Pub Date : 2001-05-29DOI: 10.1109/HPSR.2001.923596
Cheng Shan
While weighted fair queueing (WFQ) and many other fair scheduling methods can provide guaranteed worst case delay bounds, they are not scalable to a large number of connections due to per-flow queuing. This paper presents a new packet scheduling method-round-robin priority queuing (RRPQ). RRPQ guarantees a per-session-based delay bound without per-flow queuing. It also avoids many other implementation complexities in fair queuing scheduling. The scalability and low implementation cost of RRPQ make it a practical solution in providing guaranteed delay bounds in high-speed networks.
{"title":"A new scalable and efficient packet scheduling method in high-speed packet switch networks","authors":"Cheng Shan","doi":"10.1109/HPSR.2001.923596","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923596","url":null,"abstract":"While weighted fair queueing (WFQ) and many other fair scheduling methods can provide guaranteed worst case delay bounds, they are not scalable to a large number of connections due to per-flow queuing. This paper presents a new packet scheduling method-round-robin priority queuing (RRPQ). RRPQ guarantees a per-session-based delay bound without per-flow queuing. It also avoids many other implementation complexities in fair queuing scheduling. The scalability and low implementation cost of RRPQ make it a practical solution in providing guaranteed delay bounds in high-speed networks.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"26 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":"121161184","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923621
S. Ohta, A. Greca
The routing and wavelength assignment algorithm is indispensable to operate optical networks, which consist of the wavelength division multiplexing technique and optical cross-connect systems. This paper focuses on the routing and wavelength algorithm that minimizes the number of wavelengths for the static path demand situation. The paper shows that the routing problem and the wavelength assignment problem can be solved simultaneously by employing the multicommodity flow model, which has been comprehensively studied in the literature. On the basis of this notion, a heuristic routing and wavelength assignment algorithm is proposed. Through numerical examples, the proposed algorithm is compared with conventional algorithms that run under the same criteria.
{"title":"Comparison of routing and wavelength assignment algorithms for optical networks","authors":"S. Ohta, A. Greca","doi":"10.1109/HPSR.2001.923621","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923621","url":null,"abstract":"The routing and wavelength assignment algorithm is indispensable to operate optical networks, which consist of the wavelength division multiplexing technique and optical cross-connect systems. This paper focuses on the routing and wavelength algorithm that minimizes the number of wavelengths for the static path demand situation. The paper shows that the routing problem and the wavelength assignment problem can be solved simultaneously by employing the multicommodity flow model, which has been comprehensively studied in the literature. On the basis of this notion, a heuristic routing and wavelength assignment algorithm is proposed. Through numerical examples, the proposed algorithm is compared with conventional algorithms that run under the same criteria.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"85 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":"114472245","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923671
Jinhui Li, N. Ansari
The enhanced Birkhoff-von Neumann decomposition (EBVND) algorithm, a new class-of scheduling algorithms for input queued (IQ) switches, is introduced. Theoretical analysis shows that the performance of EBVND is better than the Birkhoff-von Neumann decomposition algorithm in terms of throughput and cell delay, and can also provide rate and cell delay guarantees. Wave front Birkhoff-von Neumann decomposition (WFBVND) algorithm and its simplified version WFBVND with logN iterations (WFBVND-logN), the special cases of EBVND, are also introduced and evaluated. Simulations show that WFBVND and WFBVND-logN have much lower average cell delay as compared to the Birkhoff-von Neumann decomposition algorithm.
{"title":"QoS guaranteed input queued scheduling algorithms with low delay","authors":"Jinhui Li, N. Ansari","doi":"10.1109/HPSR.2001.923671","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923671","url":null,"abstract":"The enhanced Birkhoff-von Neumann decomposition (EBVND) algorithm, a new class-of scheduling algorithms for input queued (IQ) switches, is introduced. Theoretical analysis shows that the performance of EBVND is better than the Birkhoff-von Neumann decomposition algorithm in terms of throughput and cell delay, and can also provide rate and cell delay guarantees. Wave front Birkhoff-von Neumann decomposition (WFBVND) algorithm and its simplified version WFBVND with logN iterations (WFBVND-logN), the special cases of EBVND, are also introduced and evaluated. Simulations show that WFBVND and WFBVND-logN have much lower average cell delay as compared to the Birkhoff-von Neumann decomposition algorithm.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"36 12 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":"129395942","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923615
V. Oberle, H. Ritter, Klaus Wehrle
We focus on the distribution of pricing information between autonomous systems in the Internet. Therefore we propose the border pricing protocol (BPP) as a protocol for exchanging pricing information in analogy to the border gateway protocol, the current Internet standard for inter-autonomous system routing protocols. The open pricing framework is sketched as well as its relation to the differentiated services architecture. The mechanisms of the BPP are presented and the content of the pricing and charging information base is discussed.
{"title":"BPP: a protocol for exchanging pricing information between autonomous systems","authors":"V. Oberle, H. Ritter, Klaus Wehrle","doi":"10.1109/HPSR.2001.923615","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923615","url":null,"abstract":"We focus on the distribution of pricing information between autonomous systems in the Internet. Therefore we propose the border pricing protocol (BPP) as a protocol for exchanging pricing information in analogy to the border gateway protocol, the current Internet standard for inter-autonomous system routing protocols. The open pricing framework is sketched as well as its relation to the differentiated services architecture. The mechanisms of the BPP are presented and the content of the pricing and charging information base is discussed.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"3 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":"131335497","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923612
X. Hou, H. Mouftah
The implementation of wavelength converters in the optical crossconnect (OXC) is a key step towards the all-optical networks. Optical switches are the building blocks of OXC with the functions of wavelength routing, reconfigurability, network protection and restoration. Wavelength converters are used to effectively alleviate the blocking probability and help to resolve contention at the output port of switches by shifting one incoming wavelength to another outgoing wavelength at an OXC node. The deployment of wavelength converters within optical switches offers robust routing, switching functions and network management in the optical layer. This is critical to the emerging all-optical Internet. However, the high cost of wavelength converters at the current stage of fabrication technology remains an issue when we design node architectures for an optical network. This paper proposes a new wavelength-convertible switch and investigates the most cost-effective use of wavelength converters.
{"title":"Design of wavelength-convertible optical switches for the all-optical next-generation Internet","authors":"X. Hou, H. Mouftah","doi":"10.1109/HPSR.2001.923612","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923612","url":null,"abstract":"The implementation of wavelength converters in the optical crossconnect (OXC) is a key step towards the all-optical networks. Optical switches are the building blocks of OXC with the functions of wavelength routing, reconfigurability, network protection and restoration. Wavelength converters are used to effectively alleviate the blocking probability and help to resolve contention at the output port of switches by shifting one incoming wavelength to another outgoing wavelength at an OXC node. The deployment of wavelength converters within optical switches offers robust routing, switching functions and network management in the optical layer. This is critical to the emerging all-optical Internet. However, the high cost of wavelength converters at the current stage of fabrication technology remains an issue when we design node architectures for an optical network. This paper proposes a new wavelength-convertible switch and investigates the most cost-effective use of wavelength converters.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"197 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":"134461602","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923645
K. Yoshigoe, Ken Christensen
Input buffered switches with virtual output queues (VOQ) are scalable to very high speeds, but require switch matrix scheduling algorithms to achieve high throughput. Existing scheduling algorithms based on parallel request grant-accept cycles cannot natively support variable length Ethernet packets. A parallel-polled VOQ (PP-VOQ) architecture is proposed that natively supports variable length packets. Small amounts of FIFO buffering within a crossbar are used. Using simulation, the PP-VOQ with buffered crossbar switch is shown to have lower switch delay at high offered loads than an iSLIP switch for both cell and variable-length packet traffic. The PP-VOQ switch does not require internal speed-up or complex reassembly mechanisms. The priority mechanism implemented in both the iSLIP and PP-VOQ switches are demonstrated to provide guaranteed rate and bounded delay for schedulable traffic.
{"title":"A parallel-polled virtual output queued switch with a buffered crossbar","authors":"K. Yoshigoe, Ken Christensen","doi":"10.1109/HPSR.2001.923645","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923645","url":null,"abstract":"Input buffered switches with virtual output queues (VOQ) are scalable to very high speeds, but require switch matrix scheduling algorithms to achieve high throughput. Existing scheduling algorithms based on parallel request grant-accept cycles cannot natively support variable length Ethernet packets. A parallel-polled VOQ (PP-VOQ) architecture is proposed that natively supports variable length packets. Small amounts of FIFO buffering within a crossbar are used. Using simulation, the PP-VOQ with buffered crossbar switch is shown to have lower switch delay at high offered loads than an iSLIP switch for both cell and variable-length packet traffic. The PP-VOQ switch does not require internal speed-up or complex reassembly mechanisms. The priority mechanism implemented in both the iSLIP and PP-VOQ switches are demonstrated to provide guaranteed rate and bounded delay for schedulable traffic.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"25 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":"130100968","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923667
G. Riley, M. Ammar, E. Zegura
We introduce the concept of Neighbor-Index Vector (NIx-Vector) routing which gives efficient and consistent routing of packets in Internet routers. A NIx-Vector is a compact representation of a routing path, which is small enough to be included in a packet header. We show how, by including the routing information in the packet header, routing decisions can be made in O(1) time at each router, with no caching or state in the routers (other than the existing routing tables). The creation of a NIx-Vector for a source and destination pair requires one round trip time, but once the NIx-Vector is known, it can be retained by the source and reused indefinitely with no further setup time required (or until it becomes no longer valid, which is easily detected). In addition to O(1) routing decisions, the use of NIx-Vectors to record and specify routes has other benefits. NIx-Vectors provide for route pinning which is beneficial for reservation protocols and mitigates the effect of routing flaps on long lived flows. A variation of NIx-Vectors can insure symmetrical routes from a source to a destination and back to the source, which is also beneficial to some reservation protocols. We give a complete description of how NIx-Vectors are created and used in Internet routers; and give empirical data showing the number of bits required to represent NIx-Vectors for typical Internet paths.
{"title":"Efficient routing using NIx-Vectors","authors":"G. Riley, M. Ammar, E. Zegura","doi":"10.1109/HPSR.2001.923667","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923667","url":null,"abstract":"We introduce the concept of Neighbor-Index Vector (NIx-Vector) routing which gives efficient and consistent routing of packets in Internet routers. A NIx-Vector is a compact representation of a routing path, which is small enough to be included in a packet header. We show how, by including the routing information in the packet header, routing decisions can be made in O(1) time at each router, with no caching or state in the routers (other than the existing routing tables). The creation of a NIx-Vector for a source and destination pair requires one round trip time, but once the NIx-Vector is known, it can be retained by the source and reused indefinitely with no further setup time required (or until it becomes no longer valid, which is easily detected). In addition to O(1) routing decisions, the use of NIx-Vectors to record and specify routes has other benefits. NIx-Vectors provide for route pinning which is beneficial for reservation protocols and mitigates the effect of routing flaps on long lived flows. A variation of NIx-Vectors can insure symmetrical routes from a source to a destination and back to the source, which is also beneficial to some reservation protocols. We give a complete description of how NIx-Vectors are created and used in Internet routers; and give empirical data showing the number of bits required to represent NIx-Vectors for typical Internet paths.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"16 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":"131090134","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923628
S. Van den Bosch, P. de La Vallee-Poussin, G. Van Hoey, H. de Neve, G. Petit
Two mixed-integer linear programming (MILP) formulations of a centralized, off-line traffic engineering tool are compared with respect to their scalability properties. Such an algorithm can be used to optimize the routing configuration in a DiffServ network. It is shown that a considerable reduction of the number of variables and constraints can be obtained by pairing primary and protection path candidates prior to the routing optimization. Simulations of networks with different sizes evidence this result,.
{"title":"Study of scalability properties of two LP-based algorithms for traffic engineering in a flat differentiated services network","authors":"S. Van den Bosch, P. de La Vallee-Poussin, G. Van Hoey, H. de Neve, G. Petit","doi":"10.1109/HPSR.2001.923628","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923628","url":null,"abstract":"Two mixed-integer linear programming (MILP) formulations of a centralized, off-line traffic engineering tool are compared with respect to their scalability properties. Such an algorithm can be used to optimize the routing configuration in a DiffServ network. It is shown that a considerable reduction of the number of variables and constraints can be obtained by pairing primary and protection path candidates prior to the routing optimization. Simulations of networks with different sizes evidence this result,.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"198 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":"131210510","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923624
K. Shiomoto, M. Omotani, M. Uga, S. Shimizu
This paper proposes a label switch router architecture using ATM switch core. Hardware-based forwarding engines are attached to the switch core in a scalable fashion in the proposed label switch router. Dimensioning the number of forwarding engines is proposed to achieve scalable performance. An ATM virtual circuit is used as label switched path in the MPLS network. Traffic management and OAM capabilities are applied to achieve carrier-grade network control for the network infrastructure the next generation Internet is to provide.
{"title":"Parallel packet forwarding architecture using ATM switch core for scalable performance","authors":"K. Shiomoto, M. Omotani, M. Uga, S. Shimizu","doi":"10.1109/HPSR.2001.923624","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923624","url":null,"abstract":"This paper proposes a label switch router architecture using ATM switch core. Hardware-based forwarding engines are attached to the switch core in a scalable fashion in the proposed label switch router. Dimensioning the number of forwarding engines is proposed to achieve scalable performance. An ATM virtual circuit is used as label switched path in the MPLS network. Traffic management and OAM capabilities are applied to achieve carrier-grade network control for the network infrastructure the next generation Internet is to provide.","PeriodicalId":308964,"journal":{"name":"2001 IEEE Workshop on High Performance Switching and Routing (IEEE Cat. No.01TH8552)","volume":"65 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":"130290584","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 : 2001-05-29DOI: 10.1109/HPSR.2001.923629
K. Chandrayana, B. Sikdar, S. Kalyanaraman
Active queue management policies and in particular random early drop (RED) are being pushed as additional mechanisms in Internet routers to control congestion and keep network utilization high. RED's performance is highly dependent on the settings of its control parameters. However, no firm guidelines exist on configuring RED parameters and the current suggestions fail to provide the desired performance scalability. We propose a mechanism to configure RED parameters based on evaluating the expected instantaneous length of a M/sup x//M/1/K queue. We show that by setting the RED parameters min/sub th/ and max/sub th/ to lie on either side of this expected queue length, we can ensure that the queue is not underutilized and flows cut their rates before the onset of congestion. This setting also allows the operating point to perform satisfactorily over a wide range of flow counts thereby allowing for a higher degree of scalability. We also show that our proposed mechanism increases the queue goodput, reduces losses and timeouts and increases the fairness when compared to existing guidelines. Our proposals have been verified using extensive simulations.
{"title":"Scalable configuration of RED queue parameters","authors":"K. Chandrayana, B. Sikdar, S. Kalyanaraman","doi":"10.1109/HPSR.2001.923629","DOIUrl":"https://doi.org/10.1109/HPSR.2001.923629","url":null,"abstract":"Active queue management policies and in particular random early drop (RED) are being pushed as additional mechanisms in Internet routers to control congestion and keep network utilization high. RED's performance is highly dependent on the settings of its control parameters. However, no firm guidelines exist on configuring RED parameters and the current suggestions fail to provide the desired performance scalability. We propose a mechanism to configure RED parameters based on evaluating the expected instantaneous length of a M/sup x//M/1/K queue. We show that by setting the RED parameters min/sub th/ and max/sub th/ to lie on either side of this expected queue length, we can ensure that the queue is not underutilized and flows cut their rates before the onset of congestion. This setting also allows the operating point to perform satisfactorily over a wide range of flow counts thereby allowing for a higher degree of scalability. We also show that our proposed mechanism increases the queue goodput, reduces losses and timeouts and increases the fairness when compared to existing guidelines. Our proposals have been verified using extensive simulations.","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":"130339981","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：