A Scheduling Method Based on Service Duration in Hybrid Switching Network

Abstract

A scheduling method based on service duration in a hybrid switching network, the steps are as follows: 1) Obtain a network physical topology G that requires path resource allocation, and the hybrid network corresponding to the network physical topology G includes N nodes; 2) Collect network The resource state information of the hybrid network corresponding to the physical topology G constructs a heterogeneous link state table; 3) constructs a hierarchical graph model according to the different characteristics of the heterogeneous links; The services are numbered and classified according to the service characteristics of each service in the service request matrix D; 5) According to different types of services, the scope of path query is narrowed down, and resource scheduling based on service duration is performed.

Description

technical field [0001] The invention relates to a scheduling method based on service duration in a hybrid switching network, which belongs to data communication transmission technology. Background technique [0002] Currently, commonly used network scheduling methods include scheduling algorithms for assigning fixed time slots and scheduling algorithms based on priority queues. The scheduling algorithm for allocating time slots allocates fixed time slots for each service by establishing a time slot table, which has problems such as complex synchronization control of the entire network and weak adaptability to burst services; the scheduling algorithm based on priority queues is based on a certain According to the priority rules, different input service flows are temporarily stored in the corresponding queues for scheduling and service, which can quickly solve the scheduling problem of electrical services, but it is difficult to apply to optical services with difficult bufferi...

AI Technical Summary

Problems solved by technology

The scheduling algorithm for allocating time slots allocates fixed time slots for each service by establishing a time slot table, which has problems such as complex synchronization control of the entire network and weak adaptability to burst services; the scheduling algorithm based on priority queues is based on a certain According to the priority rules, different input service flows are temporarily stored in the corresponding queues for scheduling and service, which can quickly solve the scheduling problem of electrical services, but it is difficult to apply to optical services with difficult buffering and large granularity

However, the existing hybrid switching network resource scheduling method adopts a cascaded / separated scheduling method and a joint scheduling method. The cascaded / separated resource scheduling method adopts different scheduling methods for multi-layer and multi-domain link resources, and does not separate the resources between layers. The grooming link resources between domains are considered uniformly, and the link resources between layers and domains lack necessary coordination

The core idea of ​​the joint scheduling method is that each layer and each domain in the switching network is not designed independently. By comprehensively considering the resource parameters between adjacent layers or adjacent domains, each link resource is integrated through a cross-layer and cross-domain decision-making module. Resource allocation, through the coordination and fusion of resource parameters scattered in each layer, but the variety of link resources makes this method complex in technology and complex in resource scheduling methods.

Images