Pub Date : 2002-11-07DOI: 10.1109/HPSR.2002.1024232
Jinchul Kim, Younghee Lee
Gaining lessons from the IntServ/RSVP experiences, more simplified differentiated services (DiffServ) has emerged as the core network architecture with enhanced scalability for implementation and deployment. In order to provision differentiated services on a DiffServ domain, it is necessary to provision resources appropriately in the control or management plane of the DiffServ domain, in addition to the data plane service protocols implementation. But, efforts in the control or management plane are not matured in the DiffServ domain. While the static provisioning may be sufficient for the quantitative traffic, the qualitative provisioning is more difficult to support, so that the provisioning parameters should be estimated based on the accurate traffic characteristics, which can be obtained from possibly real time measurements. The paper aims to resolve dynamic resource provisioning, especially for the qualitative traffic by measurement-based admission control scheme having some implicit signaling component using control plane DSCPs. The proposed scheme has been simulated on the ns-2 simulator, using traffic with burst characteristics. The results shows that the proposed scheme has the capability to support a fine-grained, dynamic admission control with flexibility in DiffServ domain.
{"title":"A dynamic admission control scheme in a DiffServ domain","authors":"Jinchul Kim, Younghee Lee","doi":"10.1109/HPSR.2002.1024232","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024232","url":null,"abstract":"Gaining lessons from the IntServ/RSVP experiences, more simplified differentiated services (DiffServ) has emerged as the core network architecture with enhanced scalability for implementation and deployment. In order to provision differentiated services on a DiffServ domain, it is necessary to provision resources appropriately in the control or management plane of the DiffServ domain, in addition to the data plane service protocols implementation. But, efforts in the control or management plane are not matured in the DiffServ domain. While the static provisioning may be sufficient for the quantitative traffic, the qualitative provisioning is more difficult to support, so that the provisioning parameters should be estimated based on the accurate traffic characteristics, which can be obtained from possibly real time measurements. The paper aims to resolve dynamic resource provisioning, especially for the qualitative traffic by measurement-based admission control scheme having some implicit signaling component using control plane DSCPs. The proposed scheme has been simulated on the ns-2 simulator, using traffic with burst characteristics. The results shows that the proposed scheme has the capability to support a fine-grained, dynamic admission control with flexibility in DiffServ domain.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128823794","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024205
M. Nabeshima
In stateless core (SCORE) networks, edge routers maintain per-flow state while core routers do not. Core stateless fair queuing (CSFQ) has been proposed for approximating the operation of per-flow queuing techniques in SCORE networks. However, the packet dropping probability offered by CSFQ suits only UDP flows. Thus, CSFQ cannot achieve fair bandwidth allocation for TCP flows. This paper proposes adaptive CSFQ (ACSFQ). The packet dropping probability in ACSFQ is determined adaptively based on the flow arrival rate, the fair share rate, and the current queue length. It well supports TCP flows as well as UDP flows. We compare ACSFQ to CSFQ in terms of fair bandwidth allocation.
{"title":"Adaptive CSFQ: a new fair queuing mechanism for SCORE networks","authors":"M. Nabeshima","doi":"10.1109/HPSR.2002.1024205","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024205","url":null,"abstract":"In stateless core (SCORE) networks, edge routers maintain per-flow state while core routers do not. Core stateless fair queuing (CSFQ) has been proposed for approximating the operation of per-flow queuing techniques in SCORE networks. However, the packet dropping probability offered by CSFQ suits only UDP flows. Thus, CSFQ cannot achieve fair bandwidth allocation for TCP flows. This paper proposes adaptive CSFQ (ACSFQ). The packet dropping probability in ACSFQ is determined adaptively based on the flow arrival rate, the fair share rate, and the current queue length. It well supports TCP flows as well as UDP flows. We compare ACSFQ to CSFQ in terms of fair bandwidth allocation.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129141850","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024209
Yihan Li, S. Panwar, H. J. Chao
Virtual output queuing is widely used by fixed-length high-speed switches to overcome head-of-line blocking. This is done by means of matching algorithms. Maximum matching algorithms have good performance, but their implementation complexity is quite high. Maximal matching algorithms need speedup to guarantee good performance. Iterative algorithms (such as PIM and iSLIP) use multiple iterations to converge on a maximal match. The dual round-robin matching (DRRM) scheme has performance similar to iSLIP and lower implementation complexity. The objective of matching algorithms is to reduce the matching overhead for each time slot. In this paper we present the exhaustive service dual round-robin matching (EDRRM) algorithm, which amortizes the cost of a match over multiple time slots. While EDRRM suffers from a throughput below 100% for small switch sizes, it is conjectured to achieve an asymptotic 100% throughput under uniform traffic. Simulations show that it achieves high throughput under nonuniform traffic. Its delay performance is not sensitive to traffic burstiness, switch size and packet length. In an EDRRM switch cells belonging to the same packet are transferred to the output continuously, which leads to good packet delay performance and simplifies the implementation of packet reassembly.
{"title":"The dual round robin matching switch with exhaustive service","authors":"Yihan Li, S. Panwar, H. J. Chao","doi":"10.1109/HPSR.2002.1024209","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024209","url":null,"abstract":"Virtual output queuing is widely used by fixed-length high-speed switches to overcome head-of-line blocking. This is done by means of matching algorithms. Maximum matching algorithms have good performance, but their implementation complexity is quite high. Maximal matching algorithms need speedup to guarantee good performance. Iterative algorithms (such as PIM and iSLIP) use multiple iterations to converge on a maximal match. The dual round-robin matching (DRRM) scheme has performance similar to iSLIP and lower implementation complexity. The objective of matching algorithms is to reduce the matching overhead for each time slot. In this paper we present the exhaustive service dual round-robin matching (EDRRM) algorithm, which amortizes the cost of a match over multiple time slots. While EDRRM suffers from a throughput below 100% for small switch sizes, it is conjectured to achieve an asymptotic 100% throughput under uniform traffic. Simulations show that it achieves high throughput under nonuniform traffic. Its delay performance is not sensitive to traffic burstiness, switch size and packet length. In an EDRRM switch cells belonging to the same packet are transferred to the output continuously, which leads to good packet delay performance and simplifies the implementation of packet reassembly.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116769566","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024233
Timucin Ozugur, Dominique Verchere
In this paper, we propose an upstream resource management propagation (URMP) algorithm for optical burst switching to provide congestion control in the GMPLS control-plane infrastructure. The URMP establishes the transmission control between congested OBS node and the ingress edge routers. The congested node includes an ingress edge router in the URMP algorithm if any of the edge router's label switched paths participate in the congestion. The congested OBS node transmits a URMP message upstream for the involving ingress edge routers, and sets up the URMP states at the OBS nodes throughout the path. Using the URMP states at the OBS nodes, the related congestion events may merge into one congestion. At the receipt of the URMP message, ingress edge routers change their transmissions from asynchronous to slot-based transmission with controlled data rate.
{"title":"Upstream resource management propagation algorithm for optical burst switching","authors":"Timucin Ozugur, Dominique Verchere","doi":"10.1109/HPSR.2002.1024233","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024233","url":null,"abstract":"In this paper, we propose an upstream resource management propagation (URMP) algorithm for optical burst switching to provide congestion control in the GMPLS control-plane infrastructure. The URMP establishes the transmission control between congested OBS node and the ingress edge routers. The congested node includes an ingress edge router in the URMP algorithm if any of the edge router's label switched paths participate in the congestion. The congested OBS node transmits a URMP message upstream for the involving ingress edge routers, and sets up the URMP states at the OBS nodes throughout the path. Using the URMP states at the OBS nodes, the related congestion events may merge into one congestion. At the receipt of the URMP message, ingress edge routers change their transmissions from asynchronous to slot-based transmission with controlled data rate.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116502929","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024250
D. Schupke
We consider the off-line routing of lightpaths in WDM networks where the nodes individually provide either full, partial or no wavelength conversion. We develop a flexible network model and use it to find the lightpaths in a national-scale network for different demand patterns by a flow-based approach. Our results indicate that due to low network utilization no or few converters in the network can yield the same performance as full wavelength conversion in all the nodes.
{"title":"Off-line lightpath routing in WDM networks with different wavelength converter configurations","authors":"D. Schupke","doi":"10.1109/HPSR.2002.1024250","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024250","url":null,"abstract":"We consider the off-line routing of lightpaths in WDM networks where the nodes individually provide either full, partial or no wavelength conversion. We develop a flexible network model and use it to find the lightpaths in a national-scale network for different demand patterns by a flow-based approach. Our results indicate that due to low network utilization no or few converters in the network can yield the same performance as full wavelength conversion in all the nodes.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121971714","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024228
Jeng-Farn Lee, Yeali S. Sun, Meng Chang Chen
Maximum rate control in a shared channel is important to service providers and carriers for various reasons. Previous approaches either used a concatenation of regulator and scheduler, which employs two set of queues and two management systems, or a policer in front of scheduler. The former requires extra buffer space and high overhead, and the latter causes inaccuracy. In this paper, we propose a new scheduling algorithm, called WF/sup 2/Q-M (worst-case fair weighted fair queueing with maximum rate control), to simultaneously support maximum rate control and provide minimum service rate guarantee. WF/sup 2/Q-M employs a WF/sup 2/Q like scheduler without policer or regulator that it is designed to provide accurate scheduling with low overhead. We prove that in WF/sup 2/Q-M the packet's eligible time can be merged into its virtual starting time, and propose virtual clock adjustment to distribute the excess bandwidth of saturated sessions to other sessions without recalculating their virtual starting and finishing times. We also prove that WF/sup 2/Q-M performance is theoretically bounded by a fluid reference mode, and experiments show WF/sup 2/Q-M performs just as claimed.
{"title":"On maximum rate control of worst-case weighted fair queueing","authors":"Jeng-Farn Lee, Yeali S. Sun, Meng Chang Chen","doi":"10.1109/HPSR.2002.1024228","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024228","url":null,"abstract":"Maximum rate control in a shared channel is important to service providers and carriers for various reasons. Previous approaches either used a concatenation of regulator and scheduler, which employs two set of queues and two management systems, or a policer in front of scheduler. The former requires extra buffer space and high overhead, and the latter causes inaccuracy. In this paper, we propose a new scheduling algorithm, called WF/sup 2/Q-M (worst-case fair weighted fair queueing with maximum rate control), to simultaneously support maximum rate control and provide minimum service rate guarantee. WF/sup 2/Q-M employs a WF/sup 2/Q like scheduler without policer or regulator that it is designed to provide accurate scheduling with low overhead. We prove that in WF/sup 2/Q-M the packet's eligible time can be merged into its virtual starting time, and propose virtual clock adjustment to distribute the excess bandwidth of saturated sessions to other sessions without recalculating their virtual starting and finishing times. We also prove that WF/sup 2/Q-M performance is theoretically bounded by a fluid reference mode, and experiments show WF/sup 2/Q-M performs just as claimed.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116161271","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024633
Y. Yamabayashi, Y. Kanayama, T. Kawai
{"title":"Reliable PLC thermo-optic switches for optical network systems","authors":"Y. Yamabayashi, Y. Kanayama, T. Kawai","doi":"10.1109/HPSR.2002.1024633","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024633","url":null,"abstract":"","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132657474","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024627
P. Yan, T. McLaughlin, Y. Park
ASET's R&D efforts on opto-electronics packaging technology for terabit throughput are reviewed. To establish opto-electronics packaging technologies for high-speed routers and information appliances, we have been researching the following areas; an OE-MCM (Opto-Electronic Multi Chip Module) for low-cost high-speed processing, an AIP (Active-Interposer) for LSIs that increases bit rate (>1.25 Gbps) and makes for a smaller interface with optical signals, and an OE-board with optical right-angled multi-channel connectors for three-dimensional board-level assembly structures. We will demonstrate the following packaging technologies: ( 1 ) Optical waveguide film lamination technique: This technique is versatile and reduces the cost of the OE-MCM. The waveguide film has a 0.3-0.5 dB/c:m loss and is laminated on the printed circuit board. (2) AIP: The driver and receiver ICs, photo diode and laser diode are three-dimensionally stacked and integrated. (3) Self-written Waveguide: The waveguide exploits a unique automatic fiber-coupling method. Optical coupling is complete even if there is a few dozen of microns miss-alignment between fibers. (4) SMOP (Small MultdDemultiplexer consisting of Optical Elements): This is a compact MultVDemultiplexer for four-wavelength CWDM (Coarse Wavelength Division Multiplexing) of the same dimension as the MT ferrule. ( 5 ) Opto-electronic Subrack: This device consists of an optical backplane, fiber bending right-angled multi-channel connectors, optical boards and OE-MCMs. This work was performed under the management of ASET and was supported by NED0 (New Energy and industrial technology Development Organization)
{"title":"Eriang 40 Gb/s full-duplex multi-services router reference system","authors":"P. Yan, T. McLaughlin, Y. Park","doi":"10.1109/HPSR.2002.1024627","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024627","url":null,"abstract":"ASET's R&D efforts on opto-electronics packaging technology for terabit throughput are reviewed. To establish opto-electronics packaging technologies for high-speed routers and information appliances, we have been researching the following areas; an OE-MCM (Opto-Electronic Multi Chip Module) for low-cost high-speed processing, an AIP (Active-Interposer) for LSIs that increases bit rate (>1.25 Gbps) and makes for a smaller interface with optical signals, and an OE-board with optical right-angled multi-channel connectors for three-dimensional board-level assembly structures. We will demonstrate the following packaging technologies: ( 1 ) Optical waveguide film lamination technique: This technique is versatile and reduces the cost of the OE-MCM. The waveguide film has a 0.3-0.5 dB/c:m loss and is laminated on the printed circuit board. (2) AIP: The driver and receiver ICs, photo diode and laser diode are three-dimensionally stacked and integrated. (3) Self-written Waveguide: The waveguide exploits a unique automatic fiber-coupling method. Optical coupling is complete even if there is a few dozen of microns miss-alignment between fibers. (4) SMOP (Small MultdDemultiplexer consisting of Optical Elements): This is a compact MultVDemultiplexer for four-wavelength CWDM (Coarse Wavelength Division Multiplexing) of the same dimension as the MT ferrule. ( 5 ) Opto-electronic Subrack: This device consists of an optical backplane, fiber bending right-angled multi-channel connectors, optical boards and OE-MCMs. This work was performed under the management of ASET and was supported by NED0 (New Energy and industrial technology Development Organization)","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124859268","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024248
J. Karvo, S. Aalto, J. Virtamo
We present two new algorithms for calculating call blocking probabilities for multi-layer multicast streams with the assumption that blocked calls are lost. Users may join and leave the multicast connections freely, thus creating dynamic multicast trees. We define the state space, and give two recursive algorithms; for the general case and for the special case where all multicast channels are statistically indistinguishable. Our recursive algorithms are linear with respect to the number of links. The special case is also polynomial with respect to the number of channels.
{"title":"Blocking probabilities of multi-layer multicast streams","authors":"J. Karvo, S. Aalto, J. Virtamo","doi":"10.1109/HPSR.2002.1024248","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024248","url":null,"abstract":"We present two new algorithms for calculating call blocking probabilities for multi-layer multicast streams with the assumption that blocked calls are lost. Users may join and leave the multicast connections freely, thus creating dynamic multicast trees. We define the state space, and give two recursive algorithms; for the general case and for the special case where all multicast channels are statistically indistinguishable. Our recursive algorithms are linear with respect to the number of links. The special case is also polynomial with respect to the number of channels.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128675380","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 : 2002-11-07DOI: 10.1109/HPSR.2002.1024632
T. Ichikawa
{"title":"IEEE 802.11a-compliant high-speed wireless LAN","authors":"T. Ichikawa","doi":"10.1109/HPSR.2002.1024632","DOIUrl":"https://doi.org/10.1109/HPSR.2002.1024632","url":null,"abstract":"","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126907646","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}