System for Parameter Analysis and Traffic Monitoring in Asynchronous Transfer Mode (ATM) Networks

The present invention relates to networks in general, and specifically to operational parameter analysis and traffic monitoring in asynchronous transfer mode (ATM) networks.

Various types of networks are well-known in the art. Common network types include Ethernet, token ring, and asynchronous transfer mode (ATM) networks. In each of these types of networks, it is desirable to analyze network operational parameters, such as network traffic and network errors, for planning, fault detection, and other purposes. Apparatus and methods for network operational analysis are well-known for Ethernet networks and are described, for example, in RMON MIB, RFC 1757, IETF standard, Ed. S. Waldbusser, Carnegie Mellon University, publication date February 1995, available, for example, from WWW Internet site http://ftp.internic.net. Traffic analysis apparatus and methods for token ring networks are also well-known and are described, for example, in Token Ring Extensions to RMON MIB, RFC 1513, Ed. S. Waldbusser, Carnegie Mellon University, publication date September 1993, available, for example, from WWW Internet site http://ftp.internic.net.

For ATM networks, suggestions have been made for traffic analysis such as, for example, the VC Steering MIB, a contribution to the ATM Forum. The VC Steering MIB suggests a method where a specific connection is copied to a monitoring device; that is, where all traffic passing over a specific connection is also sent to the monitoring device. Copying a specific connection to a monitoring device is believed to not be optimum in all cases because, for example, in the case where a large amount of traffic passes over the connection, the amount of traffic to be copied is large and hence a significant load may be placed on the ATM network by the copying itself.

The AToM MIB, RFC 1695, IETF standard, Ed. M. Ahmed and K. Tesnik, Bell Communications Research, publication date August 1994, available, for example, from WWW Internet site http://ftp.intenic.net, suggests a method for traffic monitoring in ATM networks where monitoring is performed on the individual circuit level. No suggestion is made in the AToM MIB to group together monitoring information from different circuits.

A press release entitled "Cisco Collaborates with RMON and Analysis Tool Vendors to Develop Specification for ATM Remote Monitoring", dated Feb. 20, 1996, describes an intention to present a draft specification for ATM traffic analysis to the Internet Engineering Task Force (IETF).

A proposal entitled "Proposal for Remote Network Monitoring MIB Extensions for ATM Networks" was distributed at the ATMF's "Test Sub-Working Group" and "Network Managing Sub-Working Group" meeting on Sep. 23, 1996. The proposal number is ATM.sub.-- Forum/96-0808R1.

Methods and apparatus used in ATM networking are described in the following sample list of publications:

User Network Interface (UNI) 3.1, published by the ATM Forum, September 1994; and

Specification for LAN Emulation over ATM--Version 1.0, Ed. Bill Ellington, IBM, published by the ATM Forum, publication date not available, particularly page 44.

The SNMP protocol is described in The Simple Book--An Introduction to Management of TCP/IP based Internets, Marshall T. Rose, published by Prentice Hall, N.J., USA, 1991.

Further publications may be accessed via the ATM Forum whose Internet address is: http://www.atmforum.com.

The disclosures of the above publications and of the publications cited therein are hereby incorporated by reference. The disclosures of all publications mentioned in this specification and of the publications cited therein are hereby incorporated by reference.

The present invention seeks to provide an improved system for analyzing operational parameters on an ATM network. Operational parameters typically include network traffic and network errors, and may also include any other suitable operational parameters of the ATM network. The present invention is intended to overcome known problems of prior art systems and of previous suggestions for analyzing ATM operational parameters, such as the problems described above in the method suggested by the VC Steering MIB.

There is thus provided in accordance with a preferred embodiment of the present invention a method for analyzing traffic on an ATM network, the network including a first plurality of stations to be interconnected by a second plurality of ATM connections, the method including intercepting at least one ATM connection characteristic of preferably each of the second plurality of ATM connections and storing the at least one ATM connection characteristic in an ATM connection characteristic memory, monitoring the amount of traffic passing over each of the second plurality of ATM connections and accumulating the amounts of traffic in the ATM connection characteristic memory, for at least one ATM connection subset including at least two of the second plurality of ATM connections which complies with a retrieval key, the retrieval key including a combination of at least one of the ATM connection characteristics, determining the total amount of traffic that has passed over the ATM connection subset by retrieving from the memory, using the retrieval key, the amounts of traffic passing over each ATM connection in the ATM connection subset and summing the amounts, and generating an output representation of the total amount of traffic.

Further in accordance with a preferred embodiment of the present invention the at least one ATM connection characteristic includes the identity of at least one of the stations interconnected by an individual ATM connection.

Still further in accordance with a preferred embodiment of the present invention at least one of the ATM connections includes a LAN-emulation ATM connection and the at least one ATM connection characteristic includes a LAN-emulation protocol ID.

Additionally in accordance with a preferred embodiment of the present invention the at least one ATM connection characteristic includes a traffic quality of service type.

Moreover in accordance with a preferred embodiment of the present invention the at least one ATM connection characteristic includes a function which the ATM connection serves.

There is also provided in accordance with another preferred embodiment of the present invention a method for analyzing performance of an ATM network, the network including a first plurality of stations to be interconnected by a second plurality of ATM connections, the method including intercepting at least one ATM connection characteristic of each of the second plurality of ATM connections and storing the at least one ATM connection characteristic in an ATM connection characteristic memory, monitoring at least one operation parameters of each of the second plurality of ATM connections and storing the operation parameters in the ATM connection characteristic memory, for at least one ATM connection subset including at least two of the second plurality of ATM connections which complies with a retrieval key, the retrieval key including a combination of at least one of the ATM connection characteristics, retrieving from the memory, using the retrieval key, the operation parameters for each ATM connection in the ATM connection subset and combining the operation parameters into a subset operation characterizing value which characterizes the operation of the subset, and generating an output representation of the subset operation characterizing value.

Further in accordance with a preferred embodiment of the present invention the at least one operation parameter includes at least one parameter characterizing errors transmitted over the ATM connection.

Still further in accordance with a preferred embodiment of the present invention the step of generating an output representation includes generating a computer display.

Additionally in accordance with a preferred embodiment of the present invention the step of generating an output representation includes generating an alarm when a threshold value is reached.

There is also provided in accordance with another preferred embodiment of the present invention ATM network apparatus including a first plurality of network elements including a plurality of stations and at least one switch, and a second plurality of ATM connections, wherein the first plurality of stations is interconnected by the second plurality of ATM connections associated with at least one switch, and wherein at least one of the first plurality of network elements includes operation parameter analysis apparatus, the operation parameter analysis apparatus including an ATM connection characteristic memory, an ATM connection characteristic monitor operative to intercept at least one ATM connection characteristic of each of the second plurality of ATM connections and to store the at least one ATM connection characteristic in the ATM connection characteristic memory, an ATM connection operation parameter monitor operative to monitor at least one operation parameters of each of the second plurality of ATM connections and to store the operation parameters in the ATM connection characteristic memory, an ATM connection operation parameter analyzer operative, for at least one ATM connection subset including at least one of the second plurality of ATM connections which complies with a retrieval key, the retrieval key including a combination of at least one of the ATM connection characteristics, to retrieve from the memory, using the retrieval key, the operation parameters for each ATM connection in the ATM connection subset and to combine the operation parameters into a subset operation characterizing value which characterizes the operation of the subset, and output apparatus operative to generate an output representation of the subset operation characterizing value.

Further in accordance with a preferred embodiment of the present invention the at least one network element which includes the operation parameter analysis apparatus includes at least one station.

Still further in accordance with a preferred embodiment of the present invention the at least one network element which includes the operation parameter analysis apparatus includes at least one switch.

Additionally in accordance with a preferred embodiment of the present invention the at least one network element which includes the operation parameter analysis apparatus includes a plurality of network elements each including the operation parameter analysis apparatus and also including an output value processing unit operative to receive a fifth plurality of subset operation characterizing values from a fifth plurality of ATM connection operation parameter analyzers and to combine the fifth plurality of subset operation characterizing values into a single value.

Moreover in accordance with a preferred embodiment of the present invention the retrieval key includes a Boolean combination of a plurality of the ATM connection characteristics.

Further in accordance with a preferred embodiment of the present invention the step of monitoring the amount of traffic includes monitoring the traffic passing, in one direction, over each of the second plurality of ATM connections.

Still further in accordance with a preferred embodiment of the present invention the traffic service type includes at least one of the following: CBR, RT-VBR, NRT-VBR, VBR, ABR.

Additionally in accordance with a preferred embodiment of the present invention the function which the ATM connection serves includes at least one of the following: PNNI, signaling, LAN-emulation.

Moreover in accordance with a preferred embodiment of the present invention the step of monitoring the amount of traffic includes separately monitoring the amount of traffic passing, in each of two directions, over each of the second plurality of ATM connections.

Further in accordance with a preferred embodiment of the present invention each of the second plurality of ATM connections includes a plurality of sub-connections, and the step of monitoring the amount of traffic comprises separately monitoring the amount of traffic passing over each sub-connection of each of the second plurality of ATM connections.

Still further in accordance with a preferred embodiment of the present invention the ATM connection characteristics include a sub-connection identifier.

Additionally in accordance with a preferred embodiment of the present invention the sub-connection identifier includes a direction associated with the sub-connection.

There is also provided in accordance with another preferred embodiment of the present invention ATM operation parameter analysis apparatus for use in conjunction with an ATM network, the ATM network including a first plurality of network elements including a plurality of stations and at least one switch, and a second plurality of ATM connections, wherein the first plurality of stations is interconnected by the second plurality of ATM connections associated with at least one switch, the operation parameter analysis apparatus including an ATM connection characteristic memory, an ATM connection characteristic monitor operative, as each of the second plurality of ATM connections is established, to intercept at least one ATM connection characteristic of each of the second plurality of ATM connections and to store the at least one ATM connection characteristic in the ATM connection characteristic memory, an ATM connection operation parameter monitor operative to monitor at least one operation parameters of each of the second plurality of ATM connections and to store the operation parameters in the ATM connection characteristic memory, an ATM connection operation parameter analyzer operative, for at least one ATM connection subset including at least one of the second plurality of ATM connections which complies with a retrieval key, the retrieval key including a combination of at least one of the ATM connection characteristics, to retrieve from the memory, using the retrieval key, the operation parameters for each ATM connection in the ATM connection subset and to combine the operation parameters into a subset operation characterizing value which characterizes the operation of the subset, and output apparatus operative to generate an output representation of the subset operation characterizing value.

Intercepting of ATM connection characteristics may either be during or after establishment of the ATM connections.

The present invention will be understood and appreciated from the following detailed description, taken in conjunction with the drawings in which:

FIGS. 1A and 1B, taken together, comprise a simplified partly pictorial, partly block-diagram illustration of a system for analyzing operational parameters on an ATM network constructed and operative in accordance with a preferred embodiment of the present invention;

FIG. 2 is a simplified block diagram illustration of a preferred implementation of the traffic analysis unit 110 of FIGS. 1A and 1B;

FIG. 3 is a simplified flowchart illustration of a preferred method of operation of the traffic analysis unit 110 of FIGS. 1A and 1B;

FIG. 4 is a simplified flowchart illustration of a preferred implementation of step 200 of FIG. 3;

FIG. 5 is a simplified flowchart illustration of a preferred implementation of a portion of step 230 of FIG. 3;

FIG. 6 illustrates the typical connections which are reported in a preferred embodiment of the switch;

FIG. 7 illustrates the various states of a connection in a preferred embodiment of the present invention; and

FIG. 8 illustrates a simplified block diagram of the interfaces between AMON and the switch.

Reference is now made to FIGS. 1A and 1B which illustrates a system for analyzing operational parameters in an asynchronous transfer mode (ATM) network constructed and operative in accordance with a preferred embodiment of the present invention. The terms "operational parameters" or "operation parameters", as used herein, refer to any one or more suitable operational parameters of an ATM network such as, for example, network traffic or network errors. Throughout the present specification, traffic is used as a particular example of an operational parameter when describing specific embodiments of the present invention. It is appreciated, however, that whenever the term "traffic" is used in describing specific embodiments of the present invention any other appropriate operational parameter, such as errors or such as a combination of one or more operational parameters, may also be used in place of or in addition to traffic.

The system of FIGS. 1A and 1B preferably comprises a first ATM switch 100, which may be any suitable ATM switch such as, for example, a Collage 740 Switch, commercially available from Madge Networks UK, Ltd., facsimile number (+044)-1-628-858016. The first ATM switch 100 preferably comprises a traffic analysis unit 110. Typically, the traffic analysis unit 110 comprises a suitably programmed processor 115 comprising suitable memory (not shown in FIGS. 1A and 1B), which processor 115 may preferably be an existing processor comprised in a suitable commercially available ATM switch used as the first ATM switch 110 or may alternatively be a suitable processor added to such suitable commercially available ATM switch. The traffic analysis unit also preferably comprises a traffic counter unit 120, which preferably comprises traffic counter hardware embedded in a suitable commercially available ATM such as a suitable commercially available ATM switch used as the first ATM switch 100. It is appreciated that, in the case where the particular ATM switch used as the first ATM switch 100 does not comprise embedded traffic counter hardware, such traffic counter hardware may be provided as part of the traffic counter unit 120, as is well known in the art. A preferred implementation of the traffic analysis unit 110 is described more fully below with reference to FIG. 2.

The term "ATM connection", as used herein, refers to the provision, over an ATM network, of a connection between two or more stations on the network, the connection typically comprising switching of cells such as ATM cells 122, each ATM cell 122 typically comprising data, from at least one source station to one or more destination stations, using ATM methods as are well known in the art. The switching of cells is typically provided in an ATM network by an ATM switch such as the first ATM switch 100. ATM connections are well known in the art and are described, for example, in UNI 3.1, published by the ATM Forum, referred to above.

It is appreciated that an ATM connection may be one-way, that is, may carry data only in one direction from, for example, a first station to a second station. Alternatively, an ATM connection may be two-way, that is, may carry data in both directions between, for example, a first station and a second station. Furthermore, it is appreciated that a two-way connection may comprise two one-way connections, one in each direction. Use of the term "ATM connection" herein is not meant to be limiting, and may be taken to refer, for example, to either a one-way connection or a two-way connection.

It is also appreciated that an ATM connection, typically a two-way connection, may be considered to comprise a plurality of sub-connections, each such sub-connection typically being a one-way connection. The term "ATM connection", as used herein, is understood to include such a sub-connection. Each sub-connection may be taken as having an identifier, typically comprising an indication of the direction of the subconnection.

As is well known in the art and as is described in, for example, ATM UNI 3.1, referred to above, cells typically include data information. Typically, all cells associated with an ATM connection include an ATM connection ID identifying the ATM connection.

A process known herein as an ATM connection establishment process is used to establish an ATM connection. A plurality of cells are typically used in the ATM connection establishment process. For the purposes of clarity of description, a cell used to establish a connection is termed herein a "connection establishment cell", such as connection establishment cells 124; it is appreciated that connection establishment cells 124 are typically a special case of ATM cells 122, differing from other ATM cells 122 primarily in that connection establishment cells 124 are used as part of an ATM connection establishment process. Typically, some connection establishment cells 124 may comprise data describing the ATM connection to be established such as, for example, the identity of the stations to be interconnected by the ATM connection being established.

It is appreciated that, generally, extraction of meaningful data from cells includes assembling a plurality of cells into a frame. The process of assembling a plurality of cells into a frame is well-known in the art.

As is well known in the art and as is described, for example, in Specification for LAN Emulation over ATM, referred to above, in an ATM network ATM connections can be used to emulate other kinds of network connections, typically in order to allow existing software which relies on such other connections to participate in an ATM network. Such emulation generally occurs using protocols which define the rules for carrying out the emulation. One example of such emulation is emulation of a local area network (LAN). An ATM connection used for emulation of a LAN, for example, is termed herein a LAN-emulation ATM connection, and a particular function performed by the LAN-emulation protocol connection and defining the LAN-emulation functionality of the connection is termed herein a LAN-emulation protocol type or LAN-emulation protocol ID.

As is well known in the art and is described, for example, in the ATM UNI 3.1, referred to above, ATM connections have various associated characteristics, including:

traffic service type, which may typically comprise one of the following: CBR, RT-VBR, NRT-VBR, VBR, ABR, or another appropriate type; and

application or function served, which may typically comprise one of the following: PNNI, signaling, LAN-emulation, or another appropriate function.

As is well known in the art, an ATM connection may have errors in the sense that at least one of the cells comprised in the ATM connection may be marked as erroneous, that is, as an errored cell, which is typically termed in the art "a bad HEC cell".

The traffic analysis unit 110 comprised in the first ATM switch 100 is preferably operable to analyze ATM traffic on ATM connections passing through the first ATM switch 100, such as a first ATM connection 125 and a second ATM connection 130. The analysis preferably occurs during creation of the ATM connection, also known as signaling setup phase. In order to perform said analysis, the traffic analysis unit 110 comprised in the first ATM switch 100 typically is also operative to store ATM connection characteristic information relating to ATM connections passing through the first ATM switch 100, the ATM connection characteristic information typically comprising the following:

an ATM connection ID identifying the ATM connection;

one or more ATM connection characteristics characterizing the ATM connection, the ATM connection characteristics typically comprising one or more of the following:

the identity of at least one of the stations interconnected by the ATM connection;

whether the ATM connection is a LAN-emulation ATM connection and, if so, the LAN-emulation protocol ID of the ATM connection;

traffic service type of the ATM connection, indicating the type of traffic service being provided by the ATM connection;

function served by the ATM connection such as, for example, a type of application service being provided by the ATM connection, such as, for example, LAN-emulation or PNNI.

While only two ATM connections are depicted in FIGS. 1A and 1B as passing through the first ATM switch 100, it is appreciated that, generally, a very large number of ATM connections may pass through the first ATM switch 100, such as, for example, 5,000 ATM connections. The operation of the traffic analysis unit 110 is described more fully below with reference to FIG. 3.

The system of FIGS. 1A and 1B optionally also comprises a second ATM switch 135. The second ATM switch 135, if present, may be similar to the first ATM switch 100, described above. While two ATM switches 100 and 135 are depicted in FIGS. 1A and 1B it is appreciated that only one ATM switch or, alternatively, a plurality of ATM switches may be used. The traffic analysis unit 110 comprised in the second ATM switch 135 is preferably operable to analyze ATM traffic on ATM connections passing through the second ATM switch 135, such as the second ATM connection 130 and a third ATM connection 140.

The system of FIGS. 1A and 1B preferably also comprises a plurality of ATM stations, shown in FIGS. 1A and 1B as ATM end stations 145, 150, and 155. It is appreciated that while, for the sake of simplicity, three ATM end stations 145, 150, and 155 are shown in FIGS. 1A and 1B, a larger number of ATM end stations such as, for example, 5,000 ATM end stations may be also used. Each of the ATM end stations 145, 150, and 155 may be any suitable ATM end station, such as a Collage 530, available from Madge Networks UK Ltd., referred to above.

The ATM end stations 150 and 155 each preferably comprise a traffic analysis unit 110, each traffic analysis unit 110 being preferably similar to the traffic analysis unit 110 described above with reference to the first ATM switch 100 and the second ATM switch 135. The traffic analysis unit 110 comprised in the ATM end station 150 is preferably operable to analyze ATM traffic on one or more ATM connections in which the ATM end station 150 participates, such as the first ATM connection 125 and the second ATM connection 130. The traffic analysis unit 110 comprised in the ATM end station 155 is similarly preferably operable to analyze ATM traffic on one or more ATM connections in which the ATM end station 155 participates, such as the second ATM connection 130 and the third ATM connection 140.

The ATM connections 125, 130, and 140 are each depicted in FIGS. 1A and 1B for the sake of simplicity as being two-way ATM connections, each being capable of carrying data in both directions between stations which participate in the ATM connection. It is appreciated, however, that one or more of the ATM connections 125, 130, and 140, for example, need not be two-way ATM connections but may be one-way ATM connections capable of carrying data only in one direction, or may be any other appropriate type of ATM connection available in an appropriate ATM network.

The system of FIGS. 1A and 1B also preferably comprises a data presentation unit 160, which may be any suitable pro-rammed computer such as a suitably programmed SPARC 5 Station, commercially available from Sun Microsystems, USA. The data presentation unit 160 is preferably operable to receive, typically via one or more ATM data connections 165, at least one result of traffic analysis performed by one or more of the traffic analysis units 110. The result typically comprises information about one or more ATM connections, the information typically comprising ATM connection characteristics and other information resulting from analysis of said one or more ATM connections by said one or more traffic analysis units 110, as described above.

The data presentation unit 160 is also preferably operable to display one or more results of said analysis. The data presentation unit 160 may preferably be operable to perform one or more of the following: to display said results on a display screen; to print said results; to store said results on an electronic medium or any other medium; to further process said results; or to send said results to another unit (not shown) via the ATM network or by any other means.

The data presentation unit 160 may also be operable to combine and process the data from one or more of the traffic analysis units 110 prior to, during, or after any of the previously mentioned operations performed by the data presentation unit 160. Such combining and processing may preferably include any appropriate combining and processing, such as, for example, one or more of the following: combining related data by summing or otherwise, computing an average, or computing a statistical result.

The data presentation unit 160 is shown in FIGS. 1A and 1B as a unit separate from other elements of FIGS. 1A and 1B. It is appreciated, however, that the data presentation unit may alternatively be comprised in an ATM switch such as the first ATM switch 100 or the second ATM switch 135, or with an ATM end station such as one or more of ATM end stations 145, 150, and 155. It is also appreciated that a plurality of data presentation units 160 may be used.

The operation of the system of FIGS. 1A and 1B is now briefly described. Except when otherwise stated, the description of the operation of the system of FIGS. 1A and 1B, when making reference to one of the traffic analysis units 110, may preferably apply to any one or more of the traffic analysis units 110 in the system of FIGS. 1A and 1B, including traffic analysis units 110 comprised in an ATM switch such as the first ATM switch 100 or the second ATM switch 135, and traffic analysis units 110 comprised in an ATM end station such as one or more of the ATM end stations 145, 150, and 155.

One of the traffic analysis units 110 such as, for example, the traffic analysis unit 110 comprised in the ATM switch 100, listens to cells as the cells pass through the ATM switch 100. The terms "listen" and "intercept" are used interchangeably throughout the present specification to refer to analyzing certain characteristics of cells as they pass through the ATM switch 100 comprising the traffic analysis unit 110, without interrupting or disturbing the passing of the cells through the ATM switch 100. Preferably, the traffic analysis unit 100 is operative to listen to all such cells.

The traffic analysis unit 110 comprised in the ATM switch 100 is preferably operable, in intercepting such cells, to listen for a connection setup process comprising at least one connection establishment cell 124 , as described above. A connection setup process, as described above, is a process setting up an ATM connection between two or more stations, typically using a signaling protocol as defined, for example, in ATM UNI 3.1, referred to above. A connection setup process typically comprises transmitting ATM connection characteristic information comprising one or more of the following:

the identity of the stations interconnected by the ATM connection;

whether the ATM connection is a LAN-emulation ATM connection and, if so, the LAN-emulation protocol ID of the ATM connection;

traffic service type of the ATM connection, indicating the type of traffic service being provided by the ATM connection;

function served by the ATM connection such as, for example, a type of application service being, provided by the ATM connection, such as, for example, LAN-emulation or PNNI.

The traffic analysis unit 110 is preferably also operative, upon identifying a connection setup process, to store one or more item of ATM connection characteristic information comprised in the connection setup process in an ATM connection characteristic memory (not shown in FIGS. 1A and 1B) typically comprised in the traffic analysis unit 110. It is a particular feature of the present invention that, by listening to connection setup processes, ATM connection characteristic information such as that mentioned above may be identified and stored for later use in analyzing one or more operational parameters of an ATM connection.

The traffic analysis unit 110, having thus identified an ATM connection, is preferably operative to monitor ATM cells 122 and to store traffic information from the traffic counter 120, and/or information on other operation parameters thereof as described above, in the ATM connection characteristic memory. Said stored information is preferably stored according to ATM connection characteristic; that is, for example, in the case of ATM connection characteristic "the ATM connection comprises traffic sent from end station 150", sufficient information is preferably stored to enable retrieval of stored information for all such ATM connections having the ATM connection characteristic "the ATM connection comprises traffic sent from end station 150". Another example of an ATM connection characteristic would be "the ATM sub-connection direction is from end station 150". It is appreciated that there is a wide variety of possible examples.

Upon request, typically upon request from the data presentation unit 160, the traffic analysis unit 110 is preferably operable to transmit a result of said request to the data presentation unit 160, typically via one of the ATM data connections 165. The request typically comprises a retrieval key indicating ATM connection characteristics for which the operational parameters of ATM connections with the ATM connection characteristics indicated by the retrieval key, are of interest. Such a retrieval key might, for example, comprise "the ATM connection comprises traffic sent from end station 150". Such a retrieval key might also comprise a Boolean combination of ATM connection characteristics such as, for example "the ATM connection comprises traffic sent from end station 150 OR the ATM connection comprises traffic sent from end station 155".

In producing the result, the traffic analysis unit 110 preferably combines the stored operation parameters for one or more ATM connections which match the retrieval key. In the case where the stored operation parameter is traffic, combining typically comprises adding to-ether to form a sum. It is appreciated, however, that in the case of traffic and in the case of other operation parameters, such combining may comprise applying any of a wide variety of appropriate multi-variable functions to said stored operation parameters such as, for example, a maximum or minimum function or other appropriate multi-variable function.

The data presentation unit 160, upon receiving the result, preferably generates an output representation of the result. The output representation may include any appropriate output representation such as, for example:

generating a computer display;

generating a hardcopy display;

storing the result on a magnetic computer medium or any other appropriate computer medium; or

generating an alarm when a threshold value is reached, the threshold value typically representing a maximum or minimum desirable value for the operation parameter.

Reference is now made to FIG. 2, which is a simplified block diagram illustration of a preferred implementation of the traffic analysis unit 110 of FIGS. 1A and 1B. The apparatus of FIG. 2 preferably comprises an ATM connection characteristic monitor 168, preferably operative to monitor ATM connection characteristic parameters of ATM connections passing through the unit which comprises the traffic analysis unit 110.

The apparatus of FIG. 2 also preferably comprises an ATM connection operation parameter monitor 170, preferably operative to monitor ATM connection operation parameters of ATM cells 122 passing through the unit which comprises the traffic analysis unit 110.

The apparatus of FIG. 2 also preferably comprises an ATM connection characteristic memory 175, preferably operative as described above with reference to FIGS. 1A and 1B to store ATM connection characteristics, typically received from the ATM connection characteristic monitor 168, and/or ATM connection operation parameters, typically received from the ATM connection operation parameter monitor 170.

The apparatus of FIG. 2 also preferably comprises an ATM connection operation parameter analyzer 180 which is preferably operative to process a request including a retrieval key and to combine information stored in said ATM connection characteristic memory 175 to form a result of said request.

The apparatus of FIG. 2 also preferably comprises output apparatus 185 which is preferably operable to transmit the result to the unit which made the request, typically the data presentation unit 160. Typically, the results are transmitted over the ATM network using the SNMP protocol, described, for example, in The Simple Book, referred to above. It is appreciated that functions described above with reference to the data presentation unit 160 may, alternatively or in addition, be performed by the output apparatus 185.

Reference is now made to FIG. 3, which is a simplified flowchart illustration of a preferred method of operation of the traffic analysis unit 110 of FIGS. 1A and 1B. The method of FIG. 3 preferably comprises the following steps:

The traffic analysis unit 110 captures and stores ATM connection characteristics during ATM connection setup (step 200), thus storing identification and characteristic information for one or more of the existing ATM connections which pass through the unit in which the traffic analysis unit 110 is comprised. The information is typically stored in the ATM connection characteristic memory 175 (FIG. 2).

The traffic analysis unit 110 collects information, preferably on each ATM connection for which information was stored in step 200 (step 210). Typically, in the case of traffic information, the information is collected by accumulating traffic information from embedded traffic counter hardware, which may be provided, as described above, as part of the traffic counter unit 120. In the case of other information such as, for example, error information, the information may typically be collected from other counter hardware which may be provided as part of the traffic counter unit 120 or otherwise as part of the ATM switch 100, or may be obtained by intercepting ATM cells 122 as described above. The information collected in step 210 is typically stored in the ATM connection characteristic memory 175 (FIG. 2).

Typically, upon request from the data presentation unit 160 or otherwise, the traffic analysis unit 110 groups the information stored in steps 200 and 210 according to a search key provided as part of the request (step 230). It is appreciated that, as explained above, grouping may comprise summing or some other appropriate multi-variable function of the information. The grouped information is then output (step 235), typically to the source of the request such as, for example, the data presentation unit 160.

Reference is now made to FIG. 4, which is a simplified flowchart illustration of a preferred implementation of step 200 of FIG. 3. The method of FIG. 4 preferably comprises the following steps:

The traffic analysis unit 110 listens, as described above, for one or more connection setup processes (step 240). The traffic analysis unit 110 then extracts identifying characteristics from the one or more connection setup processes (step 250), and stores the identifying characteristics, typically in the ATM connection characteristic memory 175 (FIG. 2) (step 260).

Reference is now made to FIG. 5, which is a simplified flowchart illustration of a preferred implementation of a portion of step 230 of FIG. 3, in which grouping or summing is performed. The method of FIG. 5 preferably includes the following steps:

ATM connections are preferably grouped according to a search key as follows. The next ATM connection, which may be the first ATAM connection, is preferably examined (step 270). If no ATM connections remain, the method of FIG. 5 preferably terminates.

A check is preferably made to see whether the stored characteristics of the ATM connection match the search key (step 280). If not processing continues with step 270.

If the stored characteristics of the ATM connection did match the search key in step 280, the stored ATM connection is preferably combined, as described above, with stored data from other ATM connections corresponding to the search key. Processing then continues with step 270.

There follows a detailed description of a preferred embodiment of the traffic analysis unit 110 of FIGS. 1A and 1B, particularly for use with the Collage 740 Switch referred to above or with any ATM switch constructed as defined in ATM Forum specifications, obtainable by joining the ATM FORUM. The ATM Forum's Internet address is: http://www.atmforum.com.

1.1. Overview

AMON--ATM Remote Monitoring, is a method which provides monitoring capabilities of ATM networks in a fashion similar to that provided for shared networks by RMON compliant probes. AMON uses the RMON MIB (to the possible extent), and incorporates a hierarchical monitoring strategy.

AMON may be provided by embedded software running in the ATM switch in parallel to the switch's other packages, and by a graphical management application used for displaying the information.

The applications provided by AMON are:

ATM Switch Statistics: provides aggregate statistics on the behavior of the switch.

ATM Port Statistics: provides statistics on the traffic passing through all the switch's ports.

ATM Host Statistics: provides statistics on the traffic produced by all ATM hosts communicating through the switch. ATM Hosts are identified by their ATM addresses and not by their MAC addresses.

ATM Host Matrix: provides a host matrix on all sessions passing through the switch. The matrix relates to ATM addresses and not to MAC addresses.

The first phase of AMON provides monitoring, capabilities for the following, ATM forum standards:

UNI 3.0 and 3.1

LANE 1.0

1.2. Host/Matrix Support

The AMON Host and Matrix applications are heavy users of memory. As a result, the amount of addresses and address pairs which are stored and supported by each application are a function of the amount of memory which is installed on the switch:

1.3. Limitations

1. ATM Host and Matrix statistics are typically partially dependent on the transmission of the Calling Address IE during signaling setup. LANE stations typically transmit this IE, and are therefore fully analyzed by AMON. However, other ATM stations may not transmit this IE, and as a result the ATM statistics which are collected for these stations (in the ATM Host and Matrix applications) may be partial.

2. In Signaling 4.0, when leaf initiated join without root notification is used by a host, and the join is performed by a switch which is downstream to the switch in which AMON is performed, AMON has no knowledge of the new station, and therefore is not able to report traffic pertaining to this ATM station.

1.4. Definition & Abbreviation

AMON--ATM remote MONitoring

RMON--Remote MONitoring

Eth10--Ethernet 10 Mbps

Eth100--Ethernet 100 Mbps

NMS--Network Management Station

BUS--the LAN Emulation Broadcast and Unknown Server

ELAN--Emulated LAN

LANE--the LAN Emulation protocol

LEC--LAN Emulation Client

LECS--LAN Emulation Configuration Server

LES--LAN Emulation Server

2.1. Overview

The AMON probe is a part of the ATM switch's software, running in parallel to other functions performed by the CPU. The probe collects ATM level statistics. These are collected by analyzing various properties of the VCs during signaling setup (using information available in the setup process), and correlating this data with the per-VC counters collected by the switch's HW. This enables the CPU to collect RMON-like statistics at the ATM level.

2.2. ATM Statistics

ATM level statistics are collected by analyzing properties of the VCs during signaling setup, and correlating these properties with the cell counters provided per VC by the switch's HW.

The ATM level functions provided by the AMON probe are:

ATM Switch Statistics,

ATM Port Statistics

ATM Host Statistics

ATM Host Top N,

ATM Host Matrix,

ATM Switch statistics and ATM Port statistics are collected by the probe at all times. The other 3 groups--ATM Host, ATM Host Top N, and ATM Matrix statistics are collected by the probe upon request from the management station. The probe is capable of collecting all groups simultaneously.

2.2.1. Required Signaling Information

To enable the probe to collect ATM level statistics, the switch's signaling entity typically transfers to the AMON entity for all connections passing through the switch, data which typically but not necessarily includes the following:

ATM addresses connected at both sides of the connection,

The connection's calling party,

For Point-to-Multipoint connections--the root (for UNI 3.0 and 3.1 this is actually redundant with the calling party).

Service used by the connection (ABR, UBR, VBR, or CBR),

Amount of bandwidth allocated to the VC,

Point-to-Point or Point-to-Multipoint,

The usage of the VC: PNNI, Signaling, LANE Control/Configuration, LANE 802.3 Data Direct, LANE 802.5 Data Direct, LANE 802.3 Multicast, LANE 802.5 Multicast, or "other".

The creation order of the VC (in relation to other VCs).

Notes:

1. The above list defines the data that will be available internally to the AMON task, and not externally via MIB.

2.2.1.1. Connection Types

This section describes one possible implementation for collecting data on point-to-multipoint connections.

Point-to-Multipoint connections maintained by the connection manager typically have one of 3 types:

Root: this is the end to which the calling ATM entity is connected.

Branch: this is an internal branching. The maximum possible number of branches is the number of physical ports in the switch.

Leaf: a single leaf exists for each of the multipoint stations.

FIG. 6 illustrates the typical connections which are reported in a switch of a preferred embodiment of the present invention.

The connection types which are typically reported to AMON are:

Conn. #1--Tunck

Conn. #2--Branch

Conn. #3--Branch

Conn. #4--Leaf

Conn. #5--Leaf

Conn. #6--Leaf

Each connection remains in the data base for approx. 30 seconds after it is actually deleted from the switch. Each connection has a status to represent this. The various states are represented in FIG. 7.

For point to multi point connections, the probe will use a specific connection type for calculating specific counters, n UNI 3.0 and UNI 3.1. This is described in the following table:

This section and all other MIB definitions disclosed herein provide examples of the information that can be collected by AMON. These examples are not intended to be limiting and in particular additional other information can be collected.

The MIB items represent preferred embodiments of the present invention and are not essential features of the invention.

The AMON probe collects statistics related to the switch as a global entity.

Switch statistics which are collected by summing the relevant VC counters of all VCs active on the switch. Statistics are typically collected by the probe at all times. The following table defines the counters which are supported at the switch level

2.2.2. ATM Port Statistics

The AMON probe collects statistics related to the physical ports of the ATM switch. Statistics are collected at all times for all ports in the PortTable. When a port is added to the switch or removed from it in run-time, the table is updated accordingly to reflect the correct status of the switch.

Port statistics are collected by summing the relevant VC counters of all VCs active on the port. The counters which are supported for each port are:

2.2.3. ATM Host Statistics

ATM Host Statistics are provided by the RMON Host group, and preferably by additional non-RMON mibs. Statistics regarding an ATM host are computed as the sum of the traffic on all VCCs connected to this host. Collection of statistics are activated by the RMON MIB's hostControlTable, both for the RMON MIB counters and for the non-RMON table of the non-RMON mibs.

2.2.3.1. RMON MIB Support

The probe typically supports the following RMON MIB tables:

hostControlTable

hostTable

hostTimeTable

Support for the RMON tables are is provided according to the definitions in the RMON MIB, with the following modifications which are required for ATM support:

The following implementation specific issues are typically utilized:

1. The maximum size of the host table is limited, as defined above. Any hosts over this discovered by the probe are not written to the host tables.

2. The datasource of the hostControlTable can be typically: mib2.interfaces.ifTable.ifEntry.ifIndex.1.

3. When the AMON probe is instructed to start collection of ATM Host statistics, the host tables (hostTable, hostTimeTable, and AtmHostTimeTable) are initialized with entries for all the ATM hosts which are active on the switch at that time.

4. The tables are updated whenever a new ATM address is identified, i.e., when a new VC is created and either of the two hosts communicating on this VC have not yet been registered in the host table.

5. When an existing ATM host disconnects all its connections and stops communicating over the switch, it is removed from the RMON Host tables, and the counters for this entry remain frozen with their last values. If this host later opens new connection on the switch, the probe continues counting using the previously frozen counters.

Trigger:

The collection of ATM Host Statistics is triggered when a valid entry is created in the hostControlTable.

AMON Probe Actions:

When an ATM Host collection session begins, the actions performed by the probe preferably include some or all of the Following actions:

1. Scans the list of existing VCs, and for each of them creates an entry in the hostTable and in the non-RMON atmHostTable.

2. Periodically polls all VCs to update the host counters in both the RMON and non-RMON tables (Note: this is an implementation issue, and polling of all VCs may be performed in other fashions, e.g., in response to an SNMP get request).

3. When a new VC is created, checks if the hosts at the ends of this VC are already in the tables, and if not adds them.

4. When an existing VC is deleted, removes this VC from the list of VCs to be polled, but does not remove the hosts connected to it from the tables.

2.2.4.2 Non-RMON MIB Support

The probe supports the non-RMON atmHostTimeTable. This table is indexed by the same indices as the RMON hostTimeTable, and is created by the RMON hostControlTable. The counters which are typically collected are:

2.2.4. ATM Host Top N

ATM Host Top N provides an agent based sorting mechanism on the data found for ATM hosts. This is done using the standard RMON MIB and non-RMON additions to it.

The AMON probe typically supports all tables required by the Host Top N mechanism, i.e.:

hostTopNControlTable,

hostTopNTable.

To support the Top N functionality for the additional counters provided by the non-RMON additions to the RMON MIB's Hosts Group, some new values are added to hostTopNRateBase. The values RateBase values defined in the standard RMON MIB are not supported. The values which are added to hostTopNRateBase are:

hostTopNVcs(101)

hostTopNInCells(102)

hostTopNOutCells(103)

hostTopNOutUbrCells(104)

hostTopNOutAbrCells(105)

hostTopNOutRtVbrCells(106)

hostTopNOutNrtVbrCells(107)

hostTopNOutCbrCells(108)

hostTopNOutLaneCells(109)

Top N statistics are typically limited to a maximum of 30 hosts.

Trigger:

The collection of Host Top N statistics is typically triggered when one of the following happens:

1. A valid entry is created in the hostTopNControlTable, with timeRemaining.noteq.0.

2. The value of timeRemaining in a valid entry is changed to a non-zero value.

NOTE: The following implementation specific issues are typically utilized:

1. The maximum size of the report may typically be 30 hosts. If a manager requests more, the value of hostTopNGrantedSize is typically set to 30 by the agent.

2. The HostTopNTable is updated only when hostTopNTimeRemaning reaches 0, and then remain s frozen.

3. The minimum allowed value for hostTopNTimeRemaining is typically 10 seconds. Attempts to set values smaller than this are rejected by a badValue response.

4. The item hostTopNGrantedSize typically reports the exact number of hosts that are collected in the hostTopNReport. This means that the value of this item may change after the report has been created.

2.2.4.1. Relationship Between Tables

The following relationship between the host table and the hostTopN table can typically be implemented:

2.2.5. ATM Host Matrix

The ATM Host Matrix is provided by the RMON Matrix group. The probe supports all tables defined in the RMON MIB's matrix group, i.e.:

matrixControlTable

matrixSDTable

matrixDSTable

In addition, the probe supports the non-RMON amonMatrix SDTable.

Statistics regarding a session are computed as the sum of the traffic on all VCCs between the 2 hosts. The contents of the MIB counters which are typically collected for every session are defined below:

The following implementation specific issues are can typically be utilized:

1. The maximum size of the Matrix table can typically be limited as defined above. Any sessions over this discovered by the probe are not written to the Matrix tables.

2. The datasource of the matrixControlTable may typically be the following: mib2.interfaces.ifTable.ifEntry.ifIndex.1.

3. When the AMON probe is instructed to start data collection via the matrixControlTable, the SD and DS tables are initialized with entries for all the sessions which are active on the switch at that time.

4. The tables are updated whenever a new session is identified, i.e., when a new VC is created and the 2 hosts communicating on this VC are not already listed in the matrix tables.

5. When an existing session is disconnected for any reason, it is not removed from the RMON Matrix tables, and the counters for this entry remain frozen with their last values. If this session becomes active at a later time, the probe continues counting using the previously frozen counters.

Trigger:

The collection of ATM Matrix Statistics is triggered when a valid is created in the matrixControlTable.

AMON Probe Actions:

When an ATM Matrix collection session begins, the probe performs the following actions:

1. Scans the list of existing VCs, and for each of them creates an entry in the matrixTable.

2. Periodically polls all VCs to update the matrix counters. (Note: this is an implementation issue, and polling of all VCs may be performed in other fashions, e.g., in response to an SNMP get request).

3. When a new VC is created, checks if this session is already in the table, and if not adds it.

4. When an existing VC is deleted, removes this VC from the list of active VCs, but does not remove the session defined by this VC from the RMON tables.

2.2.5.1. Non-RMON MIB Support

The probe supports the non-RMON atmMatrixSDTable. This table is indexed by the same indices as the RMON MatrixSDTable, and is created by the RMON MatrixControlTable. The counters which are typically collected are:

2.2.6. General--Updating Lists in MIB Tables

For all ATM related applications which contain lists of entities (i.e., ATM Port Stats, ATM Host Stats, and ATM Host Matrix) the agent updates the relevant MEB tables when a new entity is discovered--i.e., a new host, session, port or VC. This typically happens in two cases:

When a control entry is activated, all the VCs active on the switch at the time is scanned, and the table is initialized to contain the list of all entities already active on the switch.

When a new VC is created the relevant MIB tables are updated to include any new entities discovered when this happened: hosts--if the VC is used by hosts that are not already in the table, and sessions--if the VC is between two hosts that are not yet communicating with each other via the switch.

When a VC is removed from the switch, the entries pertaining to that VC are typically treated as follows:

1. ATM Hosts are not removed from the RMON and non-RMON host tables,

2. ATM Sessions are not removed from the RMON Matrix tables and the non-RMON tables.

When a port is removed or added to the switch, the port tables are updated to reflect the status of the switch.

2.3. Interfaces to Other Switch Functions

This section describes an interface suitable for the Madge (NAME OF..) Collage 740 ATM switch and commercially available from Madge Networks Ltd., Atidim Technology Park, Building 3, Tel-Aviv 61131, Israel, and is typically different for other commercially available switches.

The AMON application interfaces with a number of switch functions. A simplified block diagram of a preferred interface between AMON and the switch is illustrated in FIG. 8.

Examples of the information that can be collected by AMON are described by the following MIB: ##SPC1##

It is appreciated that the software components of the present invention may, if desired, be implemented in ROM (read-only memory) form. The software components may, generally, be implemented in hardware, if desired, using conventional techniques.

It is appreciated that various features of the invention which are, for clarity, described in the contexts of separate embodiments may also be provided in combination in a single embodiment. Conversely, various features of the invention which are, for brevity, described in the context of a single embodiment may also be provided separately or in any suitable subcombination.

It will be appreciated by persons skilled in the art that the present invention is not limited to what has been particularly shown and described hereinabove. Rather, the scope of the present invention is defined only by the claims that follow: