Routing method for module-expansion-based data center network topology system

Abstract

The invention discloses a routing method for a module-expansion-based data center network topology system, and mainly aims to solve the problem that the conventional routing methods in the Ethernet and a data center tree network cannot be well applied to the module-expansion-based data center network topology system. The method is implemented by the following steps of: (1) performing initialization, and identifying each piece of network equipment in the topology system by adopting a hierarchical addressing mechanism; (2) reading the addressing information of current network equipment, and judging the type of the current network equipment; (3) generating data, and transmitting the data to edge layer switches by using a source server; (4) determining data output ports according to the destination address of the data and a port with the data by using the edge layer switches, convergence layer switches and intermediate layer switches; and (5), determining a data output port according to the destination address of the data, the hop number of a path hop number and the cache information of the port by using core layer switches. The method has the advantages that: network throughput is improved; traffic balancing is realized; the method can be used for the path selection of a data center network and the provision of high-efficiency data transmission service; and the shortcomings of static routing are overcome.

Description

technical field [0001] The invention belongs to the technical field of the Internet, and in particular relates to a routing method of a data center network topology system based on module expansion, which can be used to realize efficient communication between data center network servers. Background technique [0002] With the rapid development of network and information technology, the data center has become the service center of scientific research institutes, financial institutions and other industries in modern society, carrying the core business of enterprises and meeting the service requirements of data mining, storage, high-performance computing and so on. Therefore, the construction of a data center suitable for the development needs of enterprises has become a key IT construction project for service operators, scientific research institutes, and various portal websites to achieve the purpose of improving service efficiency, reducing operating costs, and improving cent...

AI Technical Summary

Problems solved by technology

The working mechanism of open shortest path first OSPF is not suitable for large-scale data center networks, because collecting network status information has the following defects for data center networks: waste of storage space, collection of status information is not real-time, and routing decisions are relatively lagging; RIP protocol When working, the current device needs to exchange routing table information with its adjacent devices. Because the data center network is large in scale, the routing table entries of each device are huge, which wastes storage space and exchanges routing table information in real time, resulting in high system overhead. , and the update of routing information is relatively lagging behind, which is not conducive to routing decision-making

[0009] The above two routing methods all have the following deficiencies: First, the ECMP and VLB routing methods can only make full use of the equal-cost paths in the network topology, but cannot take advantage of the non-equivalent paths in the topology. Here, the equal-cost paths refer to data transmission Paths passing through the same number of switches; second, the ECMP and VLB routing methods fail to take into account the real-time status information of the network when making routing decisions. The ECMP routing method specifies the same transmission path for data flows with the same source and destination. The VLB routing method is to randomly specify the path for the data flow with equal probability, which cannot fully avoid the occurrence of network hotspots, and thus cannot obtain routing results with good performance.

These deficiencies lead to the fact that the above two routing methods cannot be well adapted to the data center network topology system based on module expansion, because in the data center network topology system based on module expansion, there are multiple parallel equal-cost paths and non-parallel paths between server pairs. equal-cost path, and the downlink can be dynamically selected, but the above two routing methods cannot fully explore the structural characteristics of the network topology to achieve optimal transmission

Images