Generalized stochastic high-level Petri net (GSHLPN)-based network data transmission modeling and performance analysis method

Abstract

The invention discloses a generalized stochastic high-level Petri net (GSHLPN)-based network data transmission modeling and performance analysis method, and belongs to the fields of network analysis and network management control in the Internet. A modeling and analysis process comprises the four steps of: acquiring the status of a target network; constructing a GSHLPN model of the target network; solving the constructed GSHLPN model; and calculating performance parameters of the target network according to a GSHLPN model solving result. The method also comprises the following steps of: mapping a network terminal node and data packet production consumption into a GSHLPN substructure; mapping the waiting queue length, routing capability and forwarding capability of a network switching node into a GSHLPN substructure; calculating the utilization rate, packet loss rate, throughput and delay of the network switching node according to the position and transition status of the solved GSHLPN; and calculating the delay characteristics, packet loss rate and network availability of service provided by the network for a specified load flow according to performance indexes of the network switching node and network topology.

Description

technical field [0001] The invention belongs to the field of network analysis and network management control in the Internet, in particular to a GSHLPN-based network data transmission modeling and performance analysis method. Background technique [0002] The application of computer network has spread to all fields of society and has become the basic requirement of national development and social progress. In the process of computer network design, deployment and application, it is necessary to analyze the performance of network data transmission according to the topology structure, equipment capability and operating load of the target network, and optimize the system design according to the analysis results. [0003] In the process of performance analysis of existing network systems, there are mainly methods of estimation based on experience, methods of building prototype system experiments, and methods of numerical simulation experiments. Design based on experience has lo...

AI Technical Summary

Problems solved by technology

Design based on experience has low accuracy and poor reliability; building a prototype system consumes a lot of money and takes a long time to prepare; the existing numerical simulation system is either expensive (such as the network simulation technology software package OPNET), or very difficult to use (such as the network simulation platform NS2 )

Images