Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing – Routing data updating
Reexamination Certificate
2000-09-14
2004-11-23
Thompson, Marc D. (Department: 2144)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
Routing data updating
C370S235000
Reexamination Certificate
active
06823395
ABSTRACT:
BACKGROUND
The present invention relates to an arrangement for managing/monitoring routing in a communications network, comprising a routing domain within a number of routing areas containing network nodes communicating via transmission links, which implements link state routing. The invention also relates to a routing controlling device for controlling routing in a routing area in a network implementing link state routing at least for that part (area) of the network. The invention also relates to a method of controlling routing within a routing area implementing link state routing. By controlling is here meant managing and/or monitoring.
Through routing a path through a network is determined. This can be done in different manners. Generally the monitoring and the management of the routing process is performed through direct communication with all network nodes which are associated with the routing process. According to one known solution this is done through actual login on each network element and retrieval of information. This is disadvantageous since each network node has to be contacted which means that network resources will be required which means that bandwidth as well as CPU (Central Processing Unit) power is demanded within the network nodes which is disadvantageous since preferably bandwidth as well as CPU power are scarce resources which should be used, to an extent which is as high as possible, for actual traffic handling. Another disadvantage consists in that the actual login on each network node generally is slow and different procedures are needed for different kinds of network nodes, for example different procedures are required for different manufactures of network nodes.
Alternatively remote monitoring may be implemented. Then may for example the SNMP (Simple Network Management Protocol) be used.
This is for example described in RFC 1157. This is however slow due to the fact that network nodes treat SNMP messages with the low priority as compared to other messages relating to other functions. A particular issue relates to Internet routing. Routing is a process used by Internet hosts for delivery of packets. Internet uses a hop-by-hop routing model which means that each host or router handling a packet examines the destination address in an IP header, computes the next hop address which will bring the packet one step closer to its destination and delivers the packet to the next hop address. There the above mentioned process is performed over again. A routing table is then required for matching destination addresses with next hops and routing protocols determine the contents of the tables. One particular kind of a routing protocol is a link-state routing protocol. Such a protocol requires each router to maintain at least a partial map of the network. When a link in the network changes its state (up to down or vice versa), a notification called a Link State Advertisement (LSA) is flooded throughout the network. This means that all the network nodes, in the case of Internet routers, note the change and in agreement therewith compute the routes. Examples on Internet links state routing protocols are OSPF (Open Shortest Path First) and IS—IS (OSI, Open Systems Interconnection). OSPF version 2 is for example described in RFC 1583. OSPF uses small so called hello packets for verification of link operations without transferring large tables. According to OSPF the routing domain is divided into different routing areas comprising a backbone area which divides interior routing into two levels and if inter area traffic is required, the packets are first routed to the backbone which is such may cause non-optimal routes since the inter area routing is not carried out until the packet reaches the backbone.
Link state routing of Internet means that each router has to find out which its neighbours are and obtain information about their addresses, measure the delay or the cost to each of said neighbours, construction of a packet with the above mentioned information which packet is to be sent to all the other routers and computation of the shortest path to all the other routers.
In addition to the hop-by-hop model another model is known which is denoted “source based routing”. The difference is that, the route that a packet should take through a network is determined at the network ingress point and not like in the hop-by-hop model where it is determined at every node that the packet traverses. This model is used e.g. in ATM and PNNI.
U.S. Pat. No. 5,687,168 for example describes link-state routing in an ATM communication system. A link-state routing device is here used to reduce the amount of link state information that is exchanged in the network irrespectively of the number of links connecting adjacent switches. An abstracted link a plurality of links are virtually aggregated to generate a link state with respect to the abstracted link uniting link state information of a plurality of links. The link state update protocol portion has link topology the information synchronized on the network through distribution of the abstracted link-state information with the flooding mechanism. Because of the fact that the abstracted link represents a plurality of links, the amount of information to be distributed can be reduced. However, this document discloses no solution to the above mentioned problems.
SUMMARY
What is needed it is therefore an arrangement as referred to above for managing/monitoring routing in a communications network comprising a number of routing domains in turn comprising a number of routing areas with a number of network nodes which communicate via transmission links. Each routing domain is administrated as one unit by administrating means and link state routing is implemented through which the management, which may comprise monitoring and/or managing of new situations on transmission links within the network, in an efficient, reliable and cheap manner, can be provided. An arrangement is also needed through which routing monitoring/management can be handled in such a manner that bandwidth can be saved both generally in the network and for the respective network nodes and through which CPU power can be saved for the respective network nodes. An arrangement is also needed through which routing managing and monitoring can be handled in a fast manner as well as independently of kind of network node, manufacturer of network node etc.
Therefore is also a routing controlling device for managing/monitoring routing within a routing area needed through which the above mentioned objects can be fulfilled.
Still further a method of managing/monitoring routing within a routing domain or in a routing area as referred to above is needed through which the above mentioned objects can be met.
Therefore an arrangement as referred to above is disclosed and there is one link state database for each routing area within the routing domain. A domain may of course comprise only one routing are as well. A link database is maintained by each network node of the routing area that each network node belongs to since each network node belongs to at least one routing area. For at least some of the routing areas a separate routing controlling device is provided such that a domain with a plurality of routing areas may comprise also a plurality of routing devices; and there may be one routing controlling device for each area but it is not necessarily the case. The numbers may also differ. Such routing controlling device belongs to the routing process of the respective routing area and it contains a copy of, preferably an identical copy of, the link state database of the routing area or areas and it is/are identical to the link state database of the network nodes of the respective area. Each routing controlling device is connected to a network node of the respective routing area it belongs to and the routing controlling device comprises means for rejecting non-routing information/traffic and for injecting routing information into the link state routing process respective routing area. More generally m
Telefonaktiebolaget LM Ericsson (publ)
Thompson Marc D.
LandOfFree
Arrangement and method relating to routing in a network does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Arrangement and method relating to routing in a network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Arrangement and method relating to routing in a network will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3302138