The medium access control (MAC) protocol is one of the key components for providing quality of service (QoS) in wireless ATM (W-ATM) networks. In this paper, we propose a hierarchical scheduling scheme coupled with fair queueing algorithms with adaptive weights. This scheme is intended to be applicable to a TDMA/TDD-based MAC protocol. Specifically, the performance of the fair-queueing algorithm using fixed weights and adaptive weights are investigated. Simulation results show that the proposed hierarchical fair-queueing scheduling with adaptive weights (HAW) can yield a lower cell transfer delay and higher channel utilization while maintaining fairness among multiple users.
{"title":"Hierarchical scheduling with adaptive weights for W-ATM","authors":"Hui Huang, D. Tsang, R. Sigle, P. Kuehn","doi":"10.1109/ATM.1999.786878","DOIUrl":"https://doi.org/10.1109/ATM.1999.786878","url":null,"abstract":"The medium access control (MAC) protocol is one of the key components for providing quality of service (QoS) in wireless ATM (W-ATM) networks. In this paper, we propose a hierarchical scheduling scheme coupled with fair queueing algorithms with adaptive weights. This scheme is intended to be applicable to a TDMA/TDD-based MAC protocol. Specifically, the performance of the fair-queueing algorithm using fixed weights and adaptive weights are investigated. Simulation results show that the proposed hierarchical fair-queueing scheduling with adaptive weights (HAW) can yield a lower cell transfer delay and higher channel utilization while maintaining fairness among multiple users.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128447809","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 presents a new distributed scheduling architecture for wireless ATM networks. Usually, in WATM scheduling architectures, a fixed order is defined among the different connections through their ATM service category (CBR, VBR, ABR, UBR). We argue that although this static priority is easier to implement, this type of precedence is not necessarily a good choice for the MAC layer. The MAC layer scheduling should define an order such that it uses efficiently the resources while providing quality of service (QoS) guarantees. In this spirit, our architecture delays (without violating their QoS) the real-time connections in order to improve the performance of non-real-time connections.
{"title":"Efficient distributed scheduling architecture for wireless ATM networks","authors":"C. Tse, B. Bensaou, K. C. Chua","doi":"10.1109/ATM.1999.786891","DOIUrl":"https://doi.org/10.1109/ATM.1999.786891","url":null,"abstract":"This paper presents a new distributed scheduling architecture for wireless ATM networks. Usually, in WATM scheduling architectures, a fixed order is defined among the different connections through their ATM service category (CBR, VBR, ABR, UBR). We argue that although this static priority is easier to implement, this type of precedence is not necessarily a good choice for the MAC layer. The MAC layer scheduling should define an order such that it uses efficiently the resources while providing quality of service (QoS) guarantees. In this spirit, our architecture delays (without violating their QoS) the real-time connections in order to improve the performance of non-real-time connections.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130960993","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}
In this paper, we report on how we conducted a detailed analysis of a very large corporate ATM backbone network. The performance analysis tasks were conducted in three stages: call route analysis, traffic flow analysis and network performance analysis. The call route analysis was to determine if the routes from the edge switch to the core switch took the optimal path. This analysis revealed several sub-optimal routes for the IISP protocol configuration. The traffic flow analysis is based on the need for traffic to flow from one ELAN to another ELAN through a virtual router. The analysis showed that several routes require additional hops to reach their respective primary and secondary routers. The performance analysis was conducted using the data collected through the SNMP management system for a two month period. By using several statistical analytical methods, the high utilization segments in the edge networks as well as in the core networks were identified. The results showed that the LAN segment can accommodate two times more bandwidth for the currently shared network users, and can accommodate 50% more traffic when these users are migrated to the switched ports. For outer cores, the current utilization remained at 20%, i.e., four times more than the current traffic can be handled. Finally, the inner core traffic stayed at about 16% of the network utilization for 80% of the time. This analysis demonstrated a systematic method of demonstrating current network utilization to the network manager for current network operation and future growth planning.
{"title":"Performance analysis of an ATM-based local area network","authors":"W. Lu, T. Reyes, E. Krieser, V. Musi","doi":"10.1109/ATM.1999.786870","DOIUrl":"https://doi.org/10.1109/ATM.1999.786870","url":null,"abstract":"In this paper, we report on how we conducted a detailed analysis of a very large corporate ATM backbone network. The performance analysis tasks were conducted in three stages: call route analysis, traffic flow analysis and network performance analysis. The call route analysis was to determine if the routes from the edge switch to the core switch took the optimal path. This analysis revealed several sub-optimal routes for the IISP protocol configuration. The traffic flow analysis is based on the need for traffic to flow from one ELAN to another ELAN through a virtual router. The analysis showed that several routes require additional hops to reach their respective primary and secondary routers. The performance analysis was conducted using the data collected through the SNMP management system for a two month period. By using several statistical analytical methods, the high utilization segments in the edge networks as well as in the core networks were identified. The results showed that the LAN segment can accommodate two times more bandwidth for the currently shared network users, and can accommodate 50% more traffic when these users are migrated to the switched ports. For outer cores, the current utilization remained at 20%, i.e., four times more than the current traffic can be handled. Finally, the inner core traffic stayed at about 16% of the network utilization for 80% of the time. This analysis demonstrated a systematic method of demonstrating current network utilization to the network manager for current network operation and future growth planning.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114411030","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}
Current ATM standards regarding the procedures for delay accumulation appear to be incomplete, and overestimate actual accumulated delay. This paper provides a summary of delay bound theory for packet networks and reviews the signaling support for delay bounds. A proposal is made for improved procedures regarding delay accumulation. Additional procedural improvements regarding delay parameters are also proposed.
{"title":"ATM network procedures supporting delay QoS","authors":"S. Wright, Y. Viniotis","doi":"10.1109/ATM.1999.786883","DOIUrl":"https://doi.org/10.1109/ATM.1999.786883","url":null,"abstract":"Current ATM standards regarding the procedures for delay accumulation appear to be incomplete, and overestimate actual accumulated delay. This paper provides a summary of delay bound theory for packet networks and reviews the signaling support for delay bounds. A proposal is made for improved procedures regarding delay accumulation. Additional procedural improvements regarding delay parameters are also proposed.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123052922","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}
We propose a new explicit-rate (ER) available bit rate (ABR) algorithm (load-based explicit rate control, or LBERC) that can maintain high link utilization and reduce congestion in asynchronous transfer mode (ATM) networks. In order to evaluate the algorithm, we model World-Wide Web (WWW) traffic using the file transfer distribution of the SPECweb96 benchmark. We simulate Web traffic via a client-server model that consists of Web clients, Web servers, and a finite source function (FSF) to represent client (user) behavior. When network congestion occurs, users access fewer Web pages (and open fewer new transmission control protocol (TCP) sessions). The FSF simulates this user behavior by counting the number of active TCP sessions. We present TCP-over-ABR/LBERC simulation results to verify the compatibility of interactions between TCP's window-based flow control and ABR's rate-based flow control.
{"title":"Performance evaluation of TCP over ATM using World Wide Web traffic","authors":"T. Soumiya, K. Nakamichi, A. Bragg","doi":"10.1109/ATM.1999.786905","DOIUrl":"https://doi.org/10.1109/ATM.1999.786905","url":null,"abstract":"We propose a new explicit-rate (ER) available bit rate (ABR) algorithm (load-based explicit rate control, or LBERC) that can maintain high link utilization and reduce congestion in asynchronous transfer mode (ATM) networks. In order to evaluate the algorithm, we model World-Wide Web (WWW) traffic using the file transfer distribution of the SPECweb96 benchmark. We simulate Web traffic via a client-server model that consists of Web clients, Web servers, and a finite source function (FSF) to represent client (user) behavior. When network congestion occurs, users access fewer Web pages (and open fewer new transmission control protocol (TCP) sessions). The FSF simulates this user behavior by counting the number of active TCP sessions. We present TCP-over-ABR/LBERC simulation results to verify the compatibility of interactions between TCP's window-based flow control and ABR's rate-based flow control.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123236826","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}
VC merging allows VC to be mapped onto the same VC label. It provides a scalable solution for network growth. It enables features such as stream merging in MPLS and RSVP wildcard filter to be efficiently supported over ATM. It also plays an important role in providing multipoint-to-multipoint multicast and share-tree multicasting paradigm. VC merging involves distinguishing cells from an identical merged VC label. Various approaches have been proposed to help this identification process. However, most of them incur additional buffering, protocol overhead and/or variable delay. They make the provision of QoS difficult to achieve. In this paper, we propose a novel merge-capable scheduler to support VC merging. The proposed scheduler consists of a core scheduler and a number of sub-queue sequencers forming a 2-level hierarchical structure. Using the proposed merge-capable scheduler, we can uniquely identify incoming cells while: (1) allowing cell cut-through forwarding and interleaving; (2) maintaining per-flow QoS; and (3) requiring no additional buffer and protocol overhead. We show analytically that the scheduler can guarantee a worst-case fairness and deliver a worst-case delay bound. We further analyze the scheduler performance in term of the service received by a queue and the average queuing delay experienced by a cell. The results indicate that the amount of services is fairly distributed among merging and non-merging flows according to their reservations. Moreover, the proposed merge-capable scheduler performs equally as well as its non-merging counterpart when reservations are highly utilized, even though it may introduce a minimal additional delay when utilization is low.
{"title":"VC-merge capable scheduler design","authors":"H. Chow, A. Leon-Garcia","doi":"10.1109/ATM.1999.786851","DOIUrl":"https://doi.org/10.1109/ATM.1999.786851","url":null,"abstract":"VC merging allows VC to be mapped onto the same VC label. It provides a scalable solution for network growth. It enables features such as stream merging in MPLS and RSVP wildcard filter to be efficiently supported over ATM. It also plays an important role in providing multipoint-to-multipoint multicast and share-tree multicasting paradigm. VC merging involves distinguishing cells from an identical merged VC label. Various approaches have been proposed to help this identification process. However, most of them incur additional buffering, protocol overhead and/or variable delay. They make the provision of QoS difficult to achieve. In this paper, we propose a novel merge-capable scheduler to support VC merging. The proposed scheduler consists of a core scheduler and a number of sub-queue sequencers forming a 2-level hierarchical structure. Using the proposed merge-capable scheduler, we can uniquely identify incoming cells while: (1) allowing cell cut-through forwarding and interleaving; (2) maintaining per-flow QoS; and (3) requiring no additional buffer and protocol overhead. We show analytically that the scheduler can guarantee a worst-case fairness and deliver a worst-case delay bound. We further analyze the scheduler performance in term of the service received by a queue and the average queuing delay experienced by a cell. The results indicate that the amount of services is fairly distributed among merging and non-merging flows according to their reservations. Moreover, the proposed merge-capable scheduler performs equally as well as its non-merging counterpart when reservations are highly utilized, even though it may introduce a minimal additional delay when utilization is low.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127788442","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}
Numerous research has been done on modeling the traffic flows in a multiservice network. An alternative proposal was made by Addie et al. (1998). The authors argued that because the central limit theorem can be applied to heavy traffic processes, therefore as more traffic is aggregated together through the sharing of a transmission link, the aggregated traffic will weakly converge to a Gaussian process. In this paper, we apply this Gaussian model to the connection admission control (CAC) of a multiservice network. We consider a store-and-forward (packet or cell-switched) connection-oriented multiservice network that can be based on the concept of asynchronous transfer mode. We also make efficiency comparison with a CAC based on an effective bandwidth model. Two scenarios were considered: the first assumes that all a priori traffic specifications are known before a CAC acceptance/rejection decision is made, while the second assumes that only the peak rate of a new (requesting admission) connection is declared to the CAC mechanism. The latter scenario entails a measurement-based CAC approach.
对多业务网络中交通流的建模进行了大量的研究。Addie et al.(1998)提出了另一种建议。作者认为，由于中心极限定理可以应用于大流量过程，因此当更多的流量通过共享传输链路聚集在一起时，聚集的流量将弱收敛到高斯过程。本文将该高斯模型应用于多业务网络的连接允许控制(CAC)。我们考虑一个基于异步传输模式概念的面向存储转发(分组或单元交换)连接的多业务网络。并与基于有效带宽模型的CAC进行了效率比较。我们考虑了两种场景:第一种假设在做出CAC接受/拒绝决定之前，所有先验的流量规范都是已知的，而第二种假设只向CAC机制声明新(请求允许)连接的峰值速率。后一种情况需要基于度量的CAC方法。
{"title":"An efficiency study of different model-based and measurement-based connection admission control techniques using heterogeneous traffic sources","authors":"Teck Kiong Lee, M. Zukerman","doi":"10.1109/ATM.1999.786789","DOIUrl":"https://doi.org/10.1109/ATM.1999.786789","url":null,"abstract":"Numerous research has been done on modeling the traffic flows in a multiservice network. An alternative proposal was made by Addie et al. (1998). The authors argued that because the central limit theorem can be applied to heavy traffic processes, therefore as more traffic is aggregated together through the sharing of a transmission link, the aggregated traffic will weakly converge to a Gaussian process. In this paper, we apply this Gaussian model to the connection admission control (CAC) of a multiservice network. We consider a store-and-forward (packet or cell-switched) connection-oriented multiservice network that can be based on the concept of asynchronous transfer mode. We also make efficiency comparison with a CAC based on an effective bandwidth model. Two scenarios were considered: the first assumes that all a priori traffic specifications are known before a CAC acceptance/rejection decision is made, while the second assumes that only the peak rate of a new (requesting admission) connection is declared to the CAC mechanism. The latter scenario entails a measurement-based CAC approach.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128182689","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 presents a new design of an ATM multicast protocol with a dynamic codes injection mechanism for customization of service. We use this mechanism to implement an efficient congestion control of multi-layered encoded video multicast streams.
{"title":"An active congestion control in ATM multicast","authors":"H. Pung, N. Bajrach","doi":"10.1109/ATM.1999.786888","DOIUrl":"https://doi.org/10.1109/ATM.1999.786888","url":null,"abstract":"This paper presents a new design of an ATM multicast protocol with a dynamic codes injection mechanism for customization of service. We use this mechanism to implement an efficient congestion control of multi-layered encoded video multicast streams.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1999-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121882898","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 shows a new class of shared-medium access mechanism, named reset-window with buffer insertion protocol, applied to the ATM customer premises network architecture. The broadband local network consists of a unidirectional slotted ring network with spatial reuse by the destination release procedure. Under this access mechanism two levels of traffic with requested quality of service (QoS) are supported: real-time and non-real-time traffic.
{"title":"Supporting broadband distributed accessing by using a fair control","authors":"M. Soto, S. Sallent","doi":"10.1109/ATM.1999.786880","DOIUrl":"https://doi.org/10.1109/ATM.1999.786880","url":null,"abstract":"This paper shows a new class of shared-medium access mechanism, named reset-window with buffer insertion protocol, applied to the ATM customer premises network architecture. The broadband local network consists of a unidirectional slotted ring network with spatial reuse by the destination release procedure. Under this access mechanism two levels of traffic with requested quality of service (QoS) are supported: real-time and non-real-time traffic.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124799537","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 rate monotonic algorithm has recently been proposed to be used along with traffic shapers to provide quality of service guarantees in ATM networks. A necessary and sufficient condition for admission control has been derived assuming that cell arrivals are periodic (i.e., CBR traffic). For bursty traffic, this criterion is obviously conservative and may significantly reduce bandwidth utilization. In this paper, we present an efficient admission criterion for bursty or VBR sources. We assume that the traffic generated by a VBR source is regulated with a token bucket regulator. Our admission criterion takes into consideration traffic characteristics. Since traffic characteristics change from hop to hop, every network node is required to know how many hops a connection has traversed. Numerical results show that, compared with the conservative criterion which assumes periodic cell arrivals, system utilization can be largely improved using our admission criterion.
{"title":"An efficient admission criterion of traffic-shaped rate monotonic scheduler for VBR traffic","authors":"Tsern-Huei Lee, Yaw-Wen Kuo","doi":"10.1109/ATM.1999.786766","DOIUrl":"https://doi.org/10.1109/ATM.1999.786766","url":null,"abstract":"The rate monotonic algorithm has recently been proposed to be used along with traffic shapers to provide quality of service guarantees in ATM networks. A necessary and sufficient condition for admission control has been derived assuming that cell arrivals are periodic (i.e., CBR traffic). For bursty traffic, this criterion is obviously conservative and may significantly reduce bandwidth utilization. In this paper, we present an efficient admission criterion for bursty or VBR sources. We assume that the traffic generated by a VBR source is regulated with a token bucket regulator. Our admission criterion takes into consideration traffic characteristics. Since traffic characteristics change from hop to hop, every network node is required to know how many hops a connection has traversed. Numerical results show that, compared with the conservative criterion which assumes periodic cell arrivals, system utilization can be largely improved using our admission criterion.","PeriodicalId":266412,"journal":{"name":"IEEE ATM Workshop '99 Proceedings (Cat. No. 99TH8462)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123954364","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：