Pub Date : 2011-07-04DOI: 10.1109/HPSR.2011.5986002
C. Minkenberg, R. Luijten, G. Rodríguez
Based on a realistic, yet simple cost model, we compute the switch radix that minimizes the cost of a fat tree network to support a given number of end nodes. The cost model comprises two parameters indicating the relative cost of a crosspoint vs. a link, and the crosspoint-independent base cost of a switch. These parameters can be adapted to represent a given technology used to implement links and switches. Based on these inputs, the resulting model allows a quick evaluation of the switch radix that minimizes the overall cost of the network. We demonstrate that the optimum radix depends most strongly on the relative cost of a link, and turns out to be largely independent of the network size. Using a first-order cost bounds analysis based on current CMOS and link technology, our model indicates that the optimum switch radix for large fat trees is driven almost entirely by link cost and as a result lies in the range of hundreds of ports, rather than the tens of ports being offered today by most commercial switch products today.
{"title":"On the optimum switch radix in fat tree networks","authors":"C. Minkenberg, R. Luijten, G. Rodríguez","doi":"10.1109/HPSR.2011.5986002","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986002","url":null,"abstract":"Based on a realistic, yet simple cost model, we compute the switch radix that minimizes the cost of a fat tree network to support a given number of end nodes. The cost model comprises two parameters indicating the relative cost of a crosspoint vs. a link, and the crosspoint-independent base cost of a switch. These parameters can be adapted to represent a given technology used to implement links and switches. Based on these inputs, the resulting model allows a quick evaluation of the switch radix that minimizes the overall cost of the network. We demonstrate that the optimum radix depends most strongly on the relative cost of a link, and turns out to be largely independent of the network size. Using a first-order cost bounds analysis based on current CMOS and link technology, our model indicates that the optimum switch radix for large fat trees is driven almost entirely by link cost and as a result lies in the range of hundreds of ports, rather than the tens of ports being offered today by most commercial switch products today.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127502354","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986005
P. Pedroso, J. Perelló, M. Klinkowski, D. Careglio, S. Spadaro, J. Solé-Pareta
This paper introduces a Generalized Multi-Protocol Label Switching (GMPLS)-enabled Optical Burst Switched (OBS) network architecture featuring end-to-end QoS-aware burst transport services. The proposed GMPLS/OBS architecture guarantees specific burst loss probability requirements, independently of the offered network load. This is achieved through the configuration of burst Label Switched Paths (LSPs) over the network which are set-up at the GMPLS control plane level only, while it is the OBS node controller the responsible for eventually committing data plane resources upon burst arrival. In such a way, statistical multiplexing feature is preserved and, at the same time, the end-user QoS demands are met by tightly dimensioning the LSPs. A Mixed Integer Linear Programming (MILP) model is devised to provide the optimal burst LSPs dimensioning according to the offered network load and QoS requirements. Moreover, GMPLS-driven capacity reconfigurations are dynamically triggered whenever unfavorable network conditions are detected. The performance of the proposed architecture is extensively assessed on several network scenarios. Future work on the topic is finally outlined.
{"title":"A GMPLS/OBS network architecture enabling QoS-aware end-to-end burst transport","authors":"P. Pedroso, J. Perelló, M. Klinkowski, D. Careglio, S. Spadaro, J. Solé-Pareta","doi":"10.1109/HPSR.2011.5986005","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986005","url":null,"abstract":"This paper introduces a Generalized Multi-Protocol Label Switching (GMPLS)-enabled Optical Burst Switched (OBS) network architecture featuring end-to-end QoS-aware burst transport services. The proposed GMPLS/OBS architecture guarantees specific burst loss probability requirements, independently of the offered network load. This is achieved through the configuration of burst Label Switched Paths (LSPs) over the network which are set-up at the GMPLS control plane level only, while it is the OBS node controller the responsible for eventually committing data plane resources upon burst arrival. In such a way, statistical multiplexing feature is preserved and, at the same time, the end-user QoS demands are met by tightly dimensioning the LSPs. A Mixed Integer Linear Programming (MILP) model is devised to provide the optimal burst LSPs dimensioning according to the offered network load and QoS requirements. Moreover, GMPLS-driven capacity reconfigurations are dynamically triggered whenever unfavorable network conditions are detected. The performance of the proposed architecture is extensively assessed on several network scenarios. Future work on the topic is finally outlined.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131206291","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986024
N. Correia, G. Schütz
Wireless-optical approaches, leveraging on the strengths of both wireless and optical technologies, are seen as promising approaches for future access networks. For wireless-optical access networks to provide geographically continuous wireless coverage, fault-tolerance must be integrated in the design of these networks. While at the optical back end different fiber failure impacts must be considered, at the wireless front end the disruption of wireless communication links must be taken into consideration. Here we address the problem of designing a fault-tolerant multi-radio wireless-optical access network, a problem that is hard to solve. Our goal here is to extract guidelines for the reduction of such complex problem, which is done through the analysis of wireless and optical failure impacts on network congestion. Two network scenarios, having different risk groups for gateways, are analysed.
{"title":"A problem reduction approach for the design of fault-tolerant wireless-optical access networks","authors":"N. Correia, G. Schütz","doi":"10.1109/HPSR.2011.5986024","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986024","url":null,"abstract":"Wireless-optical approaches, leveraging on the strengths of both wireless and optical technologies, are seen as promising approaches for future access networks. For wireless-optical access networks to provide geographically continuous wireless coverage, fault-tolerance must be integrated in the design of these networks. While at the optical back end different fiber failure impacts must be considered, at the wireless front end the disruption of wireless communication links must be taken into consideration. Here we address the problem of designing a fault-tolerant multi-radio wireless-optical access network, a problem that is hard to solve. Our goal here is to extract guidelines for the reduction of such complex problem, which is done through the analysis of wireless and optical failure impacts on network congestion. Two network scenarios, having different risk groups for gateways, are analysed.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133172211","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986026
Jiajia Liu, Xiaohong Jiang, Hiroki Nishiyama, N. Kato
Two-hop relay is a class of attractive routing protocols for mobile ad hoc networks (MANETs) due to its efficiency and simplicity. This paper extends the conventional two-hop relay and proposes a more general group-based two-hop relay algorithm with redundancy. In such an algorithm with redundancy ƒ and group size g (2HR-(ƒ, g) for short), each packet is delivered to at most ƒ distinct relay nodes and can be accepted by its destination if it is among the group of g packets the destination is currently requesting. The 2HR-(ƒ, g) covers the available two-hop relay protocols as special cases, like the in-order protocols (ƒ ≥ 1, g = 1), the out-of-order protocols with redundancy (ƒ > 1, g = ∞) or without redundancy (ƒ = 1, g = ∞), and it enables a more flexible control of packet delivery process to be made in the challenging MANET environment. A general theoretical framework is further developed to explore how the control parameters ƒ and g affect the expected packet delivery delay in an 2HR-(ƒ, g) MANET, where the important medium contention, interference and traffic contention issues are carefully incorporated into the analysis. Finally, extensive simulation and theoretical results are provided to demonstrate the efficiency of the 2HR-(ƒ, g) scheme and the corresponding theoretical framework.
两跳中继以其高效和简单的特点成为移动自组网中极具吸引力的路由协议。本文对传统的两跳中继进行了扩展，提出了一种更通用的基于组的冗余两跳中继算法。在这种具有冗余度和组大小g(简称2HR-(f, g))的算法中，每个数据包被传送到最多f个不同的中继节点，并且如果它在目的地当前请求的g个数据包组中，则可以被其目的地接受。2HR-(f, g)涵盖了可用的两跳中继协议作为特殊情况，如有序协议(f≥1,g = 1)，冗余无序协议(f > 1, g =∞)或无冗余(f = 1, g =∞)，它能够在具有挑战性的MANET环境中更灵活地控制数据包传输过程。进一步开发了一个通用理论框架，以探索控制参数f和g如何影响2HR-(f, g) MANET中的预期分组传输延迟，其中重要的介质争用、干扰和流量争用问题被仔细地纳入分析。最后，提供了大量的仿真和理论结果来证明2HR-(f, g)方案和相应的理论框架的有效性。
{"title":"Group-based two-hop relay with redundancy in MANETs","authors":"Jiajia Liu, Xiaohong Jiang, Hiroki Nishiyama, N. Kato","doi":"10.1109/HPSR.2011.5986026","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986026","url":null,"abstract":"Two-hop relay is a class of attractive routing protocols for mobile ad hoc networks (MANETs) due to its efficiency and simplicity. This paper extends the conventional two-hop relay and proposes a more general group-based two-hop relay algorithm with redundancy. In such an algorithm with redundancy ƒ and group size g (2HR-(ƒ, g) for short), each packet is delivered to at most ƒ distinct relay nodes and can be accepted by its destination if it is among the group of g packets the destination is currently requesting. The 2HR-(ƒ, g) covers the available two-hop relay protocols as special cases, like the in-order protocols (ƒ ≥ 1, g = 1), the out-of-order protocols with redundancy (ƒ > 1, g = ∞) or without redundancy (ƒ = 1, g = ∞), and it enables a more flexible control of packet delivery process to be made in the challenging MANET environment. A general theoretical framework is further developed to explore how the control parameters ƒ and g affect the expected packet delivery delay in an 2HR-(ƒ, g) MANET, where the important medium contention, interference and traffic contention issues are carefully incorporated into the analysis. Finally, extensive simulation and theoretical results are provided to demonstrate the efficiency of the 2HR-(ƒ, g) scheme and the corresponding theoretical framework.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114405757","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986015
Xiaoliang Wang, Xiaohong Jiang, A. Pattavina
The lack of optical buffer is still one of the main problems that hinder the development of all optical networks. One approach to this problem is to emulate the behavior of optical buffers by using optical switches and fiber delay lines (SDL). Current works on this topic have demonstrated the feasibility of constructing SDL-based First In First Out (FIFO) buffer, Priority buffer, etc. The Last In First Out (LIFO) buffer is another important network component for congestion control and QoS guarantee, and parallel and cascade architectures have been peoposed for the efficient design of such optical buffer [1], [2]. The recent work in [3] showed that it is possible to use M fiber delay lines (FDLs) to construct a LIFO buffer of size B = (3/2) · 2M/2 − 1 and B = 2(M+1)/2 − 1 when M is even and odd, respectively. In this paper, we improve the work in [3] by providing a more efficient construction of SDL-based optical LIFO buffer. We first show that with a single stage feedback structure consisted of one (M + 1) × (M + 1) crossbar switch and M FDLs connecting M outputs of the crossbar back to M its inputs, we are able to construct a LIFO buffer of size B = 2 · 2M/2 − 2 and B = (3/2) · 2(M+1)/2 − 2 when M is even and odd, respectively. This is achieved through adopting a properly delay length setting for each FDL and a careful packets scheduling among FDLs, as well as exploiting the nice function of simultaneous packet reading and witting a FDL can support. We further show that if we adopt a cascade of smaller switches rather than a single (M+1)×(M+1) big switch, the new LIFO design can be implemented with much lower complexity in terms of the total number of basic 2 × 2 switch elements.
光缓冲器的缺乏仍然是阻碍全光网络发展的主要问题之一。解决这个问题的一种方法是通过使用光开关和光纤延迟线(SDL)来模拟光缓冲器的行为。目前关于该主题的工作已经证明了构建基于sdl的先进先出(FIFO)缓冲区、优先级缓冲区等的可行性。后进先出(Last In First Out, LIFO)缓冲器是实现拥塞控制和QoS保证的另一个重要网络组件，为了高效设计这种光缓冲器，人们提出了并行和级联架构[1]，[2]。最近在[3]中的工作表明，当M为偶数和奇数时，可以使用M光纤延迟线(FDLs)分别构建大小为B =(3/2)·2M/2−1和B = 2(M+1)/2−1的LIFO缓冲区。在本文中，我们改进了[3]中的工作，提供了一种更有效的基于sdl的光学LIFO缓冲器结构。我们首先证明了单级反馈结构由一个(M+1) × (M+1)交叉杆开关和M个fdl组成，将交叉杆的M个输出连接回M个输入，当M为偶数和奇数时，我们能够分别构建大小为B = 2·2M/2 - 2和B =(3/2)·2(M+1)/2 - 2的LIFO缓冲区。这是通过为每个FDL采用适当的延迟长度设置和在FDL之间仔细的数据包调度，以及利用FDL可以支持的同时读取和写入数据包的良好功能来实现的。我们进一步表明，如果我们采用一系列较小的开关而不是单个(M+1)×(M+1)大开关，就基本2 × 2开关元件的总数而言，新的LIFO设计可以以更低的复杂性实现。
{"title":"An improved design of optical LIFO buffer with switched delay lines","authors":"Xiaoliang Wang, Xiaohong Jiang, A. Pattavina","doi":"10.1109/HPSR.2011.5986015","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986015","url":null,"abstract":"The lack of optical buffer is still one of the main problems that hinder the development of all optical networks. One approach to this problem is to emulate the behavior of optical buffers by using optical switches and fiber delay lines (SDL). Current works on this topic have demonstrated the feasibility of constructing SDL-based First In First Out (FIFO) buffer, Priority buffer, etc. The Last In First Out (LIFO) buffer is another important network component for congestion control and QoS guarantee, and parallel and cascade architectures have been peoposed for the efficient design of such optical buffer [1], [2]. The recent work in [3] showed that it is possible to use M fiber delay lines (FDLs) to construct a LIFO buffer of size B = (3/2) · 2M/2 − 1 and B = 2(M+1)/2 − 1 when M is even and odd, respectively. In this paper, we improve the work in [3] by providing a more efficient construction of SDL-based optical LIFO buffer. We first show that with a single stage feedback structure consisted of one (M + 1) × (M + 1) crossbar switch and M FDLs connecting M outputs of the crossbar back to M its inputs, we are able to construct a LIFO buffer of size B = 2 · 2M/2 − 2 and B = (3/2) · 2(M+1)/2 − 2 when M is even and odd, respectively. This is achieved through adopting a properly delay length setting for each FDL and a careful packets scheduling among FDLs, as well as exploiting the nice function of simultaneous packet reading and witting a FDL can support. We further show that if we adopt a cascade of smaller switches rather than a single (M+1)×(M+1) big switch, the new LIFO design can be implemented with much lower complexity in terms of the total number of basic 2 × 2 switch elements.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124471769","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986009
O. Pedrola, D. Careglio, M. Klinkowski, J. Solé-Pareta
In this paper, we deal with the physical layer impairments (PLIs) in optical burst switching (OBS). In particular we present a formulation of the routing and regenerator placement and dimensioning (RRPD) problem for a feasible translucent OBS (T-OBS) network architecture. Since addressing the joint RRPD problem results in an extremely complex undertaking, we decouple the problem, and hence, we eventually provide formal models to solve routing and RPD separately in the socalled R+RPD problem. Thus, making use of mixed integer linear programming (MILP) formulations, we first address the routing problem with the aim of minimizing congestion in bottleneck network links, and second, we tackle the issue of performing a sparse placement of electrical regenerators in the network. Since the RPD formulation requires high computational effort for large problem instances, we also propose two alternative heuristic strategies that provide good near-optimal solutions within reasonable time limits. To be precise, we evaluate the trade-off between optimality and complexity provided by these methods. Finally, we conduct a series of simulation experiments on the T-OBS network that prove that the R+RPD strategies effectively deal with burst losses caused by the impact of PLIs, and therefore, ensure that the overall T-OBS network performance remains unaffected.
{"title":"RRPD strategies for a T-OBS network architecture","authors":"O. Pedrola, D. Careglio, M. Klinkowski, J. Solé-Pareta","doi":"10.1109/HPSR.2011.5986009","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986009","url":null,"abstract":"In this paper, we deal with the physical layer impairments (PLIs) in optical burst switching (OBS). In particular we present a formulation of the routing and regenerator placement and dimensioning (RRPD) problem for a feasible translucent OBS (T-OBS) network architecture. Since addressing the joint RRPD problem results in an extremely complex undertaking, we decouple the problem, and hence, we eventually provide formal models to solve routing and RPD separately in the socalled R+RPD problem. Thus, making use of mixed integer linear programming (MILP) formulations, we first address the routing problem with the aim of minimizing congestion in bottleneck network links, and second, we tackle the issue of performing a sparse placement of electrical regenerators in the network. Since the RPD formulation requires high computational effort for large problem instances, we also propose two alternative heuristic strategies that provide good near-optimal solutions within reasonable time limits. To be precise, we evaluate the trade-off between optimality and complexity provided by these methods. Finally, we conduct a series of simulation experiments on the T-OBS network that prove that the R+RPD strategies effectively deal with burst losses caused by the impact of PLIs, and therefore, ensure that the overall T-OBS network performance remains unaffected.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128388570","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986028
A. Bukva, R. Casellas, R. Martínez, R. Muñoz
In this paper we propose a dynamic multi-layer path computation algorithm which favors the re-usage of FA LSPs (Forwarding Adjacency Label Switched Path) and virtual links over a new LSP establishment. To evaluate the algorithm, we explore several approaches for the dynamic configuration of the Virtual Network Topology (VNT) within a multi-layer network (Connection - Oriented Ethernet transport over a Wavelength Switched Optical Network) controlled by a unified GMPLS control plane. Specifically, three VNT approaches are compared: the virtual approach, where some lambda FA LSPs are pre-computed but not established (i.e., virtual TE links), the semi-dynamic approach, where a specific set of lambda FA LSPs is pre-established and advertised as Ethernet TE links (i.e., FA TE links) and the dynamic approach, where there are neither pre-established nor pre-computed lambda FA LSPs.
本文提出了一种动态多层路径计算算法，该算法有利于FA LSP(转发邻接标签交换路径)和虚拟链路的重用，而不是建立新的LSP。为了评估该算法，我们探索了由统一的GMPLS控制平面控制的多层网络(波长交换光网络上的面向连接的以太网传输)中虚拟网络拓扑(VNT)动态配置的几种方法。具体来说，比较了三种VNT方法:虚拟方法，其中一些lambda FA lsp是预先计算的，但没有建立(即虚拟TE链路);半动态方法，其中一组特定的lambda FA lsp是预先建立的，并作为以太网TE链路发布(即FA TE链路);动态方法，其中既没有预先建立也没有预先计算lambda FA lsp。
{"title":"A dynamic on-line path computation algorithm for VNT configuration in GMPLS controlled multi-layer (Ethernet/WSON) network","authors":"A. Bukva, R. Casellas, R. Martínez, R. Muñoz","doi":"10.1109/HPSR.2011.5986028","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986028","url":null,"abstract":"In this paper we propose a dynamic multi-layer path computation algorithm which favors the re-usage of FA LSPs (Forwarding Adjacency Label Switched Path) and virtual links over a new LSP establishment. To evaluate the algorithm, we explore several approaches for the dynamic configuration of the Virtual Network Topology (VNT) within a multi-layer network (Connection - Oriented Ethernet transport over a Wavelength Switched Optical Network) controlled by a unified GMPLS control plane. Specifically, three VNT approaches are compared: the virtual approach, where some lambda FA LSPs are pre-computed but not established (i.e., virtual TE links), the semi-dynamic approach, where a specific set of lambda FA LSPs is pre-established and advertised as Ethernet TE links (i.e., FA TE links) and the dynamic approach, where there are neither pre-established nor pre-computed lambda FA LSPs.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130425647","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986037
Zoran Chicha, L. Milinkovic, A. Smiljanic
The pool of available IPv4 addresses is being depleted, comprising less than 10% of all IPv4 addresses. At the same time, the bit-rates at which packets are transmitted are increasing, and the IP lookup speed must be increased as well. Consequently, the IP lookup algorithms are in the research focus again because the existing solutions were designed for IPv4 addresses, and are not sufficiently scalable. In this paper, we compare FPGA implementations of the balanced parallelized frugal lookup (BPFL) algorithm, and the parallel optimized linear pipeline (POLP) lookup algorithm that efficiently use the memory, and achieve the highest speeds.
{"title":"FPGA implementation of lookup algorithms","authors":"Zoran Chicha, L. Milinkovic, A. Smiljanic","doi":"10.1109/HPSR.2011.5986037","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986037","url":null,"abstract":"The pool of available IPv4 addresses is being depleted, comprising less than 10% of all IPv4 addresses. At the same time, the bit-rates at which packets are transmitted are increasing, and the IP lookup speed must be increased as well. Consequently, the IP lookup algorithms are in the research focus again because the existing solutions were designed for IPv4 addresses, and are not sufficiently scalable. In this paper, we compare FPGA implementations of the balanced parallelized frugal lookup (BPFL) algorithm, and the parallel optimized linear pipeline (POLP) lookup algorithm that efficiently use the memory, and achieve the highest speeds.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116145304","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986025
Xiaozhe Zhang, Xicheng Lu, Jinshu Su, Baosheng Wang, Zexin Lu
Traditional BGP implementation is based on single process or single thread model and not fit for cluster architecture of future core router. We have developed SDBGP, a distributed BGP implementation for future core router that provides excellent performance, reliability and scalability. SDBGP is designed on a fully distributed architecture, which gives equal chance for router nodes to participate in BGP routes computing and storage. SDBGP distributes BGP neighbors among cluster router nodes in a balanced way and improves BGP's performance by parallel processing of BGP neighbors. We deploy SDBGP on a software cluster router with four nodes. Performance testing shows that SDBGP can achieve great scalability in neighbor number and routes computation. It can get almost linear speedup with the increasing of cluster route size.
{"title":"SDBGP: A scalable, distributed BGP routing protocol implementation","authors":"Xiaozhe Zhang, Xicheng Lu, Jinshu Su, Baosheng Wang, Zexin Lu","doi":"10.1109/HPSR.2011.5986025","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986025","url":null,"abstract":"Traditional BGP implementation is based on single process or single thread model and not fit for cluster architecture of future core router. We have developed SDBGP, a distributed BGP implementation for future core router that provides excellent performance, reliability and scalability. SDBGP is designed on a fully distributed architecture, which gives equal chance for router nodes to participate in BGP routes computing and storage. SDBGP distributes BGP neighbors among cluster router nodes in a balanced way and improves BGP's performance by parallel processing of BGP neighbors. We deploy SDBGP on a software cluster router with four nodes. Performance testing shows that SDBGP can achieve great scalability in neighbor number and routes computation. It can get almost linear speedup with the increasing of cluster route size.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120978736","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 : 2011-07-04DOI: 10.1109/HPSR.2011.5986039
Shih-Hsuan Lin, M. Yuang
In this paper, we propose a QoS parallel incremental scheduling (QPIS) algorithm for a variant of an earlier-proposed 10-Gb/s QoS optical packet switching system. The variant system consists of optical pseudo-Banyan space switches, a handful of fiber-delay-line (FDL)-based optical buffers, and four-wave-mixing (FWM) wavelength converters. The QPIS minimizes the loss probability for high-priority packets while maximizing system throughput and satisfying two constraints (switch-contention free, and buffer-contention free). Most notably, we prove that QPIS is incremental, i.e., the computed packet sets within each time slot are monotonically non-decreasing, and then derive the computational complexity. From simulation results that pit the QPIS algorithm against two other sequential algorithms, we show that QPIS outperforms these algorithms on packet loss probability, QoS differentiation, and computational complexity.
{"title":"QoS parallel incremental scheduling for optical pseudo-Banyan packet switching system","authors":"Shih-Hsuan Lin, M. Yuang","doi":"10.1109/HPSR.2011.5986039","DOIUrl":"https://doi.org/10.1109/HPSR.2011.5986039","url":null,"abstract":"In this paper, we propose a QoS parallel incremental scheduling (QPIS) algorithm for a variant of an earlier-proposed 10-Gb/s QoS optical packet switching system. The variant system consists of optical pseudo-Banyan space switches, a handful of fiber-delay-line (FDL)-based optical buffers, and four-wave-mixing (FWM) wavelength converters. The QPIS minimizes the loss probability for high-priority packets while maximizing system throughput and satisfying two constraints (switch-contention free, and buffer-contention free). Most notably, we prove that QPIS is incremental, i.e., the computed packet sets within each time slot are monotonically non-decreasing, and then derive the computational complexity. From simulation results that pit the QPIS algorithm against two other sequential algorithms, we show that QPIS outperforms these algorithms on packet loss probability, QoS differentiation, and computational complexity.","PeriodicalId":269137,"journal":{"name":"2011 IEEE 12th International Conference on High Performance Switching and Routing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121208495","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：