Pub Date : 2009-06-22DOI: 10.1109/HPSR.2009.5307423
T. Coutelen, G. Hébuterne, B. Jaumard
Optical Burst Switching (OBS) has been proposed to increase the resource utilization achieved by Optical Circuit Switching (OCS), which is impaired by its coarse granularity. A crucial step toward OBS maturity deals with the feature of lossless guaranteed transfers. No contention resolution mechanism, even combined with efficient pro-active mechanisms has been able to provide such a guarantee so far. In this paper, we investigate further the “offset based priority” and the “streamline effect” to identify flow isolation patterns. We then propose a specific medium access protocol in order to preserve the isolation of transit bursts over ingress bursts. The isolation patterns are translated into an MILP optimization OBS model that performs routing and wavelength assignment (RWA-OBS). Using the routes obtained with RWA-OBS guarantees asynchronous loss-less transfers. It preserves the statistical multiplexing potential and the asynchronous nature of OBS, and thus addresses the drawbacks of WR-OBS and SOBS which were recently introduced in polymorphous OBS (POBS) (Qiao et al. 2006). Experimental results show the benefits of RWA-OBS over WR-OBS in terms of grade of service, while sidestepping the synchronization issues of SOBS with similar grade of service.
光突发交换(OBS)是为了提高光电路交换(OCS)的资源利用率而提出的。迈向OBS成熟的关键一步是处理无损保证传输的特性。到目前为止，没有任何争用解决机制(即使与高效的主动机制结合使用)能够提供这样的保证。在本文中，我们进一步研究了“基于偏移的优先级”和“流线效应”来识别流动隔离模式。然后，我们提出了一种特定的介质访问协议，以保持传输突发对入口突发的隔离。隔离模式被转换成执行路由和波长分配(RWA-OBS)的MILP优化OBS模型。使用RWA-OBS获得的路由可以保证异步无损传输。它保留了OBS的统计复用潜力和异步特性，从而解决了最近在多态OBS (POBS)中引入的WR-OBS和SOBS的缺点(Qiao et al. 2006)。实验结果表明，RWA-OBS在服务等级方面优于WR-OBS，同时规避了类似服务等级的SOBS的同步问题。
{"title":"An OBS RWA formulation for asynchronous loss-less transfer in OBS networks","authors":"T. Coutelen, G. Hébuterne, B. Jaumard","doi":"10.1109/HPSR.2009.5307423","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307423","url":null,"abstract":"Optical Burst Switching (OBS) has been proposed to increase the resource utilization achieved by Optical Circuit Switching (OCS), which is impaired by its coarse granularity. A crucial step toward OBS maturity deals with the feature of lossless guaranteed transfers. No contention resolution mechanism, even combined with efficient pro-active mechanisms has been able to provide such a guarantee so far. In this paper, we investigate further the “offset based priority” and the “streamline effect” to identify flow isolation patterns. We then propose a specific medium access protocol in order to preserve the isolation of transit bursts over ingress bursts. The isolation patterns are translated into an MILP optimization OBS model that performs routing and wavelength assignment (RWA-OBS). Using the routes obtained with RWA-OBS guarantees asynchronous loss-less transfers. It preserves the statistical multiplexing potential and the asynchronous nature of OBS, and thus addresses the drawbacks of WR-OBS and SOBS which were recently introduced in polymorphous OBS (POBS) (Qiao et al. 2006). Experimental results show the benefits of RWA-OBS over WR-OBS in terms of grade of service, while sidestepping the synchronization issues of SOBS with similar grade of service.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125187203","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307442
R. Bolla, R. Bruschi, G. Lamanna, A. Ranieri
Monolithic IP-routers architectures are becoming inadequate with respect to the increase in traffic volume and network function complexity coming from new services. Current technological constraints prevent this kind of design and constructive approach to go further, while decentralized and modular systems seem to offer a better solution to the quest for the next generation router architecture. In this paper we present a design and implementation of a distributed Linux SW Router (SR), based on the concept of physical separation between different functional modules. A deployment of SRs is used to validate the efficiency and reliability of the proposed solution.
{"title":"Beyond single-box SW router architectures","authors":"R. Bolla, R. Bruschi, G. Lamanna, A. Ranieri","doi":"10.1109/HPSR.2009.5307442","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307442","url":null,"abstract":"Monolithic IP-routers architectures are becoming inadequate with respect to the increase in traffic volume and network function complexity coming from new services. Current technological constraints prevent this kind of design and constructive approach to go further, while decentralized and modular systems seem to offer a better solution to the quest for the next generation router architecture. In this paper we present a design and implementation of a distributed Linux SW Router (SR), based on the concept of physical separation between different functional modules. A deployment of SRs is used to validate the efficiency and reliability of the proposed solution.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114369074","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307437
Marija Antic, A. Smiljanic
In this paper we first describe the load balanced shortest path routing (LB-SPR) protocol. Then, we present the linear program for the optimal capacity allocation, to minimize the resource consumption when LB-SPR is applied in the network. We show that when the load balancing is applied, the capacity allocation problem can be expressed only in terms of the total traffic generated by the network nodes, without the need to know the actual traffic distribution. Then, we compare the resource consumption for the proposed LB-SPR protocol and for the regular shortest path routing (SPR) protocols assuming only the traffic loads that users generate and receive at the network nodes. It will be shown that the proposed LB-SPR protocol needs less network resources. In other words, the LB-SPR protocol can route the same traffic at the lower price than the commonly used SPR protocols.
{"title":"Optimal capacity allocation for load balanced shortest path routing","authors":"Marija Antic, A. Smiljanic","doi":"10.1109/HPSR.2009.5307437","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307437","url":null,"abstract":"In this paper we first describe the load balanced shortest path routing (LB-SPR) protocol. Then, we present the linear program for the optimal capacity allocation, to minimize the resource consumption when LB-SPR is applied in the network. We show that when the load balancing is applied, the capacity allocation problem can be expressed only in terms of the total traffic generated by the network nodes, without the need to know the actual traffic distribution. Then, we compare the resource consumption for the proposed LB-SPR protocol and for the regular shortest path routing (SPR) protocols assuming only the traffic loads that users generate and receive at the network nodes. It will be shown that the proposed LB-SPR protocol needs less network resources. In other words, the LB-SPR protocol can route the same traffic at the lower price than the commonly used SPR protocols.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"259 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114529721","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307443
Minjing Mao, Bin Wu, K. Yeung
We study survivable broadcast in high-speed networks against a single link/node failure. We follow the classic approach of blue/red tree [1] to construct a pair of spanning trees (i.e. a blue tree and a red tree) such that the connectivity between the root and an arbitrary node is ensured (via at least one tree) in the presence of a single link/node failure. To ensure that the blue and red trees are constructed in a cost-efficient way, heuristic algorithms have been designed to minimize the cost involved in tree construction. In this paper, we tackle the same problem but resorting to Integer Linear Programming (ILP) for optimal solutions. Two efficient ILPs are formulated, one for protecting against single link failure (MinCost-E) and the other for node failure (MinCost-V). Numerical results show that our ILPs can generate optimal solutions in relatively short amount of time. As compared with the existing heuristic algorithms, we observe a significant improvement in performance.
{"title":"ILP-based design of survivable broadcast trees","authors":"Minjing Mao, Bin Wu, K. Yeung","doi":"10.1109/HPSR.2009.5307443","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307443","url":null,"abstract":"We study survivable broadcast in high-speed networks against a single link/node failure. We follow the classic approach of blue/red tree [1] to construct a pair of spanning trees (i.e. a blue tree and a red tree) such that the connectivity between the root and an arbitrary node is ensured (via at least one tree) in the presence of a single link/node failure. To ensure that the blue and red trees are constructed in a cost-efficient way, heuristic algorithms have been designed to minimize the cost involved in tree construction. In this paper, we tackle the same problem but resorting to Integer Linear Programming (ILP) for optimal solutions. Two efficient ILPs are formulated, one for protecting against single link failure (MinCost-E) and the other for node failure (MinCost-V). Numerical results show that our ILPs can generate optimal solutions in relatively short amount of time. As compared with the existing heuristic algorithms, we observe a significant improvement in performance.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"557 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132303700","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307422
Qiang Wu, T. Wolf
Packet forwarding operations in network systems are often performed in software so that routers can be updated as new protocols and service features are developed. To meet the processing demands of high-performance networks, multi-processor systems-on-a-chip with dozens of cores are employed to provide raw processing power. Management of these processors and other system resources to achieve high forwarding rates is a key challenge. In particular, the allocation of processing workloads and the placement of data structures in memory have an enormous impact on system performance. Our work proposes a runtime system that manages these system resources. Much related work has proposed the use of cache memory hierarchies in packet processors. In this work, we show that our dynamic placement strategy can outperform a conventional cache memory and achieve up to 1.77 times higher hit rates for small memories, which are typically found in packet processing systems.
{"title":"Runtime resource allocation in multi-core packet processing systems","authors":"Qiang Wu, T. Wolf","doi":"10.1109/HPSR.2009.5307422","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307422","url":null,"abstract":"Packet forwarding operations in network systems are often performed in software so that routers can be updated as new protocols and service features are developed. To meet the processing demands of high-performance networks, multi-processor systems-on-a-chip with dozens of cores are employed to provide raw processing power. Management of these processors and other system resources to achieve high forwarding rates is a key challenge. In particular, the allocation of processing workloads and the placement of data structures in memory have an enormous impact on system performance. Our work proposes a runtime system that manages these system resources. Much related work has proposed the use of cache memory hierarchies in packet processors. In this work, we show that our dynamic placement strategy can outperform a conventional cache memory and achieve up to 1.77 times higher hit rates for small memories, which are typically found in packet processing systems.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127241944","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307420
R. Maniyar, P. Ghosh, Arunabha Sen
Multicast communication is an efficient method of data transmission and distribution among a group, especially when network resources are inadequate and needs to be shared. Fair share of network resources, such as, bandwidth, is desirable in such cases. Although there has been an intensive research effort to design protocols and construct multicast routing graphs for a single multicast group, construction of multiple multicast groups and the fair allocation of network resources remains virtually unexplored. In this paper, a unified approach for the Multiple Multicast Tree Construction and Rate Allocation (MMTCRA) problem is addressed. The MMTCRA problem has been defined as an optimization problem with an objective of finding a Max-Min Fair rate allocation among the multiple multicast groups that co-exist in the network subject to the link-capacity constraints. The problem is proved to be NP-Complete. A Mixed Integer Linear Program (MILP) is formulated to achieve the optimal solution for this problem. A heuristic is proposed to solve the MMTCRA problem in polynomial time. The quality of the heuristic is evaluated by comparing the solution with the optimal solution for several randomly generated networks. A metric for user satisfaction, USat, has been defined in the paper. Experimental results show that 81% solutions obtained from heuristic have optimal USat, 95% solutions obtained from heuristic have optimal minimum allocated rate and the standard deviation of solutions are within 10% of optimal solutions.
{"title":"A unified approach for Multiple Multicast Tree Construction and Max-Min Fair rate allocation","authors":"R. Maniyar, P. Ghosh, Arunabha Sen","doi":"10.1109/HPSR.2009.5307420","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307420","url":null,"abstract":"Multicast communication is an efficient method of data transmission and distribution among a group, especially when network resources are inadequate and needs to be shared. Fair share of network resources, such as, bandwidth, is desirable in such cases. Although there has been an intensive research effort to design protocols and construct multicast routing graphs for a single multicast group, construction of multiple multicast groups and the fair allocation of network resources remains virtually unexplored. In this paper, a unified approach for the Multiple Multicast Tree Construction and Rate Allocation (MMTCRA) problem is addressed. The MMTCRA problem has been defined as an optimization problem with an objective of finding a Max-Min Fair rate allocation among the multiple multicast groups that co-exist in the network subject to the link-capacity constraints. The problem is proved to be NP-Complete. A Mixed Integer Linear Program (MILP) is formulated to achieve the optimal solution for this problem. A heuristic is proposed to solve the MMTCRA problem in polynomial time. The quality of the heuristic is evaluated by comparing the solution with the optimal solution for several randomly generated networks. A metric for user satisfaction, USat, has been defined in the paper. Experimental results show that 81% solutions obtained from heuristic have optimal USat, 95% solutions obtained from heuristic have optimal minimum allocated rate and the standard deviation of solutions are within 10% of optimal solutions.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121542370","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307429
M. Bando, N. S. Artan, H. J. Chao
Since the recent increase in the popularity of services that require high bandwidth, such as high-quality video and voice traffic, the need for 100-Gbps equipment has become a reality. In particular, next generation routers are needed to support 100-Gbps worst-case IP lookup throughput for large IPv4 and IPv6 routing tables, while keeping the cost and power consumption low. It is challenging for today's state-of-the-art IP lookup schemes to satisfy all of these requirements. In this paper, we propose FlashLook, a low-cost, high-speed route lookup architecture scalable to large routing tables. FlashLook allows the use of low-cost DRAMs, while achieving high throughput. Traditionally, DRAMs are not known for their high throughput due to their high latency. However, FlashLook architecture achieves high-throughput with DRAMs by using the DRAM bursts efficiently to hide DRAM latency. FlashLook has a data structure that can be evenly partitioned into DRAM banks, a novel hash method, HashTune to smooth the hash table distribution and a data compaction method called verify bit aggregation to reduce memory usage of the hash table. These features of the FlashLook results in better DRAM memory utilization and less number of DRAM accesses per lookup. FlashLook achieves 100-Gbps worst-case throughput while simultaneously supporting 2M prefixes for IPv4 and 256k prefixes for IPv6 using one FPGA and 9 DRAM chips. FlashLook provides fast real-time updates that can support updates according to real update statistics.
{"title":"FlashLook: 100-Gbps hash-tuned route lookup architecture","authors":"M. Bando, N. S. Artan, H. J. Chao","doi":"10.1109/HPSR.2009.5307429","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307429","url":null,"abstract":"Since the recent increase in the popularity of services that require high bandwidth, such as high-quality video and voice traffic, the need for 100-Gbps equipment has become a reality. In particular, next generation routers are needed to support 100-Gbps worst-case IP lookup throughput for large IPv4 and IPv6 routing tables, while keeping the cost and power consumption low. It is challenging for today's state-of-the-art IP lookup schemes to satisfy all of these requirements. In this paper, we propose FlashLook, a low-cost, high-speed route lookup architecture scalable to large routing tables. FlashLook allows the use of low-cost DRAMs, while achieving high throughput. Traditionally, DRAMs are not known for their high throughput due to their high latency. However, FlashLook architecture achieves high-throughput with DRAMs by using the DRAM bursts efficiently to hide DRAM latency. FlashLook has a data structure that can be evenly partitioned into DRAM banks, a novel hash method, HashTune to smooth the hash table distribution and a data compaction method called verify bit aggregation to reduce memory usage of the hash table. These features of the FlashLook results in better DRAM memory utilization and less number of DRAM accesses per lookup. FlashLook achieves 100-Gbps worst-case throughput while simultaneously supporting 2M prefixes for IPv4 and 256k prefixes for IPv6 using one FPGA and 9 DRAM chips. FlashLook provides fast real-time updates that can support updates according to real update statistics.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114666715","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307418
S. Ktari, Artur Hecker, H. Labiod
In peer to peer networks, the overlay topology connecting peers is an important component in addition to peer and data organization and search. In fact, topological characteristics have an important impact on the efficiency of search on such networks. In this paper, we address this issue for Chord based P2P systems. Based on the Barabasi Albert (BA) model, we propose to extend the topology of Chord to a scale free graph DHT, named PL-Chord, to provide a non hierarchical super-peer based lookup algorithm with an efficient search support. Through several simulation experiments, we found that PL-DHT could construct an overlay network that extends the search region and provides higher reachability at a very little additional cost.
在对等网络中，连接对等体的覆盖拓扑结构是对等体和数据组织与搜索的重要组成部分。事实上，拓扑特征对这类网络的搜索效率有着重要的影响。在本文中，我们针对基于Chord的P2P系统解决了这个问题。在Barabasi Albert (BA)模型的基础上，提出将Chord的拓扑结构扩展为一个无标度图DHT，命名为PL-Chord，以提供一种具有高效搜索支持的非分层超级对等查找算法。通过多次仿真实验，我们发现PL-DHT可以构建一个覆盖网络，以很小的额外成本扩展搜索区域并提供更高的可达性。
{"title":"Empowering Chord DHT overlays","authors":"S. Ktari, Artur Hecker, H. Labiod","doi":"10.1109/HPSR.2009.5307418","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307418","url":null,"abstract":"In peer to peer networks, the overlay topology connecting peers is an important component in addition to peer and data organization and search. In fact, topological characteristics have an important impact on the efficiency of search on such networks. In this paper, we address this issue for Chord based P2P systems. Based on the Barabasi Albert (BA) model, we propose to extend the topology of Chord to a scale free graph DHT, named PL-Chord, to provide a non hierarchical super-peer based lookup algorithm with an efficient search support. Through several simulation experiments, we found that PL-DHT could construct an overlay network that extends the search region and provides higher reachability at a very little additional cost.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114463749","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307430
W. Kabaciński, J. Kleban, M. Michalski, M. Żal, A. Pattavina, G. Maier
In this paper we consider the rearrangeable multi-plane banyan-type switching fabrics, called also log2(N, 0, p) switching networks, with even number of stages. For such networks different rearranging algorithms have been proposed for both: one-at-a-time and simultaneous connection models. In this paper we consider the one-at-a-time connection model, where connections arrive to the system one-by-one, and in case of blocking rearrangements are realized. To our knowledge, known algorithms require several rearrangements, and the number of such rearrangements have not been considered in the literature. We propose the new rearranging algorithm for the multi-plane banyan-type switching fabric composed of even number of stages. This algorithm leads to success using only one rearrangement. We also introduce the modified version of this new algorithm, in which rearrangement of an existing connecting path can be realized without its interruption.
{"title":"Rearranging algorithms for log2(N, 0, p) switching networks with even number of stages","authors":"W. Kabaciński, J. Kleban, M. Michalski, M. Żal, A. Pattavina, G. Maier","doi":"10.1109/HPSR.2009.5307430","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307430","url":null,"abstract":"In this paper we consider the rearrangeable multi-plane banyan-type switching fabrics, called also log2(N, 0, p) switching networks, with even number of stages. For such networks different rearranging algorithms have been proposed for both: one-at-a-time and simultaneous connection models. In this paper we consider the one-at-a-time connection model, where connections arrive to the system one-by-one, and in case of blocking rearrangements are realized. To our knowledge, known algorithms require several rearrangements, and the number of such rearrangements have not been considered in the literature. We propose the new rearranging algorithm for the multi-plane banyan-type switching fabric composed of even number of stages. This algorithm leads to success using only one rearrangement. We also introduce the modified version of this new algorithm, in which rearrangement of an existing connecting path can be realized without its interruption.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115213243","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 : 2009-06-22DOI: 10.1109/HPSR.2009.5307416
Yeim-Kuan Chang, Yung-Chieh Lin, Kuan-Ying Ho
In high performance routers design, fast IP address lookup is always a challenge. In order to obtain fast lookup speed, multi-bit tries are often used to represent the routing tables [1,2,3,6]. The drawbacks of multi-bit tries are the large memory usage and extensive update cost. To reduce the memory usage of multi-bit tries, Srinivasan and Varghese proposed a scheme called Controlled Prefix Expansion (CPE) [2] that uses the dynamic programming technique to obtain the optimal multi-bit tries in terms of memory usage. Furthermore, current backbone routers usually run the Border Gateway Protocol (BGP). BGP may cause a few hundred of updates per second. To make multi-bit tries adequate to these updates, a series of multi-bit tries nodes need to be modified. Since these updates can seriously affect the lookup speed, we need to minimize these update cost. However, CPE does not concern this issue. In this paper, we explore the optimization issue in terms of the update cost. We want to find an update-optimal multi-bit tries that still have the efficiency of lookup speed and memory usage. Contrast to CPE, our solutions achieve a 26% reduction of the update overhead and improve 38% of the search speed. Besides, we also examine our schemes in IPv6 routing tables. The experimental results show that our scheme can also scale well in IPv6.
{"title":"Update-aware Controlled Prefix Expansion for fast IP lookups","authors":"Yeim-Kuan Chang, Yung-Chieh Lin, Kuan-Ying Ho","doi":"10.1109/HPSR.2009.5307416","DOIUrl":"https://doi.org/10.1109/HPSR.2009.5307416","url":null,"abstract":"In high performance routers design, fast IP address lookup is always a challenge. In order to obtain fast lookup speed, multi-bit tries are often used to represent the routing tables [1,2,3,6]. The drawbacks of multi-bit tries are the large memory usage and extensive update cost. To reduce the memory usage of multi-bit tries, Srinivasan and Varghese proposed a scheme called Controlled Prefix Expansion (CPE) [2] that uses the dynamic programming technique to obtain the optimal multi-bit tries in terms of memory usage. Furthermore, current backbone routers usually run the Border Gateway Protocol (BGP). BGP may cause a few hundred of updates per second. To make multi-bit tries adequate to these updates, a series of multi-bit tries nodes need to be modified. Since these updates can seriously affect the lookup speed, we need to minimize these update cost. However, CPE does not concern this issue. In this paper, we explore the optimization issue in terms of the update cost. We want to find an update-optimal multi-bit tries that still have the efficiency of lookup speed and memory usage. Contrast to CPE, our solutions achieve a 26% reduction of the update overhead and improve 38% of the search speed. Besides, we also examine our schemes in IPv6 routing tables. The experimental results show that our scheme can also scale well in IPv6.","PeriodicalId":251545,"journal":{"name":"2009 International Conference on High Performance Switching and Routing","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126018803","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：