公开(公告)号：EP0814583A2

公开(公告)日：1997-12-29

申请号：EP97480020

申请日：1997-05-23

Applicant: IBM

Inventor： BERTIN OLIVIER ,  BRUN GERARD ,  GALAND CLAUDE ,  MAUREL OLIVIER ,  NICOLAS LAURENT

IPC: H04L12/70 ,  H04L12/931 ,  H04L12/935 ,  H04L12/937 ,  H04L12/947 ,  H04L12/56

CPC classification number: H04L49/30 ,  H04L49/205 ,  H04L49/254 ,  H04L49/256 ,  H04L49/3009 ,  H04L49/3081 ,  H04L49/50 ,  H04L2012/5619

Abstract: The present invention is directed to a high speed packet switching network and, in particular to a method and system for minimizing the time to establish a connection between an origin and a destination node. Due to high dynamicity of the traffic on transmission links, it is important to select a routing path according to a fully up-to-date information on all network resources. The simpler approach is to calculate a new path for each new connection request. This solution may be very time consuming because there are as many path selection operations as connection set up operations. On another hand, the calculation of paths based on an exhaustive exploration of the network topology, is a complex operation which may also take an inordinate amount of resources in large networks. Many of connections originated from a network node flow to the same destination network node. It is therefore possible to take a serious benefit in reusing the same already calculated paths for several connections towards the same node.The path calculated at the time the connection is requested is recorded in a Routing Database and updated each time a modification occurs in the network. Furthermore, alternate paths for supporting non-disruptive path switch on failure or preemption, and new paths towards potential destination nodes can be calculated and stored when the connection set up process is idle. These last operations are executed in background with a low processing priority and in absence of connection request.

Abstract translation: 本发明涉及一种高速分组交换网络，特别涉及用于最小化建立起始节点与目的地节点之间的连接的时间的方法和系统。 由于传输链路上的流量的高动态性，重要的是根据所有网络资源的完全最新信息来选择路由路径。 更简单的方法是为每个新的连接请求计算新的路径。 该解决方案可能非常耗时，因为存在与连接建立操作相同的路径选择操作。 另一方面，基于对网络拓扑的详尽探索的路径计算是复杂的操作，其在大型网络中也可能占用过多的资源。 许多来自网络节点的连接流到同一目标网络节点。 因此，可以重复使用相同的已计算出的路径到同一节点的多个连接的重要利益。在请求连接时计算出的路径被记录在路由数据库中，并且每次在网络中发生修改时更新 。 此外，当连接建立进程空闲时，可以计算并存储用于支持故障或抢占时的非中断路径切换的备用路径和朝向潜在目的地节点的新路径。 这些最后的操作在后台执行，处理优先级低，没有连接请求。

公开(公告)号：JPH1070571A

公开(公告)日：1998-03-10

申请号：JP16152797

申请日：1997-06-18

Applicant: IBM

Inventor： BERTIN OLIVIER ,  BRUN GERARD ,  GALAND CLAUDE ,  MAUREL OLIVIER ,  NICOLAS LAURENT

IPC: H04L12/56

Abstract: PROBLEM TO BE SOLVED: To minimize connection setup delay in an access node by selecting an advance calculation path for each connection request and calculating the path when the advance calculation path is not stored in a path designation database and including a process storing it to the path designation database. SOLUTION: A path controller 305 calculates an optimum path through a network to satisfy one set of given service quality designated by the user and to minimize the quantity of network resource used to make the communication path complete. Then the controller builds up a header for a packet generated at a path designating point. As an optimizing criterion, there are the number of intermediate nodes, the characteristic of a connection request, the link capability in the path, its operating state and number of intermediate nodes or the like. The optimum path is stored in a path designation database 308 for reuse.

公开(公告)号：CA2123449A1

公开(公告)日：1995-01-31

申请号：CA2123449

申请日：1994-05-12

Applicant: IBM

Inventor： ALFONSI JEAN-PIERRE ,  GALAND CLAUDE ,  LEBIZAY GERALD ,  MAUREL OLIVIER

IPC: H04L12/56

Abstract: Currently, the routing algorithms compute all the available paths in the network, from the source node to the destination node before to select the optimal route. No assumption is made on the network topology and the route computation is often time and resource consuming. Some paths which are calculated are not acceptable due to the particular geographical configuration of the network. In the real world, large transport networks are not fully meshed. The present invention is based on the observation that networks are usually built around a hierarchical structure. A set of nodes, interconnected by high throughput lines, are used to build a 'Backbone' (401) with a high degree of meshing to allow the redundancy and reliability required by the user. The other nodes or 'local' nodes (404) are attached to one or several backbone nodes. It is the network designer responsibility, at the configuration time to define for each node what is its attribution : backbone (402) or local node (404). The list of the node attributions appears in the topology table (306) and is updated each time a node is added to or dropped from the network. The routing algorithm can take advantage of the particular network topology to drastically reduce the complexity of paths computation. For a given connection, only a limited number of nodes are eligible and are taken in account by the algorithm in the optimal route search. The object of the invention is to split the network in backbone and local nodes to speed up the path selection.

公开(公告)号：CA2130394A1

公开(公告)日：1995-06-23

申请号：CA2130394

申请日：1994-08-18

Applicant: IBM

Inventor： BERTIN OLIVIER ,  MAUREL OLIVIER

IPC: H04L12/56

Abstract: The routing paths between origin and destination nodes in a packet switching network are selected more efficiently by reducing the total number of links potentially visited by the routing algorithm. The path selection technique involves two phases, the first phase for principal links identification, and the second phase for selection of the best point-to-point path in the network. Principal paths are minimum hop count paths with a transmission delay less than a specified threshold. The algorithm limits the screening of the network links in the second phase to the principal links and excludes the secondary links. The exploration of a much smaller number of links has the advantage to considerably reduce the computation time required for the path selection. The fact the number of principal links is generally small compared with the total number of links potentially explored, allows to limit the additional processing time necessary in case of unsuccessful attempt.