Pub Date : 2005-04-18DOI: 10.1109/NGI.2005.1431640
S. Martin, I. Mitrani, K. Glazebrook
We examine the problem of how best to route jobs among a number of queues whose servers are subject to random periods of unavailability. The optimal routing policy is computed by modelling the system as a discrete-time, finite-state Markov decision process and solving the resulting dynamic programming equations. In a series of numerical experiments, the performance of various heuristic policies is compared with that of the optimal policy. A particular heuristic, using an 'index policy', is shown to be close to optimal.
{"title":"Dynamic routing among several intermittently available servers","authors":"S. Martin, I. Mitrani, K. Glazebrook","doi":"10.1109/NGI.2005.1431640","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431640","url":null,"abstract":"We examine the problem of how best to route jobs among a number of queues whose servers are subject to random periods of unavailability. The optimal routing policy is computed by modelling the system as a discrete-time, finite-state Markov decision process and solving the resulting dynamic programming equations. In a series of numerical experiments, the performance of various heuristic policies is compared with that of the optimal policy. A particular heuristic, using an 'index policy', is shown to be close to optimal.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115503339","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431647
S. Borst, O. Boxma, N. Hegde
Motivated by the need to develop simple parsimonious models for evaluating the performance of wireless data systems, we consider finite-capacity processor-sharing systems. For such systems, we analyze the sojourn time distribution, which presents a useful measure for the transfer delay of documents such as Web pages. The service rates are allowed to be state-dependent, to capture the fact that the throughput in wireless data systems may vary with the number of active users due to scheduling gains or channel collisions. We derive a set of linear equations for the Laplace-Stieltjes transforms of the sojourn time distributions, conditioned on the number of users upon arrival. This set of equations is solved, and the resulting LST's are inverted, resulting in a phase-type distribution for the unconditional sojourn time. Numerical results are provided, and two types of approximations are proposed that substantially reduce the computational effort.
{"title":"Sojourn times in finite-capacity processor-sharing queues","authors":"S. Borst, O. Boxma, N. Hegde","doi":"10.1109/NGI.2005.1431647","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431647","url":null,"abstract":"Motivated by the need to develop simple parsimonious models for evaluating the performance of wireless data systems, we consider finite-capacity processor-sharing systems. For such systems, we analyze the sojourn time distribution, which presents a useful measure for the transfer delay of documents such as Web pages. The service rates are allowed to be state-dependent, to capture the fact that the throughput in wireless data systems may vary with the number of active users due to scheduling gains or channel collisions. We derive a set of linear equations for the Laplace-Stieltjes transforms of the sojourn time distributions, conditioned on the number of users upon arrival. This set of equations is solved, and the resulting LST's are inverted, resulting in a phase-type distribution for the unconditional sojourn time. Numerical results are provided, and two types of approximations are proposed that substantially reduce the computational effort.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115732124","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431682
L. Valcarenghi, L. Rossi, F. Paolucci, P. Castoldi, F. Cugini
Multimedia communications, such as real-time video and audio streaming, have stringent quality of service requirements (e.g., throughput and delay) that must be seamlessly guaranteed in spite of network state changes. Thus if connections carrying multimedia traffic become unreliable (i.e., they do not satisfy any more their QoS constraints) they must be recovered to regain their pre-failure QoS guarantees (e.g., minimum bandwidth). In this study schemes that recover a bandwidth guaranteed client-server connection, i.e. bandwidth recovery schemes, from physical network failures, link congestion, and connection pre-emption are implemented both at the multi-protocol label switching (MPLS) layer and at the application and evaluated. Experimental results show that, thanks to the different failure detection methods utilized by each scheme, the implemented schemes form a coordinated multi-layer bandwidth recovery scheme. The implemented MPLS layer and application layer bandwidth recovery schemes allow to overcome all the considered failure scenarios.
{"title":"Multi-layer bandwidth recovery for multimedia communications: an experimental evaluation","authors":"L. Valcarenghi, L. Rossi, F. Paolucci, P. Castoldi, F. Cugini","doi":"10.1109/NGI.2005.1431682","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431682","url":null,"abstract":"Multimedia communications, such as real-time video and audio streaming, have stringent quality of service requirements (e.g., throughput and delay) that must be seamlessly guaranteed in spite of network state changes. Thus if connections carrying multimedia traffic become unreliable (i.e., they do not satisfy any more their QoS constraints) they must be recovered to regain their pre-failure QoS guarantees (e.g., minimum bandwidth). In this study schemes that recover a bandwidth guaranteed client-server connection, i.e. bandwidth recovery schemes, from physical network failures, link congestion, and connection pre-emption are implemented both at the multi-protocol label switching (MPLS) layer and at the application and evaluated. Experimental results show that, thanks to the different failure detection methods utilized by each scheme, the implemented schemes form a coordinated multi-layer bandwidth recovery scheme. The implemented MPLS layer and application layer bandwidth recovery schemes allow to overcome all the considered failure scenarios.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117210492","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431662
J. Joutsensalo, A. Viinikainen, Timo Hämäläinen, Mika Wikström
The pricing of future network services should comparable to the service quality. In this paper we present a packet scheduling scheme which ensures bandwidth as a quality of service (QoS) requirement and optimizes revenue of the network service provider. We derive a closed form formula for updating the adaptive weights of a packet scheduler from a revenue-based optimization problem. The weight updating procedure is fast and independent on any assumption of the connections' statistical behavior. The algorithm is simulated and compared to constant weights with and without a call admission control (CAC) mechanism. Also, a mechanism for guaranteeing a minimum mean bandwidth for different service classes is presented as supplement to the CAC procedure.
{"title":"Bandwidth allocation and pricing for telecommunications network","authors":"J. Joutsensalo, A. Viinikainen, Timo Hämäläinen, Mika Wikström","doi":"10.1109/NGI.2005.1431662","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431662","url":null,"abstract":"The pricing of future network services should comparable to the service quality. In this paper we present a packet scheduling scheme which ensures bandwidth as a quality of service (QoS) requirement and optimizes revenue of the network service provider. We derive a closed form formula for updating the adaptive weights of a packet scheduler from a revenue-based optimization problem. The weight updating procedure is fast and independent on any assumption of the connections' statistical behavior. The algorithm is simulated and compared to constant weights with and without a call admission control (CAC) mechanism. Also, a mechanism for guaranteeing a minimum mean bandwidth for different service classes is presented as supplement to the CAC procedure.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129325446","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431680
M. Sabeur, D. Popa, T. Atmaca
In this article, we present an analytical study of an optical packet-switched network at two levels: at the border and in the core of the network. The aim of this paper is to predict the impact of handling different classes of service (CoS) on the network performance. The lack of memory in the core of optical network and the self-similarity nature of traffic are of great interest in this study. We use the equivalent bandwidth approach to carry out this goal. The analytical-based results are compared with those obtained by simulation.
{"title":"Equivalent bandwidth based-study of a multi-service optical network","authors":"M. Sabeur, D. Popa, T. Atmaca","doi":"10.1109/NGI.2005.1431680","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431680","url":null,"abstract":"In this article, we present an analytical study of an optical packet-switched network at two levels: at the border and in the core of the network. The aim of this paper is to predict the impact of handling different classes of service (CoS) on the network performance. The lack of memory in the core of optical network and the self-similarity nature of traffic are of great interest in this study. We use the equivalent bandwidth approach to carry out this goal. The analytical-based results are compared with those obtained by simulation.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132076864","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431673
F. Baccelli, D. McDonald
In this paper, we derive a closed form formula for the average rate attained by a non persistent TCP source which alternates between idle periods and download periods subject to a fixed packet loss probability. We also derive closed form expressions for the mean time to transfer a file and for the distribution of the transmission rate. Several distributions for the file sizes and idle times are considered including heavy tailed distributions. The formula for mean transmission rate is shown to boil down to the classical square root mean value formula for persistent flows when the average file size tends to infinity. The formulae are applied to predict bandwidth sharing among competing flows subject to active queue management.
{"title":"A square root formula for the rate of non-persistent TCP flows","authors":"F. Baccelli, D. McDonald","doi":"10.1109/NGI.2005.1431673","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431673","url":null,"abstract":"In this paper, we derive a closed form formula for the average rate attained by a non persistent TCP source which alternates between idle periods and download periods subject to a fixed packet loss probability. We also derive closed form expressions for the mean time to transfer a file and for the distribution of the transmission rate. Several distributions for the file sizes and idle times are considered including heavy tailed distributions. The formula for mean transmission rate is shown to boil down to the classical square root mean value formula for persistent flows when the average file size tends to infinity. The formulae are applied to predict bandwidth sharing among competing flows subject to active queue management.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129649604","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431692
N. Markovich
An important problem connected with the heavy-tailed probability density function (PDF) and high quantile estimation is the estimation of the tail index. The recursive estimator of the tail index is investigated and its accuracy is obtained. The parameter of this estimator is calculated by bootstrap method.
{"title":"On-line estimation of the tail index for heavy-tailed distributions with application to WWW-traffic","authors":"N. Markovich","doi":"10.1109/NGI.2005.1431692","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431692","url":null,"abstract":"An important problem connected with the heavy-tailed probability density function (PDF) and high quantile estimation is the estimation of the tail index. The recursive estimator of the tail index is investigated and its accuracy is obtained. The parameter of this estimator is calculated by bootstrap method.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"T162 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125667199","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431681
Q. Fu, J. Indulska
TCP parallelization uses multiple virtual TCP connections to transfer data for an application process. The virtual connections can be standard or modified TCP connections. TCP parallelization has been used to effectively utilise bandwidth for data intensive applications. However, its aggressiveness to increase window and reluctance to reduce window tend to be unfair to other traffic in the network. There are solutions tackling this fairness problem. In these solutions, the fairness is achieved at the cost of losing the robustness of TCP parallelization. Other solutions use a single TCP connection to emulate the behaviour of a set of parallel TCP connections and therefore take advantage of the robustness of TCP parallelization and its fairness bias. These approaches can be used to achieve proportional fairness (e.g. MulTCP), or just to give a better throughput. However, such solutions may not be able to achieve the desired results. In this paper, we analyse these solutions and show their potential problems. A delay-based parallelization approach is then suggested to address the fairness concern by decoupling fairness and effectiveness. It shows that the desired fairness can be achieved in presence of congestion while the bandwidth can be effectively used in absence of congestion.
{"title":"A delay-based approach for TCP parallelization","authors":"Q. Fu, J. Indulska","doi":"10.1109/NGI.2005.1431681","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431681","url":null,"abstract":"TCP parallelization uses multiple virtual TCP connections to transfer data for an application process. The virtual connections can be standard or modified TCP connections. TCP parallelization has been used to effectively utilise bandwidth for data intensive applications. However, its aggressiveness to increase window and reluctance to reduce window tend to be unfair to other traffic in the network. There are solutions tackling this fairness problem. In these solutions, the fairness is achieved at the cost of losing the robustness of TCP parallelization. Other solutions use a single TCP connection to emulate the behaviour of a set of parallel TCP connections and therefore take advantage of the robustness of TCP parallelization and its fairness bias. These approaches can be used to achieve proportional fairness (e.g. MulTCP), or just to give a better throughput. However, such solutions may not be able to achieve the desired results. In this paper, we analyse these solutions and show their potential problems. A delay-based parallelization approach is then suggested to address the fairness concern by decoupling fairness and effectiveness. It shows that the desired fairness can be achieved in presence of congestion while the bandwidth can be effectively used in absence of congestion.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"544 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123378843","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431700
B. Quoitin, O. Bonaventure
For performance or cost reasons, autonomous systems (AS) often need to control the flow of their incoming interdomain traffic. Controlling its incoming traffic is a difficult task since it often implies influencing ASes on the path. The current BGP-based techniques that an AS can use for this purpose are primitive. Moreover, their effect is often difficult to predict. In this paper, we propose to solve this problem by using Virtual Peerings. A Virtual Peering is an IP tunnel between a border router of a source AS and a border router of a destination AS. This tunnel is established upon request from the destination AS. These tunnels can be negotiated by using backward compatible modifications to the border gateway protocol (BGP). By using Virtual Peerings, the source and destination ASes can achieve various traffic engineering objectives such as traffic-balancing or reducing the latency. A key advantage of our solution is that it does not require cooperation of the intermediate ASes and that it can be incrementally deployed in today's Internet. We then show by simulations that in a load-balancing scenario, a multi-homed AS only needs to request a few dozens of Virtual Peerings to balance its incoming traffic.
{"title":"A cooperative approach to interdomain traffic engineering","authors":"B. Quoitin, O. Bonaventure","doi":"10.1109/NGI.2005.1431700","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431700","url":null,"abstract":"For performance or cost reasons, autonomous systems (AS) often need to control the flow of their incoming interdomain traffic. Controlling its incoming traffic is a difficult task since it often implies influencing ASes on the path. The current BGP-based techniques that an AS can use for this purpose are primitive. Moreover, their effect is often difficult to predict. In this paper, we propose to solve this problem by using Virtual Peerings. A Virtual Peering is an IP tunnel between a border router of a source AS and a border router of a destination AS. This tunnel is established upon request from the destination AS. These tunnels can be negotiated by using backward compatible modifications to the border gateway protocol (BGP). By using Virtual Peerings, the source and destination ASes can achieve various traffic engineering objectives such as traffic-balancing or reducing the latency. A key advantage of our solution is that it does not require cooperation of the intermediate ASes and that it can be incrementally deployed in today's Internet. We then show by simulations that in a load-balancing scenario, a multi-homed AS only needs to request a few dozens of Virtual Peerings to balance its incoming traffic.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126358391","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 : 2005-04-18DOI: 10.1109/NGI.2005.1431645
M. Koubàa, N. Puech, M. Gagnaire
In this paper we consider working and protection paths for scheduled and random lightpath demands in an optical transport network without wavelength conversion. As the network resources are limited, the objective is to minimize the rejection ratio. To achieve this goal, we use backup multiplexing techniques to minimize the spare resources required to ensure protection. Scheduled lightpath demands (SLDs) are connection demands for which the set-up and tear-down times are known in advance as opposed to random lightpath demands (RLDs) whose date arrival and holding times are not known in advance and that are established on the fly. We propose and implement two routing algorithms to deal with the routing and spare capacity assignment problem for scheduled and random lightpath demands. The former exploits the a priori knowledge of scheduled lightpath demands to compute the routing and spare capacity assignment for the SLDs before considering the RLDs. The latter considers all the demands simultaneously using a sequential algorithm that computes the routing and spare capacity assignment on the fly, that is, demand by demand. We compare the proposed algorithms in terms of the rejection ratio.
{"title":"Routing and spare capacity assignment for scheduled and random lightpath demands in all-optical networks","authors":"M. Koubàa, N. Puech, M. Gagnaire","doi":"10.1109/NGI.2005.1431645","DOIUrl":"https://doi.org/10.1109/NGI.2005.1431645","url":null,"abstract":"In this paper we consider working and protection paths for scheduled and random lightpath demands in an optical transport network without wavelength conversion. As the network resources are limited, the objective is to minimize the rejection ratio. To achieve this goal, we use backup multiplexing techniques to minimize the spare resources required to ensure protection. Scheduled lightpath demands (SLDs) are connection demands for which the set-up and tear-down times are known in advance as opposed to random lightpath demands (RLDs) whose date arrival and holding times are not known in advance and that are established on the fly. We propose and implement two routing algorithms to deal with the routing and spare capacity assignment problem for scheduled and random lightpath demands. The former exploits the a priori knowledge of scheduled lightpath demands to compute the routing and spare capacity assignment for the SLDs before considering the RLDs. The latter considers all the demands simultaneously using a sequential algorithm that computes the routing and spare capacity assignment on the fly, that is, demand by demand. We compare the proposed algorithms in terms of the rejection ratio.","PeriodicalId":435785,"journal":{"name":"Next Generation Internet Networks, 2005","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129602913","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：