China 
Africa 
Explore chapters and articles related to this topic
Extending Open Shortest Path First for Mobile Ad Hoc Network Routing
Published in Jonathan Loo, Jaime Lloret Mauri, Jesús Hamilton Ortiz, Mobile Ad Hoc Networks, 2016
Katherine Isaacs, Julie Hsieh, Melody Moh
Mobile ad hoc networks (MANETs) face many challenges due to the diverse nature of their nodes and the fluidity of their topology. Routing is central to the efficiency and scalability of MANETs. Open Shortest Path First (OSPF) is the most widely used intradomain routing protocol on the Internet. Extending OSPF to handle the unique concerns of MANETs would be ideal as OSPF is well-known and well-tested. Furthermore, this extension would more easily support the seamless traveling of nodes between MANETs and wired networks. To understand MANET routing in OSPF, we first review many of the techniques used in MANET routing and discuss widely researched protocols including ad hoc on-demand distance vector (AODV), dynamic source routing (DSR), optimized link state routing (OLSR), zone routing protocol (ZRP), and fuzzy-sighted link state (FSLS). Next, we review the standard OSPF protocol and the existing MANET extensions for it. Then, we compare the OSPF extensions for MANET, both in design and by summarizing individual simulation results. Finally, we discuss future directions by pointing out several promising approaches as well as emerging concerns.
Layer-3 and Layer-2 MPLS VPNs
Published in Nam-Kee Tan, MPLS for Metropolitan Area Networks, 2004
Traditionally, an elaborate OSPF network consists of a backbone area (Area 0) and a number of areas connected to this backbone area via an area border router (ABR). With OSPF as the L3 MPLS VPN PE-CE routing protocol, a third level in the hierarchy of the OSPF model is introduced. This third level is called the MPLS VPN super backbone. In simple cases, the MPLS VPN super backbone is regarded as the traditional Area 0, implying that the MPLS VPN super backbone assumes the role of Area 0 when Area 0 is not configured on the customer network. On the other hand, the super backbone also allows customers to use Area 0 on their respective sites. Each site can have a separate Area 0 as long as it is connected to the super backbone. The result is the same as a partitioned Area 0.
OSPF and IS-IS Synchronization Mechanisms
Published in Rui Valadas, OSPF and IS-IS, 2019
The designated routers, called DR in OSPF and DIS in IS-IS, are both elected through the Hello protocol, but the election processes are slightly different. OSPF also elects a second router as part of this process, called BDR, with the role of replacing the DR in case of failure. In both protocols, the election process is influenced by a configurable priority called Router Priority in OSPF and Priority in IS-IS; a higher priority means that the router will have more chances of being elected, and a priority of zero means that the router will not participate in the election process.
Seismic risk analysis of a data communication network
Published in Sustainable and Resilient Infrastructure, 2022
Simona Esposito, Alessio Botta, Melania De Falco, Adriana Pacifico, Eugenio Chioccarelli, Antonio Pescapè, Antonio Santo, Iunio Iervolino
Routing is based on the combination of two protocols: open shortest path first (OSPF), used for the internal reachability of the backbone, and border gateway protocol (BGP), used for the propagation of the routes regarding external networks. BGP is configured on all the routers in the first-level POPs and the edge routers of the external networks. Routes in the first-level POPs are also set as route reflector and have a full mesh of internal BGP sessions between them. The functionality of the route reflector is to logically divide the backbone in a set of clusters, to reduce the need of meshing of the BGP sessions to a single element of each cluster. The interior routing protocol, OSPF, is used for determining the address of the next hop router, while BGP carries the routing information of local networks to the backbone.
Virtualised Environment for Learning SDN-based Networking
Published in IETE Journal of Education, 2020
Oscar Polanco, Fabio G. Guerrero
Many other network concepts such as, for instance, routing (BGP, MP-BGP, OSPF, IS-IS, EIGRP, etc.), multiple protocol label switching (MPLS) networks, and services can also be practised. We have also found the virtualisation approach to be useful for network security analysis, MPLS traffic engineering, and IPv6 based networks. Using the virtualised educational network environment, students can work autonomously to implement a campus network with layer 2 technologies and protocols (e.g. VLANs, IEEE 802.1q, STP, multiple STP, link aggregation control protocol) as well as layer 3 technologies and protocols (e.g. virtual router redundancy protocol, OSPF, IPv4, and IPv6). Students can also implement scenarios for management and automation (SNMP, Ansible) of a basic ISP infrastructure and, if necessary, connect it to the physical world, along with configuring MPLS services that ISPs usually provide to users (e.g. VPN-MPLS, central services, access to the Internet, etc.). The virtualization approach has several practical advantages. Owing to its associated costs, it is difficult to keep the network equipment of a computer network laboratory always updated. Internet service providers, for obvious reasons, hardly allow outsiders to get details of their network topologies, protocols, and settings. Carrying out an instructional activity involving sensitive aspects such as routing in an operational network is not realistic.
Heuristic traffic engineering for SDN
Published in Journal of Information and Telecommunication, 2020
Khoa Truong Dinh, Sławomir Kukliński, Tomasz Osiński, Jacek Wytrębowicz
It is also worth mentioning the problem induced by the static shortest path routing in case of a failure – the routers look for a new shortest path as a candidate to forward traffic, and when such path is found the total traffic from the broken path moves to a new path. Such behaviour may lead to the congestion of the newly selected one, whereas other existing paths between the source-destination pair of interest remain under-utilized. This example shows the benefits of traffic distribution among multiple paths. Moreover, the multipath based load balancing mechanism can contribute to increased network efficiency, in terms of the amount of traffic handled, by reducing congestion of links and therefore, packet loss rate. There are two prevailing engineering approaches, i.e. MPLS-TE (Awduche et al., 2001) and OSPF-ECMP (Németh et al., 2013), which were created to handle the mentioned above shortages. MPLS-TE extends MPLS networks using the RSVP protocol in order to establish paths by taking into consideration the network constraints such as available link bandwidth. MPLS-TE can be used for traffic engineering (TE), but it is seen as a too complicated and slow solution to make dynamic reconfiguration of paths. Moreover, it can introduce high overhead. The ECMP (Equal Cost Multipath) extension of the OSPF routing protocol also has some limitations. For instance, it cannot be used to split traffic between parallel paths of different capacities. Generally, ECMP does not support configurable load ratios among equal-cost paths, which can lead to congestion along the path that carries more traffic. In the case of many large flows (elephant flows), the ECMP forwards them via the same path, which results in load imbalance and waste of network bandwidth (Dixit et al., 2013).