ITNEWS.2008.4488153","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488153","url":null,"abstract":"In this paper we analyze intra-domain routing protocols improvements to support new features required by real time services. In particular we introduce OSPF Fast Convergence and highlights the advantage of using an incremental algorithm instead of Djikstra one to compute the shortest paths. Then we propose a new multi-path incremental algorithm that we have implemented in OSPF code of Quagga open-source routing software. Analyzing an index characterizing OSPF performance we compare our algorithm with an incremental algorithm not supporting multi-path and demonstrate that, even if multi-path support, the reconfiguration times are really similar; moreover,in some cases, our algorithm performs better, especially in a link failure scenario.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121571322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488138
P. Giacomazzi, G. Saddemi
In this paper, we propose an analytical method for the resource allocation and admission control of traffic flows with statistical Quality-of-Service (QoS) guarantees in a Static Priority service discipline. We assume that at the network ingress each traffic flow is controlled by a Token Bucket, and the statistical QoS targets for each service class are expressed in terms of delay bound and delay violation probability. Under these assumptions, we solve analytically the resource allocation problem by deriving the closed-form expression of the minimum capacity to be allocated in the network in order to guarantee concurrently the QoS of all traffic flows. Moreover, we provide the solution of the admission control problem, by deriving the expression of the maximum number of flows that is possible to accept, in all priority levels, known the link capacity, with statistical QoS constraints on the delay.
{"title":"Resource allocation and admission control in static priority schedulers with statistical guarantees","authors":"P. Giacomazzi, G. Saddemi","doi":"10.1109/ITNEWS.2008.4488138","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488138","url":null,"abstract":"In this paper, we propose an analytical method for the resource allocation and admission control of traffic flows with statistical Quality-of-Service (QoS) guarantees in a Static Priority service discipline. We assume that at the network ingress each traffic flow is controlled by a Token Bucket, and the statistical QoS targets for each service class are expressed in terms of delay bound and delay violation probability. Under these assumptions, we solve analytically the resource allocation problem by deriving the closed-form expression of the minimum capacity to be allocated in the network in order to guarantee concurrently the QoS of all traffic flows. Moreover, we provide the solution of the admission control problem, by deriving the expression of the maximum number of flows that is possible to accept, in all priority levels, known the link capacity, with statistical QoS constraints on the delay.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"2013 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127304189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488131
H. Martikainen, A. Sayenko, O. Alanen, V. Tykhomyrov
In the IEEE 802.16 the number of errors and the MAC PDU size have an impact on the performance of the network. We present a way to estimate the optimal PDU size and we run a number of simulation scenarios to study these parameters and how they impact on the performance of application protocols. The simulation results reveal that the channel bit error rate has a major impact on the optimal PDU size in the IEEE 802.16 networks. Also, the ARQ block rearrangement influences the performance.
{"title":"Optimal MAC PDU Size in IEEE 802.16","authors":"H. Martikainen, A. Sayenko, O. Alanen, V. Tykhomyrov","doi":"10.1109/ITNEWS.2008.4488131","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488131","url":null,"abstract":"In the IEEE 802.16 the number of errors and the MAC PDU size have an impact on the performance of the network. We present a way to estimate the optimal PDU size and we run a number of simulation scenarios to study these parameters and how they impact on the performance of application protocols. The simulation results reveal that the channel bit error rate has a major impact on the optimal PDU size in the IEEE 802.16 networks. Also, the ARQ block rearrangement influences the performance.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131394128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488129
A. Valletta, A. Todini, A. Baiocchi
The problem of unfairness among competing flows in TCP multihop wireless networks is well known, and has often been traced to the misbehaviour of dynamic routing protocols in case of link failure. We show, through simulations, that the competition between TCP flows in a multihop wireless network leads to an unfair outcome, independently of the routing protocol involved. We focus on the impact of hidden node collisions on the performance of multi-hop TCP flows, and trace the source of the problem to collisions between RTS frames belonging to the starved flow and long data frames sent by the active flow. Such collisions lead to a rapid increase of the MAC contention window at the TCP sender of the starved flow; this, in conjunction with TCP's congestion control algorithm, allows the running flow to hold the channel for a long time, until the situation is reversed. We also show that a simple multi-channel MAC can ease the competition among flows.
{"title":"On the unfairness among TCP flows in IEEE 802.11 multi-hop ad-hoc networks","authors":"A. Valletta, A. Todini, A. Baiocchi","doi":"10.1109/ITNEWS.2008.4488129","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488129","url":null,"abstract":"The problem of unfairness among competing flows in TCP multihop wireless networks is well known, and has often been traced to the misbehaviour of dynamic routing protocols in case of link failure. We show, through simulations, that the competition between TCP flows in a multihop wireless network leads to an unfair outcome, independently of the routing protocol involved. We focus on the impact of hidden node collisions on the performance of multi-hop TCP flows, and trace the source of the problem to collisions between RTS frames belonging to the starved flow and long data frames sent by the active flow. Such collisions lead to a rapid increase of the MAC contention window at the TCP sender of the starved flow; this, in conjunction with TCP's congestion control algorithm, allows the running flow to hold the channel for a long time, until the situation is reversed. We also show that a simple multi-channel MAC can ease the competition among flows.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132280954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488126
G. Maier, L. Colombo, D. Costantino, A. Pattavina, P. Szegedi
National Research and Education Networks (NERNs) are investigating the option to migrate their infrastructure from IP over static transport network to IP over control- plane based transport network, possibly deploying GMPLS or ASON technology. In the NREN-participated MUPBED project, quality of provisioning of bandwidth on-demand services has been studied. In this paper the main results are presented showing that without an automatic control-plane, connection provisioning is impaired by setup and release delays. In managed- networks, performance under dynamic traffic may be severely limited if an insufficient number of operators are active in the network operations center. The relation between number of operators and network performance is proved by theoretical models and simulations. The reduction of network-configuration workload achievable by an ASON/GMPLS control plane may be an important motivation for NRENs because of their specific operational expenditure model.
{"title":"Quality of provisioning as an OPEX-related issue in research networks","authors":"G. Maier, L. Colombo, D. Costantino, A. Pattavina, P. Szegedi","doi":"10.1109/ITNEWS.2008.4488126","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488126","url":null,"abstract":"National Research and Education Networks (NERNs) are investigating the option to migrate their infrastructure from IP over static transport network to IP over control- plane based transport network, possibly deploying GMPLS or ASON technology. In the NREN-participated MUPBED project, quality of provisioning of bandwidth on-demand services has been studied. In this paper the main results are presented showing that without an automatic control-plane, connection provisioning is impaired by setup and release delays. In managed- networks, performance under dynamic traffic may be severely limited if an insufficient number of operators are active in the network operations center. The relation between number of operators and network performance is proved by theoretical models and simulations. The reduction of network-configuration workload achievable by an ASON/GMPLS control plane may be an important motivation for NRENs because of their specific operational expenditure model.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116347215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488156
F. Licandro, D. Marchese, M. Masi, A. Russo, G. Schembra
Multipoint distribution of real-time video constitutes a challenge for the research in the current Internet, mainly due to the poor dissemination of IP Multicast. The target of this paper is to define a low-cost flexible multipoint communication platform for the users of video sessions, which scales for a large number of communication groups with a small number of members. This paper proposes a software-based platform to provide multipoint communications in a heterogeneous unicast/multicast environment. It is constituted by a number of S-MCUs (software MCUs). Each S-MCU runs on a low-cost single board computer (SBC) and is implemented like a software router by using Click. Design, implementation and performance issues are discussed by a comparative numerical analysis at both packet level and application level.
{"title":"Low-cost SBC cards to provide multipoint to multimedia communications sessions","authors":"F. Licandro, D. Marchese, M. Masi, A. Russo, G. Schembra","doi":"10.1109/ITNEWS.2008.4488156","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488156","url":null,"abstract":"Multipoint distribution of real-time video constitutes a challenge for the research in the current Internet, mainly due to the poor dissemination of IP Multicast. The target of this paper is to define a low-cost flexible multipoint communication platform for the users of video sessions, which scales for a large number of communication groups with a small number of members. This paper proposes a software-based platform to provide multipoint communications in a heterogeneous unicast/multicast environment. It is constituted by a number of S-MCUs (software MCUs). Each S-MCU runs on a low-cost single board computer (SBC) and is implemented like a software router by using Click. Design, implementation and performance issues are discussed by a comparative numerical analysis at both packet level and application level.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117234204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488160
H. Badis
Many real-time applications, such as video conferencing, require the transmission of flows from a sender to multiple receivers subject to Quality of Service (QoS) delivery constraints (e.g. bounded delay). Traditional multicast routing algorithms used for solving the Sterner tree problem cannot be used in this scenario, because QoS constraints on links are not considered. Moreover, with QoS metrics consideration, the problem of computing optimal constrained multicast tree is known as NP-complete. In this paper, we propose a new QoS-aware Multicast Overlay Spanning Tree (QMOST) protocol to support multiple-metric routing criteria in mobile ad hoc networks. It computes low cost trees with multiple-metric bounded paths in polynomial time using an efficient source- based optimization heuristic. Unlike common multicast protocols where data packets are transmitted in a broadcast mode, in QMOST data packets are encapsulated in unicast packets before being forwarded to the overlay tunnels. Multicasting is achieved when each node forwards multicast packets to its successors in the overlay tree. QMOST must be used in conjunction with a unicast QoS link state protocol (QOLSR) and can be integrated in a QoS model. This later offers the needed service guarantee using distributed admission control to check if the multicast QoS flows are admissible, resource reservation mechanism and congestion control. Simulation results show that our QMOST protocol achieves better performance than the QOLSR protocol.
{"title":"A QoS-aware multicast overlay spanning tree protocol for multimedia applications in MANETs","authors":"H. Badis","doi":"10.1109/ITNEWS.2008.4488160","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488160","url":null,"abstract":"Many real-time applications, such as video conferencing, require the transmission of flows from a sender to multiple receivers subject to Quality of Service (QoS) delivery constraints (e.g. bounded delay). Traditional multicast routing algorithms used for solving the Sterner tree problem cannot be used in this scenario, because QoS constraints on links are not considered. Moreover, with QoS metrics consideration, the problem of computing optimal constrained multicast tree is known as NP-complete. In this paper, we propose a new QoS-aware Multicast Overlay Spanning Tree (QMOST) protocol to support multiple-metric routing criteria in mobile ad hoc networks. It computes low cost trees with multiple-metric bounded paths in polynomial time using an efficient source- based optimization heuristic. Unlike common multicast protocols where data packets are transmitted in a broadcast mode, in QMOST data packets are encapsulated in unicast packets before being forwarded to the overlay tunnels. Multicasting is achieved when each node forwards multicast packets to its successors in the overlay tree. QMOST must be used in conjunction with a unicast QoS link state protocol (QOLSR) and can be integrated in a QoS model. This later offers the needed service guarantee using distributed admission control to check if the multicast QoS flows are admissible, resource reservation mechanism and congestion control. Simulation results show that our QMOST protocol achieves better performance than the QOLSR protocol.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131159432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488133
V. Eramo, A. Germoni, C. Raffaelli, M. Savi
A Multi-fiber all-optical switch which employs wavelength converters with fixed input wavelength is presented. Being the converters tuned over a fixed input wavelength, each one is shared by channels on the same wavelength, so the resulting architecture is called shared-per-wavelength. Here the space switching matrices are simpler with respect to similar switching architectures proposed in the past (i.e. shared-per-node ones) leading to easier and less costly implementation. Moreover, fixed- input/tunable-output converters are expected to be less costly than tunable-input/tunable-output ones. A scheduling algorithm to forward optical packets in synchronous scenario and an analytical model to evaluate performance of the shared-per- wavelength architecture are presented. The analytical model is validated against simulation and the results show good accuracy in most cases of interest for optical switch design.
{"title":"Performance analysis of multi-fiber all-optical switches employing fixed-input/tunable-output wavelength converters","authors":"V. Eramo, A. Germoni, C. Raffaelli, M. Savi","doi":"10.1109/ITNEWS.2008.4488133","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488133","url":null,"abstract":"A Multi-fiber all-optical switch which employs wavelength converters with fixed input wavelength is presented. Being the converters tuned over a fixed input wavelength, each one is shared by channels on the same wavelength, so the resulting architecture is called shared-per-wavelength. Here the space switching matrices are simpler with respect to similar switching architectures proposed in the past (i.e. shared-per-node ones) leading to easier and less costly implementation. Moreover, fixed- input/tunable-output converters are expected to be less costly than tunable-input/tunable-output ones. A scheduling algorithm to forward optical packets in synchronous scenario and an analytical model to evaluate performance of the shared-per- wavelength architecture are presented. The analytical model is validated against simulation and the results show good accuracy in most cases of interest for optical switch design.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131208062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2008-04-11DOI: 10.1109/ITNEWS.2008.4488144
R. Serral-Gracià, P. Barlet-Ros, J. Domingo-Pascual
Service providers' portfolio is continuously increasing with emerging real-time services, which are offered to current and new potential clients with an associated fee. In order to properly charge such a fee, both service providers and customers must have guarantees that the offered services are correctly delivered. One way of acquiring on-line information is by monitoring some traffic parameters like one way delay or packet loss. The issue in a real scenario arises when many connections must be monitored, involving diverse underlying network technologies and different Quality of Service policies. This makes broad deployment of a measurement infrastructure very challenging. This paper presents a distributed monitoring infrastructure for tracking the QoS offered by the network. Its novelty resides on efficiently handling the resources to obtain a scalable solution that operates over heterogeneous networks. Traffic aggregation techniques are presented with the purpose of optimising the used network resources. The proposed design has been deployed and validated in a scenario using an European-wide testbed. In particular for this scenario we used heterogeneous network technologies deployed within the framework of the IST EuQoS project. This offers a convenient environment where to examine the feasibility of the solution.
{"title":"Coping with distributed monitoring of QoS-enabled heterogeneous networks","authors":"R. Serral-Gracià, P. Barlet-Ros, J. Domingo-Pascual","doi":"10.1109/ITNEWS.2008.4488144","DOIUrl":"https://doi.org/10.1109/ITNEWS.2008.4488144","url":null,"abstract":"Service providers' portfolio is continuously increasing with emerging real-time services, which are offered to current and new potential clients with an associated fee. In order to properly charge such a fee, both service providers and customers must have guarantees that the offered services are correctly delivered. One way of acquiring on-line information is by monitoring some traffic parameters like one way delay or packet loss. The issue in a real scenario arises when many connections must be monitored, involving diverse underlying network technologies and different Quality of Service policies. This makes broad deployment of a measurement infrastructure very challenging. This paper presents a distributed monitoring infrastructure for tracking the QoS offered by the network. Its novelty resides on efficiently handling the resources to obtain a scalable solution that operates over heterogeneous networks. Traffic aggregation techniques are presented with the purpose of optimising the used network resources. The proposed design has been deployed and validated in a scenario using an European-wide testbed. In particular for this scenario we used heterogeneous network technologies deployed within the framework of the IST EuQoS project. This offers a convenient environment where to examine the feasibility of the solution.","PeriodicalId":255580,"journal":{"name":"2008 4th International Telecommunication Networking Workshop on QoS in Multiservice IP Networks","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116536464","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：