Automatic traffic and quality of service control system for communications networks

An intermediate device for monitoring and controlling quality of service and availability of a discrete event system composed of a computer communications network having intermediate nodes, said intermediate device comprising memory for storing a traffic intensity surface therein, said intermediate device adapted to monitor traffic intensity by modeling the traffic intensity as a partial differential equation, said intermediate device sampling queue state information in intermediate nodes in said network and from this queue state information estimating the parameters of the partial differential equation model and solving the partial differential equation to produce the traffic intensity surface;

• said intermediate device adapted to control traffic intensity by using several actuation mechanisms that include creating routes that minimize transit through congested regions of said network and downloading said routes to intermediate nodes for use in updating their forwarding tables, that include determining optimal times and locations for content caches to operate in said network and downloading said times and locations to cache content managers in said network, that include determining whether to correct bandwidth imbalances in said network either by buying and selling of short term bandwidth or by actuation of said network's topology and resources, including links and intermediate node capacities, and downloading said bandwidth actuations to bandwidth managers in the computer communications network.

A method for monitoring and controlling the quality of service and availability variables for a discrete event system composed of a computer communications network and its traffic load, said method comprising the steps of:

• (a) collecting queue state information from intermediate nodes in the computer communications network that is being managed and storing these in memory of an intermediate device;
• (b) estimating the parameters of the partial differential equation that describes the traffic intensity dynamic of the discrete event system and storing these in memory of an intermediate device;
• (c) solving the partial differential equation that describes the traffic intensity dynamic of the computer communication network to yield the traffic intensity surface for the discrete event system and storing this in memory of an intermediate device;
• (d) generating optimal and suboptimal paths that are geodesics and non-geodesics, respectively, of the traffic intensity surface and storing these in memory of an intermediate device;
• (e) generating optimal and suboptimal routes by projecting the optimal and suboptimal paths on to the topology of the computer communications network and storing these in memory of an intermediate device;
• (f) actuating the traffic intensity by downloading the optimal routes to intermediate nodes for the purpose of updating their forwarding tables;
• and/or
• (g) determining the optimal times and locations for content caches to operate and storing these in memory of an intermediate device
• (h) actuating the traffic intensity by downloading the optimal times and locations for content caches to operate to cache content managers in the computer communications network;
• and
• (i) determining the times and locations of bandwidth imbalances in the computer communications network and their persistence and storing these in memory of an intermediate device;
• and
• (j) actuating the traffic intensity by downloading the time and locations of the bandwidth imbalances to a bandwidth manager in the computer communication network.

In a computer system having a processor coupled to a bus, a computer system coupled to said bus and having stored therein a computer program that when executed by said processor causes said computer system to implement a method for managing quality of service and availability in a computer communications network, said method comprising the steps of:

• (a) collecting queue state information from intermediate nodes in the computer communications network that is being managed and storing these in memory of an intermediate device;
• (b) estimating the parameters of the partial differential equation that describes the traffic intensity dynamics of the discrete event system and storing these in memory of an intermediate device;
• (c) solving the partial differential equation that describes the traffic intensity dynamics of the computer communication network to yield the traffic intensity surface for the discrete event system and storing this in memory of an intermediate device;
• (d) generating optimal and suboptimal paths that are geodesics and non-geodesics, respectively, of the traffic intensity surface and storing these in memory of an intermediate device;
• (e) generating optimal and suboptimal routes by projecting the optimal and suboptimal paths on to the topology of the computer communications network and storing these in memory of an intermediate device;
• (f) actuating the traffic intensity by downloading the optimal routes to intermediate nodes for the purpose of updating their forwarding tables;
• and
• (g) determining the optimal times and locations for content caches to operate and storing these in memory of an intermediate device
• (h) actuating the traffic intensity by downloading the optimal times and locations for content caches to operate to cache content managers in the computer communications network;
• and
• (i) determining the times and locations of bandwidth imbalances in the computer communications network and their persistence and storing these in memory of an intermediate device;
• and
• (j) actuating the traffic intensity by downloading the bandwidth imbalances to a bandwidth manager in the computer communications network.