The implementation of successful assured forwarding (AF) services according to the DiffServ framework remains a challenging problem today, despite the numerous proposals for AF PHB mechanisms and AF-based service implementations. In this work, we propose two modules, the TCP-window aware marker (TWAM) and the dynamic WRED (DWRED) mechanism for implementing the DiffServ AF PHB. We provide analytical models and an experimental evaluation in order to demonstrate how they succeed in enhancing the quality, increasing fairness, improving the performance and easing the deployment of a production- level AF-based service.
尽管有许多关于AF- PHB机制和基于AF的服务实现的建议，但是根据DiffServ框架实现成功的保证转发(AF)服务仍然是一个具有挑战性的问题。在这项工作中，我们提出了两个模块，tcp窗口感知标记(TWAM)和动态WRED (DWRED)机制来实现DiffServ AF PHB。我们提供了分析模型和实验评估，以证明它们如何成功地提高了质量，增加了公平性，提高了性能，并简化了生产级基于af的服务的部署。
{"title":"Performance enhancement of an AF service using TCP-aware marking and dynamic WRED","authors":"C. Bouras, A. Sevasti","doi":"10.1109/ISCC.2005.121","DOIUrl":"https://doi.org/10.1109/ISCC.2005.121","url":null,"abstract":"The implementation of successful assured forwarding (AF) services according to the DiffServ framework remains a challenging problem today, despite the numerous proposals for AF PHB mechanisms and AF-based service implementations. In this work, we propose two modules, the TCP-window aware marker (TWAM) and the dynamic WRED (DWRED) mechanism for implementing the DiffServ AF PHB. We provide analytical models and an experimental evaluation in order to demonstrate how they succeed in enhancing the quality, increasing fairness, improving the performance and easing the deployment of a production- level AF-based service.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123347405","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}
A key element for MANET performance is the routing protocol. Surprisingly, little effort has been devoted up to now to analyzing the impact of routing protocol parameter settings on MANET performance. We define and study in detail the route change latency (RCL) after link failures, a critical performance metric in a non-static ad-hoc scenario, and its dependence on routing protocol parameter settings and implementation issues using OLSR. We experiment with a set of OLSR settings in a real network environment and derive its potential effect in a generic situation, showing that end-to-end connectivity can be enhanced using different parameter settings from the default ones.
{"title":"Improving performance of a real ad-hoc network by tuning OLSR parameters","authors":"Carles Gomez, D. García, J. Paradells","doi":"10.1109/ISCC.2005.83","DOIUrl":"https://doi.org/10.1109/ISCC.2005.83","url":null,"abstract":"A key element for MANET performance is the routing protocol. Surprisingly, little effort has been devoted up to now to analyzing the impact of routing protocol parameter settings on MANET performance. We define and study in detail the route change latency (RCL) after link failures, a critical performance metric in a non-static ad-hoc scenario, and its dependence on routing protocol parameter settings and implementation issues using OLSR. We experiment with a set of OLSR settings in a real network environment and derive its potential effect in a generic situation, showing that end-to-end connectivity can be enhanced using different parameter settings from the default ones.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121716804","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}
Mátyás Martinecz, A. Kern, Zalán Heszberger, J. Bíró
Recent Internet access speeds have made it possible even for home users to reach and use previously unavailable bandwidth consuming (e.g. video on demand) services. However, the rapid spread of these applications is hindered by the lack of QoS guarantees in packet networks. In this article a QoS architecture is presented which have been designed for access networks and applies a logical overlay network to make network management tasks easier. The novelty of the architecture lies in the application of two key procedures adopted for dimensioning logical channels between edge nodes and mapping them to the physical routes. The upper load thresholds of the applied logical channels are also determined, which may form the basis of a more sophisticated call admission control policy.
{"title":"Architecture and configuration of broadband access networks supporting multimedia applications","authors":"Mátyás Martinecz, A. Kern, Zalán Heszberger, J. Bíró","doi":"10.1109/ISCC.2005.35","DOIUrl":"https://doi.org/10.1109/ISCC.2005.35","url":null,"abstract":"Recent Internet access speeds have made it possible even for home users to reach and use previously unavailable bandwidth consuming (e.g. video on demand) services. However, the rapid spread of these applications is hindered by the lack of QoS guarantees in packet networks. In this article a QoS architecture is presented which have been designed for access networks and applies a logical overlay network to make network management tasks easier. The novelty of the architecture lies in the application of two key procedures adopted for dimensioning logical channels between edge nodes and mapping them to the physical routes. The upper load thresholds of the applied logical channels are also determined, which may form the basis of a more sophisticated call admission control policy.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123996526","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}
Jhon J. Padilla Aguilar, M. Huerta, J. Aspas, X. Hesselbach
This paper proposes an approach to support end-to-end quality of service over IPv6 networks. In our solution, IPv6 flow labels are used to give better performance into the process of packet classification on network routers. In order to evaluate our proposal, a router model is obtained and it is compared with other technologies as IntServ and MPLS. As result, we obtain a solution with benefits of QoS support and label switching in IPv6 routers.
{"title":"IntServ6: an approach to support QoS over IPv6 networks","authors":"Jhon J. Padilla Aguilar, M. Huerta, J. Aspas, X. Hesselbach","doi":"10.1109/ISCC.2005.86","DOIUrl":"https://doi.org/10.1109/ISCC.2005.86","url":null,"abstract":"This paper proposes an approach to support end-to-end quality of service over IPv6 networks. In our solution, IPv6 flow labels are used to give better performance into the process of packet classification on network routers. In order to evaluate our proposal, a router model is obtained and it is compared with other technologies as IntServ and MPLS. As result, we obtain a solution with benefits of QoS support and label switching in IPv6 routers.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126200415","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}
Zsolt Kenesi, Zoltán Szabó, Zsolt Belicza, S. Molnár
The utilization of the free capacity by the TCP congestion control is not perfect due to the complex fluctuations of the background traffic resulting in reduced throughput. Different frequency components of the background traffic have different effects on the TCP performance. In this paper we present a comprehensive TCP performance evaluation study to understand the nature of the TCP adaptivity mechanism in frequency domain. Through simulations we analyze the impact of network parameters and also some alternatives for compensation of the throughput reduction. Finally, based also on the analysis of measured TCP traffic we investigate the robustness of TCP to fluctuations occur in actual network traffic.
{"title":"On the effect of the background traffic on TCP's throughput","authors":"Zsolt Kenesi, Zoltán Szabó, Zsolt Belicza, S. Molnár","doi":"10.1109/ISCC.2005.111","DOIUrl":"https://doi.org/10.1109/ISCC.2005.111","url":null,"abstract":"The utilization of the free capacity by the TCP congestion control is not perfect due to the complex fluctuations of the background traffic resulting in reduced throughput. Different frequency components of the background traffic have different effects on the TCP performance. In this paper we present a comprehensive TCP performance evaluation study to understand the nature of the TCP adaptivity mechanism in frequency domain. Through simulations we analyze the impact of network parameters and also some alternatives for compensation of the throughput reduction. Finally, based also on the analysis of measured TCP traffic we investigate the robustness of TCP to fluctuations occur in actual network traffic.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121616631","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}
InfiniBand (IBA) has been proposed as an industry-standard architecture both for I/O server and interprocessor communication. IBA employs a switched point-to-point network, instead of using a shared bus. IBA is being developed by the InfiniBand/sub SM/ Trade Association to provide present and future server systems with the required levels of reliability, availability, performance, scalability, and quality of service (QoS). In previous papers we have proposed an effective strategy for configuring the IBA networks to provide users with the required levels of QoS. This strategy is based on the proper configuration of the mechanisms IBA carries to support QoS. Specifically, our methodology configures the InfiniBand arbitration tables and uses the different service levels and virtual lanes that are available, in order to segregate the different traffic flows. Thus, each flow receives the treatment it has previously requested. Moreover, by using our methodology, applications can be assured that their requirements will be satisfied. In this paper, we review the basis of our methodology and we study the influence of the packet size on the QoS guaranteed to the applications.
{"title":"Studying the influence of the InfiniBand packet size to guarantee QoS","authors":"F. J. Alfaro, J. L. Sánchez, J. Duato","doi":"10.1109/ISCC.2005.140","DOIUrl":"https://doi.org/10.1109/ISCC.2005.140","url":null,"abstract":"InfiniBand (IBA) has been proposed as an industry-standard architecture both for I/O server and interprocessor communication. IBA employs a switched point-to-point network, instead of using a shared bus. IBA is being developed by the InfiniBand/sub SM/ Trade Association to provide present and future server systems with the required levels of reliability, availability, performance, scalability, and quality of service (QoS). In previous papers we have proposed an effective strategy for configuring the IBA networks to provide users with the required levels of QoS. This strategy is based on the proper configuration of the mechanisms IBA carries to support QoS. Specifically, our methodology configures the InfiniBand arbitration tables and uses the different service levels and virtual lanes that are available, in order to segregate the different traffic flows. Thus, each flow receives the treatment it has previously requested. Moreover, by using our methodology, applications can be assured that their requirements will be satisfied. In this paper, we review the basis of our methodology and we study the influence of the packet size on the QoS guaranteed to the applications.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"276 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120928930","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}
An anycast service in a QoS-based network requires to setup paths with bandwidth guarantees and to select a server with capacity guarantees. This paper proposes an architecture which integrates both network layer anycasting service and application layer anyscasting service in order to select both a path and a server that satisfies the client QoS constrain in DiffServ networks. To perform a path selection algorithm, the proposed protocol uses the idea of residual capacity to select all paths which match the client bandwidth requirement. To perform the server selection process, each server resource management (SRM) uses server weight of each server to select a best server that has available capacity for the client request. Furthermore, the design of multiple SRMs eliminates the single point of failure and increases reliability in a DiffServ network. The simulation results have shown that the proposed protocol has much faster server response time and has lowest request blocking rate than other existing protocols, especially in the case where the traffic load is high.
{"title":"An integrated anycast end-to-end QoS routing on DiffServ networks","authors":"Wu-Hsiao Hsu, Cheng-Lin Yang, Li-Yuan Wu, Chiun-Sheng Hsu","doi":"10.1109/ISCC.2005.28","DOIUrl":"https://doi.org/10.1109/ISCC.2005.28","url":null,"abstract":"An anycast service in a QoS-based network requires to setup paths with bandwidth guarantees and to select a server with capacity guarantees. This paper proposes an architecture which integrates both network layer anycasting service and application layer anyscasting service in order to select both a path and a server that satisfies the client QoS constrain in DiffServ networks. To perform a path selection algorithm, the proposed protocol uses the idea of residual capacity to select all paths which match the client bandwidth requirement. To perform the server selection process, each server resource management (SRM) uses server weight of each server to select a best server that has available capacity for the client request. Furthermore, the design of multiple SRMs eliminates the single point of failure and increases reliability in a DiffServ network. The simulation results have shown that the proposed protocol has much faster server response time and has lowest request blocking rate than other existing protocols, especially in the case where the traffic load is high.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126889344","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}
G. Anastasi, Emilio Ancillotti, M. Conti, A. Passarella
Several previous works have shown that TCP exhibits poor performance in mobile ad hoc networks (MANETs). The ultimate reason for this is that MANETs behave in a significantly different way from traditional wired networks, like the Internet, for which TCP was originally designed. In this paper we propose a novel transport protocol - named TPA - specifically tailored to the characteristics of the MANET environment. It is based on a completely new congestion control mechanism, and designed in such a way to minimize the number of useless transmissions and, hence, power consumption. Furthermore, it is able to manage efficiently route changes and route failures. We evaluated the TPA protocol in a static scenario where TCP exhibits good performance. Simulation results show that, even in such a scenario, TPA significantly outperforms TCP.
{"title":"TPA: a transport protocol for ad hoc networks","authors":"G. Anastasi, Emilio Ancillotti, M. Conti, A. Passarella","doi":"10.1109/ISCC.2005.147","DOIUrl":"https://doi.org/10.1109/ISCC.2005.147","url":null,"abstract":"Several previous works have shown that TCP exhibits poor performance in mobile ad hoc networks (MANETs). The ultimate reason for this is that MANETs behave in a significantly different way from traditional wired networks, like the Internet, for which TCP was originally designed. In this paper we propose a novel transport protocol - named TPA - specifically tailored to the characteristics of the MANET environment. It is based on a completely new congestion control mechanism, and designed in such a way to minimize the number of useless transmissions and, hence, power consumption. Furthermore, it is able to manage efficiently route changes and route failures. We evaluated the TPA protocol in a static scenario where TCP exhibits good performance. Simulation results show that, even in such a scenario, TPA significantly outperforms TCP.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125296495","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}
This paper addresses the problem of formalizing service level specifications (SLSs) as a first step to simplify and automate the configuration and management of multiservice IP networks. A formal representation of SLSs will allow their automatic validation and processing, fostering the dynamic negotiation of SLSs and the interoperability among service management entities. In this way, taking advantage of XML extensibility and portability, a Schema is presented describing XML SLSs sections and their contents. In addition, an XML validator tool was built to check if SLSs are correctly specified. An XML SLS for an IP telephony service is used to exemplify this proposal expressiveness.
{"title":"XML service level specification and validation","authors":"Pedro Alípio, S. R. Lima, P. Carvalho","doi":"10.1109/ISCC.2005.157","DOIUrl":"https://doi.org/10.1109/ISCC.2005.157","url":null,"abstract":"This paper addresses the problem of formalizing service level specifications (SLSs) as a first step to simplify and automate the configuration and management of multiservice IP networks. A formal representation of SLSs will allow their automatic validation and processing, fostering the dynamic negotiation of SLSs and the interoperability among service management entities. In this way, taking advantage of XML extensibility and portability, a Schema is presented describing XML SLSs sections and their contents. In addition, an XML validator tool was built to check if SLSs are correctly specified. An XML SLS for an IP telephony service is used to exemplify this proposal expressiveness.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126118171","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}
The paper deals with measurement-based CAC alternatives applicable to the novel WIND-FLEX architecture, a 17 GHz wireless LAN recently developed in the framework of an European 1ST project. We first present some existing CAC algorithms based on traffic stream measurement. Afterwards, we develop a novel CAC algorithm which is based on queue state measurement. The performance of the different CAC algorithms is then evaluated by extensive simulations. One of the main conclusions from simulative investigations is that our novel CAC algorithm is superior to other ones from the aspects of achievable utilization and parameter robustness. Moreover the proposed solution requires no additional signaling overhead and demands low implementation complexity.
{"title":"17 GHz wireless LAN: performance analysis of CAC algorithms","authors":"G. Razzano, H. T. Tran, Camilla Cantarella","doi":"10.1109/ISCC.2005.5","DOIUrl":"https://doi.org/10.1109/ISCC.2005.5","url":null,"abstract":"The paper deals with measurement-based CAC alternatives applicable to the novel WIND-FLEX architecture, a 17 GHz wireless LAN recently developed in the framework of an European 1ST project. We first present some existing CAC algorithms based on traffic stream measurement. Afterwards, we develop a novel CAC algorithm which is based on queue state measurement. The performance of the different CAC algorithms is then evaluated by extensive simulations. One of the main conclusions from simulative investigations is that our novel CAC algorithm is superior to other ones from the aspects of achievable utilization and parameter robustness. Moreover the proposed solution requires no additional signaling overhead and demands low implementation complexity.","PeriodicalId":315855,"journal":{"name":"10th IEEE Symposium on Computers and Communications (ISCC'05)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128962561","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：