Multiplex communications – Diagnostic testing – Determination of communication parameters
Reexamination Certificate
1998-12-10
2002-11-12
Ton, Dang (Department: 2661)
Multiplex communications
Diagnostic testing
Determination of communication parameters
C370S395210
Reexamination Certificate
active
06480470
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method for supporting traffic analysis, especially in an ATM (Asynchronous Transfer Mode) network, in which a transmission procedure is based on asynchronous time division multiplexing using fixed-length data packets or cells.
Technical Field
Software to support traffic analysis in an Asynchronous Transfer Mode (ATM) network has been disclosed in ref. [1] and ref. [6].
ATM is a transmission procedure based on asynchronous time division multiplexing using fixed-length data packets called cells. A cell contains a header field for control information and a payload field for user data information. The header includes, among other things, an address field that identifies a transmission channel. An ATM network comprises a set of nodes, each node being connected to one or more nodes within the network, which route the cells to their various destinations.
TECHNICAL BACKGROUND
The Problem Area
ATM networks may carry calls of a bursty traffical nature. Traffical analysis (ref. [2]) within an ATM network and during network operation may be used as an instrument to investigate the characteristics of the connections carried on the network and the characteristics of the network as a whole.
In general, such an on-line traffic analysis procedure will require:
1. A computational environment that permits the execution of traffic analysis logic.
2. One or more code blocks that serve as the logic of the traffic analysis to be performed.
3. A sheduler method that selects the next traffic analysis job to be executed.
4. A data collection method that gathers and stores the traffical data to be analysed.
5. A dispatcher method that gives an execution control to the traffic analysis job selected by the scheduler method.
The present invention relates to 3 and 4.
The difficulty with scheduling on-line traffic analysis jobs and traffical data collection in an ATM network is the transmission speed involved together with the great number of connections that may be carried simultaneously by the network.
Due to the transmission speed involved, for a traffic analysis method to be effective it must operate on, and be selected for execution at, very small time intervals. Otherwise, it will not be able to capture the traffical characteristics. The more fine the time interval granularity is, the more important the accuracy of the time intervals becomes: If imprecision is introduced in the scheduler method and the analysis is repeatedly applied to consecutive time intervals, the inaccuracy is aggregated and the analysis becomes unreliable.
Clearly, in order to perform any traffic analysis traffical data must be collected prior to the execution of the analysis job. Such a data collection method is strongly interlinked with the scheduler since, when a traffic analysis job is selected for execution, the traffical data pertaining to the particular analysis job must be handed over to that job. Since an ATM network may carry numerous connections simultaneously, the traffic analysis jobs may in total require a vast amount of traffical data. It is therefore very important that the data collection method operates efficiently in terms of storage together with the scheduler method.
The problems presented above call for a new method with minimal complexity that can handle the scheduling of numerous concurrent traffic analysis jobs on small time intervals and with high accuracy yet is efficient in terms of capacity and storage requirements.
Known Solutions and Problems with These
Two existing methods to support such on-line traffical analysis are:
Individual counting of cells pertaining to separate connections over the whole duration of each connection per link. Analysis may be scheduled per connection after disconnection has taken place.
Continuously counting of the aggregated number of cells, for all connections, on a link. Analysis may be scheduled at every cell interval, on a link basis.
The former may for instance be used to analyse the traffic of connections in terms of cells sent during connection time or in terms of the mean cell rate. The latter may be used to continuously analyse the ratio: link capacity to actual load.
The advantage of the known solutions is simplicity. However, they do not allow rigorous traffic analysis to be made. The problems with the known solutions are:
The granularity of the time interval over which the traffical analysis may be sheduled. Enabling analysis after disconnection or per cell interval are two extremes. What is required is a method that can schedule analysis jobs in a more flexible manner with respect to the time interval.
The granularity of the connection aggregation over which the analysis is performed. Allowing analysis to be scheduled on a link basis or per connection basis are two extremes. What is required is a flexible method that allows analysis to be performed on any level of traffical data aggregation.
Further Prior Art
U.S. Pat. No. 5,414,701 (Shtayer et al.) relates to a method and data structure for performing address compression in an ATM system. The system comprises a data accumulator means comprising virtual channel tables which are updated by traffic data. In the data traffic to an ATM link cells are retrieved by a cell extractor for thereafter by a cell-to-channel look-up means to be located in a virtual channel accumulator means.
SE 503 317 (Petersen/LM Ericsson) relates to a method for connecting STM cells in a circuit simulated ATM selector. The ATM selector comprises a unit which assigns a predetermined number of time slots to a cell.
FIG. 3
illustrates an STM cell at various reference points of time in the ATM selector, and the use of an associated time schedule means (chronometer) is also suggested.
Further publications related to this field of the art are: U.S. Pat. Nos. 5,317,563, 5,335,222, 5,361,253, EP 671,827 and EP 674,458.
BRIEF DISCLOSURE OF THE INVENTION
The main object of the present invention is to provide an improvement in a method as stated in the preamble, and according to the invention this object is achieved by using in combination a data accumulator means and a chronometer scheduling means, said data accumulator means comprising a channel data table which is continuously updated by the traffic data to be analyzed, and said chronometer scheduling means comprising a scheduler table which is continuously and sequentially selecting the next traffic analysis job to be executed, a scheduling of traffic analysis of a certain channel being effected when an updated channel address of said channel data table corresponds with a valid channel address in said scheduler table.
In other words, the invention presents a solution that is based on combined chronometer scheduler mechanism and data collection mechanism, both of which collaborate by means of a traffical data table and a scheduler table. The invention allows the scheduling of traffic analysis jobs and the interlinked collection of traffical data to be handled in a flexible and efficient manner.
REFERENCES:
patent: 5317563 (1994-05-01), Oouchi et al.
patent: 5335222 (1994-08-01), Kamoi et al.
patent: 5361253 (1994-11-01), Feijen et al.
patent: 5414701 (1995-05-01), Shtayer et al.
patent: 5555244 (1996-09-01), Gupta et al.
patent: 5602830 (1997-02-01), Fichou et al.
patent: 0 472 154 (1992-02-01), None
patent: 0 596 159 (1994-05-01), None
patent: 0 600 683 (1994-06-01), None
patent: 0 671 827 (1994-11-01), None
patent: 0 674 458 (1995-02-01), None
patent: 503 317 (1996-05-01), None
patent: 97/16034 (1997-05-01), None
Telefonaktiebolaget LM Ericsson (publ)
Ton Dang
LandOfFree
Method for supporting traffic analysis in an atm... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for supporting traffic analysis in an atm..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for supporting traffic analysis in an atm... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2994687