Multiplex communications – Diagnostic testing – Determination of communication parameters
Reexamination Certificate
2000-04-26
2004-06-29
Olms, Douglas (Department: 2661)
Multiplex communications
Diagnostic testing
Determination of communication parameters
C370S229000, C455S450000
Reexamination Certificate
active
06757257
ABSTRACT:
TECHNICAL FIELD
This invention relates generally to communication systems, and more particularly to data network design.
BACKGROUND OF THE INVENTION
The Internet has become a popular tool for sending and receiving information. In essence, the Internet comprises a worldwide network of communications equipment and service providers that use a common protocol for communicating. Data is transmitted from one user to another over a vast infrastructure of routers, servers, gateways and other similar devices. Typically, users on either end of the network operate computers equipped with appropriate software, devices and other components. Examples of such components include a modem and Internet browser application. Often, a user establishes a connection to the Internet through an Internet Service Provider (ISP). The underlying link level protocols stacks handle the messaging functions on both ends of the channel.
Typically, data is transmitted from a host through a router connected to an Internet Protocol (IP) network through a series of network links and switches to a receiving host. Packet-switched networks, in which messages from the transmitting host are divided into packets that are transmitted through the IP network and reassembled at the receiving host, are commonly used for the transport of digital data.
A challenge in designing a data network is determining the size and capacity of an edge router that will serve as a gateway to the IP backbone network for several terminals. An edge router requires sufficient capacity to handle data traffic to and from the IP network for the serviced terminals. In the past, the modeling of data traffic was often based on the packet arrival process, which was assumed to follow a Poisson or Erlang process. However, these modeling techniques have proven inadequate for broadband traffic analysis.
Capacity and performance of the traffic process for a router is often analyzed using a modeling formula called the M/G/1 model, used for modeling packet, frame and cell networks. However, basing an estimation of router throughput on the traditional M/G/1 model can lead to severely under-dimensioning a router. An under-dimensioned router lacks sufficient capacity to handle data traffic for its users, and may become congested due to excessive traffic entering the router for its capacity. Uncontrolled congestion can lead to data loss, long delays, and retransmissions. Traffic measurements made on real data networks have shown that the M/G/1 model is inaccurate for packet arrival processing in a data network, especially in a Local Area. Network (LAN) environment.
Due to the complexity and the highly bursty nature of the packet arrival process of data traffic, determining the required edge router throughput or capacity for data traffic is particularly difficult. Because no accurate model exists, routers are often chosen based on mere estimates of capacity requirements by counting the number of interfaces, for example, resulting in over-dimensioning the router. This results in larger routers being purchased than are necessary, creating increased costs and unused bandwidth.
What is needed in the art is a method of accurately determining the required throughput of an edge router in a data network.
SUMMARY OF THE INVENTION
The present invention achieves technical advantages as an algorithm, method, and system for dimensioning an edge router of a data network based on network parameters and the required grade of service (GoS). The throughput capacity required for the edge router is determined using calculations dependant on the number of subscriber requests rather than on the number of arriving packets, as in the prior art.
Advantages of the invention include accurate dimensioning of an access node such as an edge router in a data network, resulting in a cost savings in the purchase price of the router. A router is selectable that is capable of handling the data traffic for the subscribers or users of the data network terminals serviced by the edge router, with fewer occurrences of data loss, delays and retransmissions in the data network. The present invention provides an assessment of the performance and quality of service for a dimensioned router. The user-friendly invention is implemented based on network parameters input by the network designer.
REFERENCES:
patent: 6266322 (2001-07-01), Berger et al.
patent: 6282424 (2001-08-01), Wright et al.
Requirements, Planning and Choosing Technology,Traffic Analysis and Capacity Planning, Chapter 16, pp. 509-538.
G. Sundstrom, et al.,First results from LAN Traffic measurements at UAB—A comparison between ideal and real traffic.
Nail Adar, et al.,Telepack: An Advanced Teletraffic Analysis Package, IEEE Communication Magazine, Aug. 1998, pp. 84-87.
Ronald G. Addie et al.,Broadband Traffic Modeling: Simple Solutions to Hard Problems, IEEE Communications Magazine, Aug. 1988, pp. 88-95.
Ericsson Inc.
Olms Douglas
Wilson Robert W
LandOfFree
Traffic based dimensioning method for determining the... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Traffic based dimensioning method for determining the..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Traffic based dimensioning method for determining the... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3344786