Pub Date : 1997-06-23DOI: 10.1109/HPCS.1997.864031
J.M. Capone, I. Stavrakakis
In this work, call admission and transmis sion scheduling policies are studied for a TDMA system servicing Variable Bit Rate (VBR) applications with distinct QoS requirements and traf fic characteristics. In this environment. packets which experience excess delays are dropped (due to delay violations) at the source. In addition, packets are dropped at the receiver due to channel induced errors (interference) in the wireless link. The focus of the research is to determine the region of achievable QoS vectors for heterogeneous VBR applications in this shared resource environment and study the impact that channel quality has on the achievable performance. The region of achievable QoS vectors is central to the call admission problem and in this work, it is used to identify a class of scheduling policies capable of delivering any achievable performance.
{"title":"Achievable QoS in a shared wireless channel","authors":"J.M. Capone, I. Stavrakakis","doi":"10.1109/HPCS.1997.864031","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864031","url":null,"abstract":"In this work, call admission and transmis sion scheduling policies are studied for a TDMA system servicing Variable Bit Rate (VBR) applications with distinct QoS requirements and traf fic characteristics. In this environment. packets which experience excess delays are dropped (due to delay violations) at the source. In addition, packets are dropped at the receiver due to channel induced errors (interference) in the wireless link. The focus of the research is to determine the region of achievable QoS vectors for heterogeneous VBR applications in this shared resource environment and study the impact that channel quality has on the achievable performance. The region of achievable QoS vectors is central to the call admission problem and in this work, it is used to identify a class of scheduling policies capable of delivering any achievable performance.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123350711","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864035
M. Falkner, M. Devetsiklotis, I. Larnbadaris
A significant difficulty arising when using Monte Carlo (MC) simulation for the performance-analysis of communication networks is the long run times required to obtain accurate statistical estimates. Under the proper conditions, Importance Sampling (IS) is a technique that can speed up simulations involving rare events in network (queueing) systems [1, 2, 3, 4, 5]. Large speed-up factors in simulation run time can be obtained by using IS if the modification or bias of the underlying probability measures of certain random processes is carefully chosen. Fast simulation methods based on Large Deviation Theory [1, 3] have been successfully applied in many cases (recently, most notably in [51].
{"title":"Issues in fast simulation of networks of queues by use of effective and decoupling bandwidths","authors":"M. Falkner, M. Devetsiklotis, I. Larnbadaris","doi":"10.1109/HPCS.1997.864035","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864035","url":null,"abstract":"A significant difficulty arising when using Monte Carlo (MC) simulation for the performance-analysis of communication networks is the long run times required to obtain accurate statistical estimates. Under the proper conditions, Importance Sampling (IS) is a technique that can speed up simulations involving rare events in network (queueing) systems [1, 2, 3, 4, 5]. Large speed-up factors in simulation run time can be obtained by using IS if the modification or bias of the underlying probability measures of certain random processes is carefully chosen. Fast simulation methods based on Large Deviation Theory [1, 3] have been successfully applied in many cases (recently, most notably in [51].","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126703754","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864021
M. May, C. Diot
The Integrated Services Packet Networks are designed to offer new network services to a wide variety of applications. The architecture behind the new services consists primarily of a packet scheduler, a packet classifier, admission control, and a reservation establishment protocol. This paper describes a complete implementation of this Traffic Control extensions. We also designed and implemented a kernel interface which permits to send Quality of Service (QoS) requests to the network layer. We explain how the elements of the architecture work in concert and which mechanisms we use to guarantee a requested amount of bandwidth. Our scheduler uses a variant of the Class Based Queuing mechanism (CBQ) of Sally Floyd and Van Jacobsen [6]. For the reservation signaling and establishment we use RSVP [2]. Finally this paper provides a performance evaluation of the proposed mechanisms.
{"title":"An experimental implementation of traffic control for IP networks","authors":"M. May, C. Diot","doi":"10.1109/HPCS.1997.864021","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864021","url":null,"abstract":"The Integrated Services Packet Networks are designed to offer new network services to a wide variety of applications. The architecture behind the new services consists primarily of a packet scheduler, a packet classifier, admission control, and a reservation establishment protocol. This paper describes a complete implementation of this Traffic Control extensions. We also designed and implemented a kernel interface which permits to send Quality of Service (QoS) requests to the network layer. We explain how the elements of the architecture work in concert and which mechanisms we use to guarantee a requested amount of bandwidth. Our scheduler uses a variant of the Class Based Queuing mechanism (CBQ) of Sally Floyd and Van Jacobsen [6]. For the reservation signaling and establishment we use RSVP [2]. Finally this paper provides a performance evaluation of the proposed mechanisms.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"6 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120855406","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864037
Guang-Liang Li, Jun-hong Cui, Fang-Ming Li
In this paper, we discuss defining, computing, and guaranteeing transient loss performance in future high performance packet networks. The results apply to any fluid-type traffic model and do not rely on assumptions such as infinite connection duration, infinite buffer size, and Markov traffic processes.
{"title":"On defining, computing and guaranteeing transient loss in high-performance packet networks","authors":"Guang-Liang Li, Jun-hong Cui, Fang-Ming Li","doi":"10.1109/HPCS.1997.864037","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864037","url":null,"abstract":"In this paper, we discuss defining, computing, and guaranteeing transient loss performance in future high performance packet networks. The results apply to any fluid-type traffic model and do not rely on assumptions such as infinite connection duration, infinite buffer size, and Markov traffic processes.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128790313","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864028
Z. Zhang, I. Habib, T. Saadawi
Supporting multimedia traffic over wireless chan nels in a cellular network requires effective control of the channel's limited capacity. The challenge is to guarantee different quality of service (QoS) parameters for different types of traffic while achieving high throughput. We propose and study the performance of a bandwidth reservation multiple access (BRMA) protocol in micro (or pico) cellular networks where the propagation delays are insignificant. The admission and bandwidth allocation functionality is realized through a neural network (NN) according the available bandwidth, traffic patterns and QoS parameters of the new calls. An optimal frame length that controls the scheduling of packets transmission from multiple users is determined such that a worst-case packet loss rate or delay limit can be guaranteed while the throughput is maximized. A service ratio is determined such that the number of transmission slots/frame allocated to each user (group of users) guarantees an "minimum" service rate. This protocol achieves satisfactory performance due to the bandwidth allocation on a call level and the efficient collision-free transmission scheduling.
{"title":"Supporting multimedia with bandwidth reservation multiple access protocol","authors":"Z. Zhang, I. Habib, T. Saadawi","doi":"10.1109/HPCS.1997.864028","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864028","url":null,"abstract":"Supporting multimedia traffic over wireless chan nels in a cellular network requires effective control of the channel's limited capacity. The challenge is to guarantee different quality of service (QoS) parameters for different types of traffic while achieving high throughput. We propose and study the performance of a bandwidth reservation multiple access (BRMA) protocol in micro (or pico) cellular networks where the propagation delays are insignificant. The admission and bandwidth allocation functionality is realized through a neural network (NN) according the available bandwidth, traffic patterns and QoS parameters of the new calls. An optimal frame length that controls the scheduling of packets transmission from multiple users is determined such that a worst-case packet loss rate or delay limit can be guaranteed while the throughput is maximized. A service ratio is determined such that the number of transmission slots/frame allocated to each user (group of users) guarantees an \"minimum\" service rate. This protocol achieves satisfactory performance due to the bandwidth allocation on a call level and the efficient collision-free transmission scheduling.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129509428","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864038
D. Serpanos, M. Katevenis, E. Spyridakis
ATLAS I is a single-chip, high-performance ATM switch that implements an innovative, advanced ar chitecture with optional multilane (per VC) credit based (backpressure) flow control for data (e.g., ABR) traffic, while allowing use of rate-based flow control wherever desired. The chip is being developed as a next generation gigabit ATM switch under ACTS project "ASICCOM" of the European Union. We describe the architecture of ATLAS I with emphasis on its use as a building block for universal high-speed networking, i.e. for use as a general purpose ATM switch core for the wide, local, and system area. Its single-chip fabrication targets low cost, which will enable wider deployment of ATM technology in all areas of interconnection. Adoption of the optional credit-based flow control can result inefficient, high-speed data networks, that provide improved performance characteristics when compared to wormhole networks. We present simulation re sults indicating that ATLAS I based networks can provide significantly improved delay and through put characteristics over wormhole networks, especiatly under bursty and hot-spot traffic, which is very common in distributed and parallel processing applications.
{"title":"ATLAS I: building block for ATM networks with credit-based flow control","authors":"D. Serpanos, M. Katevenis, E. Spyridakis","doi":"10.1109/HPCS.1997.864038","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864038","url":null,"abstract":"ATLAS I is a single-chip, high-performance ATM switch that implements an innovative, advanced ar chitecture with optional multilane (per VC) credit based (backpressure) flow control for data (e.g., ABR) traffic, while allowing use of rate-based flow control wherever desired. The chip is being developed as a next generation gigabit ATM switch under ACTS project \"ASICCOM\" of the European Union. We describe the architecture of ATLAS I with emphasis on its use as a building block for universal high-speed networking, i.e. for use as a general purpose ATM switch core for the wide, local, and system area. Its single-chip fabrication targets low cost, which will enable wider deployment of ATM technology in all areas of interconnection. Adoption of the optional credit-based flow control can result inefficient, high-speed data networks, that provide improved performance characteristics when compared to wormhole networks. We present simulation re sults indicating that ATLAS I based networks can provide significantly improved delay and through put characteristics over wormhole networks, especiatly under bursty and hot-spot traffic, which is very common in distributed and parallel processing applications.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131953225","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864029
O. Sharon, E. Altman
Polling schemes are an important class of MAC protocols for Wireless LANs. A major drawback of these schemes is their inefficiency when only a small number of mobile stations have packets to transmit. In this paper we suggest a new polling TVIAC which exploits the capture phenomena and enables simultaneous polling and information packets transmissions. Simulation results show that the new TMAC overcomes the above inefficiency considerably.
{"title":"An efficient polling MAC for wireless LANs","authors":"O. Sharon, E. Altman","doi":"10.1109/HPCS.1997.864029","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864029","url":null,"abstract":"Polling schemes are an important class of MAC protocols for Wireless LANs. A major drawback of these schemes is their inefficiency when only a small number of mobile stations have packets to transmit. In this paper we suggest a new polling TVIAC which exploits the capture phenomena and enables simultaneous polling and information packets transmissions. Simulation results show that the new TMAC overcomes the above inefficiency considerably.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114211833","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864020
M. Zitterbart, T. Harbaum, D. Meier, D. Brokelmann
High performance internetworking units, such as routers and bridges are key building blocks of current and emerging high performance networks. The aggregate data rate flowing through a router can be in the range of gigabits per second. In order to keep up with very high data rates, processing within the internetworking unit needs to be optimized. This paper concentrates on routing table look up. It presents a design for a routing table look up for gigabit routers running the new version of IP, EPv6. The proposed approach, HeaRT, is viewed as the core engine of an IM router.
{"title":"Efficient routing table lookup for IPv6","authors":"M. Zitterbart, T. Harbaum, D. Meier, D. Brokelmann","doi":"10.1109/HPCS.1997.864020","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864020","url":null,"abstract":"High performance internetworking units, such as routers and bridges are key building blocks of current and emerging high performance networks. The aggregate data rate flowing through a router can be in the range of gigabits per second. In order to keep up with very high data rates, processing within the internetworking unit needs to be optimized. This paper concentrates on routing table look up. It presents a design for a routing table look up for gigabit routers running the new version of IP, EPv6. The proposed approach, HeaRT, is viewed as the core engine of an IM router.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114279603","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864023
B. Rossi, L. Delgrossi, D. Ferrari
This paper considers potential applications of a toolkit for the creation and management of virtual networks by their users. First, it shows the need for such a toolkit, and introduces the notion of supranet as the result that the application of the toolkit may yield. Then, it briefly examines the characteristics of the groups that may find the adoption of a supranet useful and convenient. Finally, it presents a complete supranet application scenario.
{"title":"The applications of a toolkit for virtual network creation and management","authors":"B. Rossi, L. Delgrossi, D. Ferrari","doi":"10.1109/HPCS.1997.864023","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864023","url":null,"abstract":"This paper considers potential applications of a toolkit for the creation and management of virtual networks by their users. First, it shows the need for such a toolkit, and introduces the notion of supranet as the result that the application of the toolkit may yield. Then, it briefly examines the characteristics of the groups that may find the adoption of a supranet useful and convenient. Finally, it presents a complete supranet application scenario.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121614105","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 : 1997-06-23DOI: 10.1109/HPCS.1997.864044
C. Schuler
Forward error correction coding (FEC) is a widely used method to improve transmission quality in modern communication systems. This paperpresents some aspects on the application of advanced error correction con cepts to recent wireless ATM developments. The target bit error rate and the mapping of the wireless connection quality to the ATM service classes are discussed. Some considerations on the issue of code selection with respect to block size, scalability, complexity and performance are presented. We propose methodsfor comparing the performance of codes at the example of various convolutional codes and Reed-Solomon codes. The impact of the radio environment on the FEC codes is addressed and the code performance is evaluated by simulations using a Gilbert Elliott channel model.
{"title":"Error correction strategies for wireless ATM","authors":"C. Schuler","doi":"10.1109/HPCS.1997.864044","DOIUrl":"https://doi.org/10.1109/HPCS.1997.864044","url":null,"abstract":"Forward error correction coding (FEC) is a widely used method to improve transmission quality in modern communication systems. This paperpresents some aspects on the application of advanced error correction con cepts to recent wireless ATM developments. The target bit error rate and the mapping of the wireless connection quality to the ATM service classes are discussed. Some considerations on the issue of code selection with respect to block size, scalability, complexity and performance are presented. We propose methodsfor comparing the performance of codes at the example of various convolutional codes and Reed-Solomon codes. The impact of the radio environment on the FEC codes is addressed and the code performance is evaluated by simulations using a Gilbert Elliott channel model.","PeriodicalId":178651,"journal":{"name":"The Fourth IEEE Workshop on High-Performance Communication Systems","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1997-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116826850","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}