Experimental validation of resilient tree-based greedy geometric routing

Geometric routing is an alternative for IP routing based on longest prefix matching. Using this routing paradigm, every node in the network is assigned a coordinate and packets are forwarded towards their intended destination following a distance-decreasing policy (greedy forwarding). This approach makes the routers significantly more memory-efficient compared to the current IP routers. In this routing, greedy embeddings are used to guarantee a 100% successful delivery to every destination in the network. Most of the existing proposals lack resiliency mechanisms to react efficiently to network changes. We propose a distributed algorithm to calculate a greedy embedding based on a spanning tree of the network. In this algorithm, nodes are triggered to re-calculate their coordinates upon a change in the topology such as link or node failures. The advantage of this approach is that it recovers from topology failures within a very short period of time. We further extend the algorithm to generate backups to apply protection in distributed setups. Different trade-offs and trends of re-convergence for geometric routing have been evaluated in an emulation environment. Realistic results are achieved through emulation as no model or abstraction is involved. The proposed routing scheme is implemented in Quagga routing software and new elements are developed in Click modular router to enable greedy forwarding. For the first time, the performance of this scheme is evaluated through emulation on a large topology of 1000 nodes and the results are compared with BGP. The experimental results indicate that the proposed scheme has interesting characteristics in terms of convergence time upon a change in the network topology.     

Learning the valid incoming direction of IP packets

Packet forwarding on the Internet is solely based on the destination address of packets, and it is easy to forge the source address of IP packets without affecting the delivery of the packets. To solve this problem, one can have routers check whether or not every packet comes from a correct direction based on its source address field. However, due to routing asymmetry in today’s Internet, a router cannot simply reverse its forwarding table to determine the correct incoming direction of a packet.In this paper, we present the source address validity enforcement protocol, SAVE, which allows routers to learn valid incoming directions for any given source address. SAVE is independent from—and can work with—any specific routing protocol. By only interfacing with the forwarding table at routers, SAVE allows routers to properly propagate valid source address information from source address spaces to all destinations, and allows each router en route to build and maintain an incoming tree to associate each source address prefix with a corresponding incoming interface. The incoming tree is further valuable in handling routing changes: although a routing change at one router could affect the incoming direction of source address spaces from many locations, only the router that sees the change needs to send out new updates. Finally, SAVE has a good performance with low overhead.     

Mitigating transient loops through interface-specific forwarding

Under link-state routing protocols such as OSPF and IS–IS, when there is a change in the topology, propagation of link-state advertisements, path recomputation, and updating of forwarding tables (FIBs) will all incur some delay before traffic forwarding can resume on alternate paths. During this convergence period, routers may have inconsistent views of the network, resulting in transient forwarding loops. Previous remedies proposed to address this issue enforce a certain order among the nodes in which they update their FIBs. While such approaches succeed in avoiding transient loops, they incur additional message overhead and/or increased convergence delay. We explore an alternative approach, loopless interface-specific forwarding (LISF), that mitigates transient loops by forwarding a packet based on both its incoming interface and destination address. LISF needs to compute and update interface-specific instead of interface-independent forwarding tables. But it requires neither the synchronization of FIB updates at different nodes nor the modification of the existing link-state routing mechanisms. LISF is easily deployable with current routers if they already maintain a FIB at each interface for lookup efficiency. This paper presents the LISF approach, illustrates its strengths and limitations, discusses four alternative implementations of it and evaluates their performance.     

Dynamic filter merging and mergeability detection for publish/subscribe

Filter-based publish/subscribe and content-based routing have been proposed for flexible information dissemination in distributed environments. A content-based router is part of an overlay structure, in which each router forwards events to neighbouring routers and local clients based on their interests. Filter merging or summarization has been proposed as an optimization strategy in this environment. These techniques combine filters to reduce the number of propagated filters and thus the size of distributed state. In this paper, we present the algorithms for generic dynamic filter merging and filter mergeability detection, and discuss integration with routing tables. The algorithms are based on a formal framework of merging rules. Experimental results are examined and analyzed for both desktop systems and small devices. The results indicate that dynamic filter merging is feasible given that the workload is mergeable.     

传感器网络中基于轨迹的多路径容错路由算法

基于轨迹提出一种新的多路径构造算法,其基本思想是源节点根据目的节点位置信息选择几条合适的转发路径,中间节点根据不同的贪婪转发策略建立相应的动态转发表转发报文。和其他算法相比,新算法具有分布式特性,路径建立只需局部信息即可;算法实现简单,无须大量的计算。理论分析和模拟实验也表明,新算法具有较好的容错性。     

Compressing Two-Dimensional Routing Tables

We consider an algorithmic problem that arises in the context of routing tables used by Internet routers. The Internet addressing scheme is hierarchical, where a group of hosts are identified by a prefix that is common to all the hosts in that group. Each host machine has a unique 32-bit address. Thus, all traffic between a source group s and a destination group d can be routed along a particular route c by maintaining a routing entry (s, d, c) at all intermediate routers, where s and d are binary bit strings. Many different routing tables can achieve the same routing behavior. In this paper we show how to compute the most compact routing table. In particular, we consider the following optimization problem: given a routing table D with N entries of the form (s, d, c) , determine a conflict-free routing table with fewest entries that has the same routing behavior as D. If the source and destination fields have up to w bits, and there are at most K different route values, then our algorithm runs in worst-case time O( NK w2) .     

XOR-Based Schemes for Fast Parallel IP Lookups

An IP router must forward packets at gigabit speed in order to guarantee a good quality of service. Two important factors make this task a challenging problem: (i) for each packet, the longest matching prefix in the forwarding table must be quickly computed; (ii) the routing tables contain several thousands of entries and their size grows significantly every year. Because of this, parallel routers have been developed which use several processors to forward packets. In this work we present a novel algorithmic technique which, for the first time, exploits the parallelism of the router also to reduce the size of the routing table. Our method is scalable and requires only minimal additional hardware. Indeed, we prove that any IP routing table T can be split into two subtables T₁ and T₂ such that: (a) |T₁| can be any positive integer k ≤ |T| and |T₂| ≤ |T| - k + 1; (b) the two routing tables can be used separately by two processors so that the IP lookup on T is obtained by simply XOR-ing the IP lookup on the two tables. Our method is independent of the data structure used to implement the lookup search and it allows for a better use of the processors L2 cache. For real routers routing tables, we also show how to achieve simultaneously: (a) |T₁| is roughly 7% of the original table T; (b) the lookup on table T₂ does not require the bestmatching prefix computation.     

Quality-of-service routing in IP networks

Multimedia applications such as video-conferencing, telemedicine, HDTV, etc. have very stringent quality-of-service (QoS) demands and require a connection-oriented service. For these applications, a path satisfying their requirements in terms of bandwidth, delay buffer, etc. needs to be found. As conventional IP routing is based only on hop counts, it is not suitable for multimedia applications. It is clear that, to route requests that have QoS requirements, existing routers should be made QoS aware and the packet forwarding should be based on QoS parameters. Also, routing protocols like OSPF and RIP must be extended suitably to facilitate QoS routing. The goal of QoS routing algorithms is to find a loop-less path satisfying a given set of constraints on parameters like bandwidth, delay, etc. The path selection process could return either the entire path to the destination or the best next hop for the request. The first case is called “source routing” and the second is referred to as “distributed routing”. In this paper, we propose a new distributed QoS routing algorithm for unicast flows, which has a very low call establishment overhead. Our algorithm makes use of existing IP routing protocols such as OSPF and RIP with minimal modifications     

基于TCAM的二级转发表管理方案

目前高性能路由器的查找方案采用基于三重内容可寻址存储器(TCAM)的硬件查找。路由查找的最长前缀匹配要求使得TCAM的表项更新复杂。该文针对当前转发表管理方案在带突发情况下的不足,提出一种基于TCAM二级转发表管理方案,并对该方案进行了计算机仿真研究,仿真结果表明该方案在突发的情况下仍能保证转发系统稳定工作。     

Efficient Management and Packets Forwarding for Multimedia Flows

Well known requirements for handling multimedia flows in routers are resource reservation and fast packets forwarding. The former takes into account the quite stable and long lasting bandwidth occupation, whereas the latter takes into account the large number of packets routed along the same path. Many techniques have been proposed and standardized to face these requirements, but their application is often complex, expensive and sometimes limited by the need of agreements between network managers of the many networks and Autonomous Systems. This paper introduces IMFM (Integrated Multimedia Flows Management), an innovative, scalable, and extremely lightweight technique to provide routers deterministic and dynamic resource reservation and a fast forwarding table lookup. It is based on a distributed linked data structure that allows direct (searchless) access to entries in the routers’ tables, extending the resource reservation algorithm REBOOK. Unlike conventional virtual circuits, IMFM does not require any interaction with (nor change in) the underlying routing protocols and autonomously recovers from errors, faults and route changes. If information stored in its data structure becomes obsolete, packet handling is reverted to best-effort, lookup-driven forwarding, so that packets are never dropped nor misrouted. IMFM can be gradually deployed, providing a framework for congestion avoidance solutions and increasing the forwarding speed in IMFM-aware router even along partially IMFM-unaware paths. IMFM has been fully implemented. Experiments have been designed to demonstrate its feasibility and the measured performance is reported and compared with existing techniques.     

RIDSI:一种融合SDN和兴趣域划分的ICN路由机制

信息中心网络(Information-Centric Networking,ICN)中的服务质量(Quality of Service,QoS)路由是分布式的,无法感知其它路由器缓存的内容,因此,基于软件定义网络(Software Defined Networking,SDN)的全局视图和集中控制功能,本文提出了一种基于兴趣域划分的ICN路由机制.这种机制旨在减轻内容节点的负载、提高路由成功率和提高PIT命中率.首先,本文提出了基于SDN的ICN网络模型,即软件定义信息中心型网络(Software Defined Information Centric Networking,SD-ICN)模型和QoS评价模型.然后,采用蜂群算法将网络中的路由器划分到不同的兴趣域.最后,基于改进的QoS依赖多播路由(QoS Dependent M ulticast Routing,QDM R)算法计算满足多个兴趣请求的转发路径.仿真实验结果表明,本文提出的路由机制与对比机制相比在路由成功率、平均路由延迟、负载均衡度等方面都具有较好的性能.     

Performance Analysis and Route Optimization: Redistribution between EIGRP,OSPF & BGP Routing Protocols

Routing is the process of data path selection of IP networks. Routers perform path selection on the basis of routing tables stored in their memory. Routing table contains IP routes for route transformation via the best path in the networks. Service providers use different routing protocols in their enterprise networks. These routing-protocols have the limitation of non-convergence in the networks. Route redistribution is the technique which overcomes this limitation. Due to this technique, service providers can get optimized communication with IP networks where multiple routing protocols are being used. This research article focuses on the performance and redistribution of different routing protocols in medium or enterprise IP networks. A simulated network model is established in GNS3 simulator. Five Cisco-7200 series routers and a switch is used in this simulated topology. All these routers are directly connected with each other via serial links. Routing protocols EIGRP, OSPF and BGP are used in this topology and then configured route redistribution on these routers. Different types of data traffic are generated and passed through the network in order to analyze network convergence, throughput and packet delay by the use of software wire shark network analyzer and debug command. EIGRP is better in convergence and through put whereas OSPF is better in packet delay.     

Routing table partitioning for speedy packet lookups in scalable routers

Most of the high-performance routers available commercially these days equip each of their line cards (LCs) with a forwarding engine (FE) to perform table lookups locally. This work introduces and evaluates a technique for speedy packet lookups, called SPAL, in such routers. The BGP routing table under SPAL is fragmented into subsets which constitute forwarding tables for different FEs so that the number of table entries in each FE drops as the router grows. This reduction in the forwarding table size drastically lowers the amount of SRAM (e.g., L3 data cache) required in each LC to hold the trie constructed according to the prefix matching algorithm. SPAL calls for caching the lookup result of a given IP address at its home LC (denoted by LC/sub ho/, using the LR-cache), such that the result can satisfy the lookup requests for the same address from not only LC/sub ho/, but also other LCs quickly. Our trace-driven simulation reveals that SPAL leads to improved mean lookup performance by a factor of at least 2.5 (or 4.3) for a router with three (or 16) LCs, if the LR-cache contains 4K blocks. SPAL achieves this significant improvement, while greatly lowering the SRAM (i.e., the L3 data cache plus the LR-cache combined) requirement in each LC and possibly shortening the worst-case lookup time (thanks to fewer memory accesses during longest-prefix matching search) when compared with a current router without partitioning the routing table. It promises good scalability (with respect to routing table growth) and exhibits a small mean lookup time per packet. With its ability to speed up packet lookup performance while lowering overall SRAM substantially, SPAL is ideally applicable to the new generation of scalable high-performance routers.     

Distributed load balancing mechanism for detouring schemes of geographic routing in wireless sensor networks

Well-known ‘routing hole’ problem of geographic routing is hardly avoided in wireless sensor networks because of various actual geographical environments. Existing geographic routing protocols use perimeter routing strategies to find a detour path around the boundary of holes when they encounter the local minimum during greedy forwarding. However, this solution may lead to uneven energy consumption around the holes since it consumes more energy of the boundary sensors. It becomes more serious when holes appear in most of routing paths in a large-scale sensor network. In this paper, we propose a novel distributed strategy to balance the traffic load on the boundary of holes by virtually changing the sizes of these holes. The proposed mechanism dynamically controls holes to expand and shrink circularly without changing the underlying forwarding strategy. Therefore, it can be applied to most of the existing geographic routing protocols which detour around holes. Simulation results show that our strategy can effectively balance the load around holes, thus prolonging the network life of sensor networks when an existing geographic routing protocol is used as the underlying routing protocol.     

Optimal State Allocation for Multicast Communications With Explicit Multicast Forwarding

In this paper, we propose a scalable and adaptive multicast forwarding mechanism based on explicit multicast (Xcast). This mechanism optimizes the allocation of forwarding states in routers and can be used to improve the scalability of traditional IP multicast and source-specific multicast. Compared with previous work, our mechanism needs fewer routers in a multicast tree to store forwarding states and therefore leads to a more balanced distribution of forwarding states among routers. We focus on two problems and formulate each of them as an optimization problem. The first problem, referred to as minstate, minimizes the total number of routers that store forwarding states in a multicast tree. The second problem, referred to as balancestate, minimizes the maximum number of forwarding states stored in a router for all multicast groups, which is proved to be an NP-hard problem. We design a distributed algorithm that obtains the optimal solution to the first problem and propose an approximation algorithm for the second problem. We also prove that the approach adopted by most existing works to allocate forwarding states in the branching routers of a multicast tree is a special case of our mechanism. The simulation results show that the forwarding state allocation provided by previous work is concentrated on the backbone routers in the Internet, which may cause the scalability problem. In contrast, our mechanism can balance forwarding states stored among routers and reduce the number of routers that store the forwarding states for a multicast tree.     

Compressing Two-Dimensional Routing Tables

We consider an algorithmic problem that arises in the context of routing tables used by Internet routers. The Internet addressing scheme is hierarchical, where a group of hosts are identified by a prefix that is common to all the hosts in that group. Each host machine has a unique 32-bit address. Thus, all traffic between a source group s and a destination group d can be routed along a particular route c by maintaining a routing entry (s, d, c) at all intermediate routers, where s and d are binary bit strings. Many different routing tables can achieve the same routing behavior. In this paper we show how to compute the most compact routing table. In particular, we consider the following optimization problem: given a routing table D with N entries of the form (s, d, c) , determine a conflict-free routing table with fewest entries that has the same routing behavior as D. If the source and destination fields have up to w bits, and there are at most K different route values, then our algorithm runs in worst-case time O( NK w ² ) .     

主动层次多目通信路由模型

许多应用需要IP多目通信.在Internet大规模应用IP Multicast时,有效的路由是关键.这样的多目路由协议必须是有效的、可伸缩的和增量可配置的.但是传统的Internet路由对性能是不敏感的,不能平衡负载和处理拥塞.现有的大多数多目通信路由协议不仅负责数据转发,还负责路由树的构造,这给路由器带来了极大的复杂性,而且协议的配置是手动的、费时费钱的工作.该文提出一个主动层次式Multicast路由的体系结构,采用主动网络技术将多目通信路由协议的数据转发和控制机制分开,根据链路的状态信息用主动报文控     

数据中心网络中一种半集中式SDN路由策略

在SDN体系架构中把网络控制功能从网络设备（交换机/路由器）里分离出来,集中到中心节点控制器上,交换机只负责数据平面的功能（通过流表进行数据转发）。在大规模的数据中心网络中,路由/流表的计算和分发完全由中心控制器完成,控制器成为网络的性能瓶颈和脆弱点。为了解决上述问题,本文提出半集中式SDN路由技术,其主要思想是每个交换机节点不需要控制器的参与,可以自主构建一个基础流表,基于基础流表,交换机可以完成基本的数据转发工作。而控制器负责更高级的路由选路（故障处理）工作,从而大大减轻控制器的负担。针对控制器的高级路由选路工作,本文通过对现有SDN网络中的故障恢复机制的特性以及限制的分析,在基础流表的基础上设计一套局部迂回故障检测恢复机制。基于该机制,控制器能够及时检测到网络故障,并在极短的时间内进行故障恢复,实现控制器的高级路由选路工作。     

快速更新IPv4/IPv6双栈路由查找算法与实现

IPv4/IPv6双栈核心路由器需要一体化高效路由查找．但常用的单步TCAM路由查找方案要求转发表的存储必须按前缀长度相对地址降序排列，这种与地址关联的排序操作影响表项的更新速度和路由查找流程的连续性．提出并实现了一种独特的对前缀范围对分搜索的IPv4／IPv6双栈一体化多步TCAM流水查找方法．突出特点是转发表不需排序、查找速率高、表项更新快、查表连续性好．可满足IPv4／IPv6双协议栈核心路由器OC-768（40Gbps）端口、48B包的线速转发．     

A simple distributed loop-free routing strategy for computercommunication networks

The loops resulting from either component failures or load changes in a computer communication network degrade the performance and the adaptability of conventional distributed adaptive routing strategies, such as ARPANET's previous routing strategy (APRS). The authors develop distributed loop-free routing strategy by adding only one additional piece of information-the total number of minimum-delay paths-to the commonly used routing messages and tables. The proposed routing strategy requires only easily obtainable information, yet removes loops completely. It is far more efficient in both time and space than its conventional counterparts, especially for sparse computer networks. The authors prove the correctness of the proposed strategy, and give several illustrative examples. The performance of this strategy is shown to be better than, or at least as good as, that of APRS and any multiorder routing strategies, in which the order of a routing strategy is determined by the amount of routing information carried in each routing message