{"title":"A high-speed router featuring minimal delay variation","authors":"M. Collier","doi":"10.1109/HPSR.2001.923653","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923653","url":null,"abstract":"This paper describes a technique for implementing the switch fabric of a high-speed router (with a throughput in excess of 600 Gb/s based on the current slate of the art), with the following properties. Delay performance is virtually identical to that of a standard output-buffered switch, and the switch fabric preserves the packet sequence, so that no resequencing is required for segmented packets. Clock rates are moderate except at ingress and egress points. This is achieved by distributing traffic across a number of crossbar switches operating at a low bit rate. The techniques used to resolve contention in the crossbar switches are described, and the bottlenecks limiting the capacity of the switch are discussed.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131086866","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}
{"title":"An improved wavelength-selective all-fiber cross-connect node","authors":"E. Mutafungwa","doi":"10.1109/HPSR.2001.923611","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923611","url":null,"abstract":"A flexible and transparent optical network, can be realised by deploying fully-optical cross-connect nodes (OXNs). We propose some modifications on previously reported all-fiber OXN. These modification aim to enhance the optical spectral efficiency, modularity and spectral response of the OXN. Simulations of the OXN and its modified version indicate a significant increase in the cascadability achieved by the modified OXN.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122084802","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}
{"title":"Theoretical network load limit when self-similarity has no adverse effect on the network","authors":"H. V. Tran, T. M. Chen","doi":"10.1109/HPSR.2001.923619","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923619","url":null,"abstract":"It is well-known that the multimedia traffic is statistically self-similar, with fractal-like behavior, that is, network traffic exhibits scale-invariance at a wide range of scales. Such scale-invariance is different from traditional models of network traffic. It is also well-known that self-similarity degrades the network performance by requiring large queueing buffers, causing delay and packet dropping problems, and that the traditional queueing theory is inadequate to predict network performance. The conventional wisdom is that the higher the load on the network, the higher the degree of self-similarity. In this paper we first show a method of how to approximate a queueing buffer size with a self-similar input process, then we derive a theoretical network load limit when self-similarity has no adverse effect on the network. This load limit is found to be about the same load value published in literature, on average delay, with actual Ethernet traffic traces compared against curves obtained from the traditional queueing network analyser.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115151695","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}
{"title":"On scheduling real-time traffic under controlled load service in an integrated services Internet","authors":"Hongyuan Shi, H. Sethu","doi":"10.1109/HPSR.2001.923595","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923595","url":null,"abstract":"The controlled load service defined within the IETF's Integrated Services architecture for QoS in the Internet requires source points to regulate the traffic while the network provides a soft guarantee on performance. Packets sent in violation of the traffic are marked so that the network may give them lower priority. We have defined the requirements of a scheduler serving packets belonging to the controlled load service. Besides efficiency and throughput goals, we define another important requirement to bound the additional delay of unmarked packets caused due to the transmission of marked packets. For any given desired bound /spl alpha/ on this additional delay, we present the CL(/spl alpha/) scheduler which achieves the bound while also achieving a per-packet work complexity of O(1). We also provide analytical proofs of these results on the CL(/spl alpha/) scheduler. The principle used in this algorithm can also be used to schedule flows with multilevel priorities, such as in some real-time video streams as well as in other emerging service models of the Internet that mark packets to identify drop precedences.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"270 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124372230","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}
{"title":"Congestion avoidance using adaptive random marking","authors":"Marissa Borrego, Na Li, G. Veciana, San-qi Li","doi":"10.1109/HPSR.2001.923605","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923605","url":null,"abstract":"Recent work on congestion control in TCP/IP networks combines improved end-user transmission mechanisms with active queue management schemes. In this paper we propose MARS (Marking based on Adaptive Random Scheme), an adaptive early packet marking active queue management mechanism. Unlike previous proposals, MARS adapts the marking probability to drive the average queue length towards a target size. By doing so, MARS implicitly reacts to changes in the network dynamics without requiring per-flow state or estimates thereof. Our simulations confirm MARS's flexibility and effectiveness in reducing packet loss, controlling queuing delay, and improving utilization and flow throughput.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127576440","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}
{"title":"Algorithm for traffic grooming in optical networks to minimize the number of transceivers","authors":"V. Konda, T. Chow","doi":"10.1109/HPSR.2001.923635","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923635","url":null,"abstract":"We study the problem of traffic grooming to reduce the number of transceivers in optical networks. We show that this problem is equivalent to a certain traffic maximization problem. We give an intuitive interpretation of this equivalence and use this interpretation to derive a greedy algorithm for transceiver minimization. We discuss implementation issues and present computational results comparing the heuristic solutions with the optimal solutions for several small example networks. For larger networks, the heuristic solutions are compared with known bounds on the optimal solution obtained using integer programming tools.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129001781","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}
{"title":"Optical packet-switched ring network with flexible bandwidth allocation","authors":"A. Smiljanic, M. Boroditsky, N. Frigo","doi":"10.1109/HPSR.2001.923609","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923609","url":null,"abstract":"We propose a novel architecture, MAC protocol and admission control for high-capacity packet-switched ring networks. Nodes access the entire link capacity by transmitting and receiving packets in parallel on all available wavelengths. Packets are added to and dropped from the ring by 2/spl times/2 optical switches. The network control is simple since the load is balanced over the wavelengths at the physical layer. The proposed MAC protocol is based on credits. The admission control algorithm requires minimal processing complexity, and can follow fast traffic changes typical in data networks.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115746307","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}
{"title":"Load balancing algorithms in MPLS traffic engineering","authors":"Keping Long, Zhongshan Zhang, Shiduan Cheng","doi":"10.1109/HPSR.2001.923627","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923627","url":null,"abstract":"This paper present three load-balancing algorithms for multiprotocol label switching (MPLS) traffic engineering: topology-based static load-balancing algorithm (TSLB), resource-based static load-balancing algorithm (RSLB) and dynamic load-balancing algorithm (DLB). We also evaluate the performance of these algorithms using NS-2 simulator. The simulation results show that the performance of our algorithms is better than the traditional shortest path algorithm. Moreover, the performance of RSLB and DLB is much better than the performance of TSLB. Although the proposed three load balancing algorithms are only suitable for non-priority unicast traffic, they can be easily extended for priority traffic conditions. The load balancing algorithms under multicast priority conditions are the subject for further study.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"09 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127306118","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}
{"title":"Architecture and performance of a multi-Tbps protocol independent switching fabric","authors":"Wei Li, Z. Dziong, Y.T. Wang","doi":"10.1109/HPSR.2001.923662","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923662","url":null,"abstract":"We describe a scalable multi-Tbps protocol independent switching fabric architecture. This architecture is cost effective for building switches supporting wirespeed switching up to 1024 ports with capacity ranging from 10's of Gbps to 10's of Tbps and QoS traffic classes. The scalability has been achieved via self-routing of cells with the routing queue number scheme coupled with an efficient backpressure and flow control. We also showed that arbitrarily high utilization of the crossbars can be achieved using a simple output buffering and re-sequencing scheme.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129578933","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}
{"title":"Fast IP table lookup and memory reduction","authors":"Y.-C. Liu, C. Lea","doi":"10.1109/HPSR.2001.923637","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923637","url":null,"abstract":"One of the time-consuming tasks in IP4 packet processing is maximum sequence matching. Fast routing requires tens of millions of routing lookups to be performed in one second. This paper describes an implementation of IP table lookup. The implementation is intended as part of the cord of an OC-192 (10 Gbps) and OC-768 (40 Gbps) rate packet processor. One key element is a memory reduction technique that applies to all lookup algorithms. For algorithms with similar complexity as the one described in the paper, the improvement in terms of memory reduction is about 20%-30%.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122262446","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}