System and method for automatic segment resolution on a local area network

ES 2 201 073 T3

DESCRIPTION

System and procedure for automatically determining a segment in a local area network. Field of the invention

The present invention is generally directed to a system and a method for determining the position of a station in a network and, more specifically, to a system and a method for determining the position of a station in a segment of a network that is bridged from transparent mode to another segment on that network.

Description of the background of the invention

Various computer network topologies and protocols are currently available. Of particular interest to the present invention, and probably the most widely used scheme for controlling a local area network in a bus structure, is carrier sense multiple access with collision detection or the IEEE 802.3 standard. The most widely used implementation of carrier sense multiple access with collision detection is found in the Ethernet specification.

Most local area networks can be interconnected by gateways, routing circuits, or bridges. A bridge provides the interconnection between two local area networks of the same type and makes decisions about switching frames between local area networks based on the address of a station. Bridges may use different algorithms when making their routing decision, some of which work transparently to end stations on the network.

The efficient operation of a computer network is critical both in terms of macrocomputer performance and the productivity of individual workstations. The ability to measure network response times associated with workstations and the ability to determine bottlenecks or other network inefficiencies that can distort network response time are important to network designers, administrators, and end users. Co-pending United States Patent Application Serial No. 992,439, which is assigned to the assignee of the present invention and which is incorporated herein by reference, describes an original system and method for obtaining operational statistics for an area network. based on the role of a station in the network and the network traffic that is motorized. Such operational statistics are useful to keep the network running efficiently.

In the token-passing ring network environment, the network monitor, as described in the aforementioned co-pending application, is able to correctly identify the addresses of the stations residing on its ring using the tables. active monitors and reserving the ring protocol with witness pass. Knowing which stations are on the local ring is essential to determine the operational statistics as described in that application. However, in the Ethernet environment it is not always possible for the network monitor to correctly identify the addresses of the stations residing on its local segment because multiple Ethernet segments can be transparently bridged to each other. Such a transparent bridge has the effect of causing the traffic to be displayed to a network monitor as being transmitted by a station on the local segment when, in fact, it may have been transmitted by a station on a bridged segment.

US 4,799,211 deals with a remote segment monitor that determines the identity of the node and the signal strength of each transaction on a segment cable of a local area network system. Each node is identified by decoding the header or preamble of the transaction. The signal strength associated with each network node is tracked over time so that components that are operating outside acceptable limits, components that exhibit erratic behavior, and components with degraded functional performance can be identified.

Document WO 88/06822 deals with a monitoring system for a local area network in which a monitor manager receives information from monitors connected to different bases of a local area network. A look-up table contains the association between the counters on each monitor and the characteristics with which they are associated such that packet processing by the monitors can be performed within a minimum packet time.

WO 92/06547 deals with a method for finding out the topological characteristics of a network that has traffic in the form of discrete packets that have source and destination information. Traffic is monitored by monitors and partial traffic matrices are built. A central processing station analyzes the partial traffic matrices to find out the topological characteristics.

US 5,179,554 deals with a method for tracing the topology of an extended local area network by augmenting a list of station addresses maintained by a bridge with a list of repeater numbers and repeater port numbers. The bridge periodically selects a station address table and passes it to all repeaters connected to the bridge. Repeaters expect the selected address to appear as a source address in a message on one of their ports. Upon detecting the address, the repeaters communicate the repeater port number to the bridge.

Accordingly, a need exists for a system and method that enables a network monitor to determine the position of a station on an Ethernet network when segments of that network are transparently bridged to other segments. Network monitors can use such information to determine the station's role and calculate operational statistics based on that determination. Ultimately, such information is critical to maintaining the efficient functional performance of that network.

Summary of the invention

The present invention is directed to a method for determining ownership of a station in a network, and to a system for determining ownership of a station, as defined in the appended claims.

Thus, the system and method of the present invention allow a network monitor to determine ownership of a station on an Ethernet network when segments of that network are transparently bridged to other segments. Such determination is useful for the network monitor to calculate operational statistics by monitoring network traffic, such as

ES 2 201 073 T3 statistics important to maintaining the efficient operation of a network. These and other advantages and benefits of the present invention will become apparent from the description of a preferred embodiment below.

Brief description of the drawings

In order for the present invention to be clearly understood and easily practiced, preferred embodiments will now be described, by way of example only, with reference to the accompanying figures, in which:

Figure 1 illustrates a simplified representation of a typical local area network bus configuration with multiple bridged segments in which the apparatus, system and method of the present invention may be employed;

Figure 2 illustrates a block diagram of the functions of an automatic segment determination program construction according to the techniques of the present invention;

Figure 3 illustrates a flow chart explaining the process flow of the present invention;

Figures 4A to 4L illustrate command block structures for an automatic segment determination program embodying the method of the present invention;

Figure 5 illustrates a first example timeline for the system and method of the present invention; and Figure 6 illustrates a second example timeline for the system and method of the present invention.

Detailed description of the preferred embodiment

Figure 1 illustrates a simplified representation of a typical computer communications network 10 in which the apparatus, system, and method of the present invention can be used to generate operational statistics. Like reference numbers are used in the various figures to designate like elements.

More particularly, and with reference to Figure 1, a local area network 10 is shown comprising m segments, a first segment identified by the number 12, a second segment identified by the number 14a and a mth segment identified by the number 14a. number 14b. Although the present invention may have application in any network whose segments are transparently bridged to each other, by way of example only, the present invention will be described as implemented in Ethernet. Those skilled in the art will understand that the number m of segments depends on the application. Each of the segments 12, 14a and 14b, respectively, is a bus structure that implements the carrier sense multiple access protocol with collision detection.

A plurality of stations 21a may be attached to the first segment 12. The number of stations 21a may vary from one to n, depending on the application. Each of the stations 21a can be, for example, a workstation attached using an appropriate Ethernet adapter board (not shown). A network monitor 20 is also attached to the first segment 12. Network monitor 20 can be any type of computer and is attached to network 10 using an appropriate Ethernet adapter board (not shown). By way of example, the network monitor 20 may be a personal computer that implements the IBM AT bus structure or the IBM microchannel, and the Ethernet adapter board may be a Model # 507 (for an AT bus) or a Model # 523 (for a microchannel) from 3Com. The Ethernet adapter board must be capable of operating in the promiscuous mode in which every frame transmitted on the network 10 is read and received by the Ethernet adapter board. The network monitor 20 houses the network monitor program functionally shown in Figure 2 and further described herein.

In a similar manner, the second segment 14a has x stations 25a and a second network monitor 24a attached to it, the number x depending on the application. Similarly, the m-th segment has y stations 25b and a m-th network monitor 24b attached to it, depending on the number and the application. In each case, the network monitors 20, 24a, and 24b are preferably configured in a similar manner. In a preferred embodiment and to ensure accuracy, each of the segments 12, 14a, 14b should have a network monitor 20, 24a, 24b attached to it.

As will be understood by those skilled in the art, the configuration of segments 12, 14a, 14b in Figure 1 is intended to be an example only. Countless other configurations of segments are possible, with the particular configuration depending on the application. The invention described here may also work in those various other configurations.

As can be seen in Figure 1, the first segment 12 and the second segment 14a are interconnected by the bridge 16. Also, the second segment 14a is connected to the same segment 14b by the bridge 16. As used herein, a bridge 16 is a device that provides interconnection between segments of a local area network 10. Bridge 16, using one of numerous possible algorithms, makes frame-switching decisions between segments based on the address of a station.

The switching algorithms can be transparent to the end stations. In an IEEE 802 local area network, interconnection between local segments 12, 14a, 14b may occur, for example, in the Media Access Control (MAC) sublayer. As will be appreciated by those skilled in the art, the carrier sense multiple access media access control frame format with collision detection includes both a destination address and a source address. A bridge 16 can use a number of different simple algorithms to make its switching decisions. By way of example only, such an algorithm is a "learn as you go" algorithm in which bridge 16 maintains a database of stations 21a, 25a, 25b that send messages on their joined segments 12, 14a, 14b, respectively. When bridge 16 monitors a message, it knows on which side of bridge 16 the message originated and adds that station address and its position relative to bridge 16 to its database. Bridge 16, when it captures a new media access control box on network 10, searches that database for the destination address contained in that box. The broadcast of the box across Bridge 16 is based on the results of that search. As such, the simple algorithm does not require any effort or knowledge by the terminal stations 21a, 25a, 25b to switch frames across the bridges 16, so that the operation of the bridges is transparent. If the algorithm is similar to the method

ES 2 201 073 T3 of "learning on the fly" as described above, it may be possible that, until steady state operation is achieved, bridge 16 cannot properly pass all frames, and so that possibility must be allowed. into account in the present invention. The present invention is preferably implemented with bridges 16 operating transparently for terminal stations 21a, 25a and 25b.

The knowledge that the network monitors 20, 24a, and 24b are within a transparently bridged group is preferable and aids in the operation of the automated segment determination of the present invention. Each of those network monitors 20, 24a, 24b is called a transparent partner and each network monitor 20, 24a and 24b preferably maintains a current list of its transparent partners. A single bit can indicate that a particular network monitor is within a transparently bridged group.

To detect other network monitors, a bridged group poll message, shown as control block 152 in Figure 4K, may be sent periodically. Control block 152 includes a command code field 154 and a request / response indicator field 156. Any other network monitor 20, 24a, and 24b that sees the bridged group poll message control block 152 will respond with a similar control block 152 with the response bit set in field 156. To maintain current status, the bridged group poll message is preferably sent periodically. To prevent disturbing other stations on the network, the destination address of that packet can be, for example, the universally administered address (UAA: universally administered address) of an Ethernet card. The Ethertype of the packet will be a unique non-routable value assigned to the implementing device for this purpose.

A host computer-based network monitor (not shown), in communication with the local area network 10, will preferably receive a periodically report of transparently bridged groups. In Figure 4L, control block 158 shows an example of such a reporting mechanism comprising an order code field 160, membership number field 161, and network monitor field 162. As such, the host computer-based network monitor (not shown) can maintain a current status of the local network monitors 20, 24a, and 24b in the joined local area network 10.

Referring to Figure 2, a functional block diagram 30 of an automatic segment determination program is shown that forms a portion of the program that operates or resides on the first network monitor 20, the second network monitor 24a, and the m -simo monitor 24b network. Functional scheme 30 is subdivided into three parts. The first portion 31 is the function performed if one of the network monitors 20, 24a, 24b is serving as the initiator of the request to determine the position of a station 21a, 25a, 25b. The second portion 32 comprises the functions performed if the network monitor 20 is functioning as the primary determining device. The third portion 46 comprises the functions performed when the network monitor 20 is not the primary determining device but is part of the transparently bridged group.

The processing flow of an implementation of the automated segment determination program will now be described with reference to the flow chart illustrated in Figure 3 and with further reference to the functional block diagram 30 in Figure 2 and the control blocks found in Figures 4A. to 4L. For the purposes of this description, the network monitor 20 will be the primary determining device and the other network monitors 24a, 24b will be part of the transparently bridged group.

The processing flow begins when a monitor 20 is requested to determine ownership of a particular station. That request can come from a host-based network monitor (not shown). Such a request may have the form shown in Figure 4E at control block 110 comprising a request code arranged to determine the station segment in field 112, a primary determination request indicated in field 114, and the station address. in field 116 representing the particular station to be determined. Alternatively, that request may be internally generated by primary determining device network monitor 20 or sent from another network monitor 24a or 24b, functionally shown as block 31 in Figure 2.

Referring to Figure 3, the first network monitor 20 will request the primary determination in step 61. The primary determining device network monitor 20 will call the "generate and send a primary probe" function 34 to generate and send a polling message to the station to be determined. In Figure 4G, control block 122 can be used to perform function 34. Control block 122 comprises a request code field 124 set equal to the station segment poll and a poll type field 126 set primary.

Once the primary determining device network monitor 20 sends the polls in step 61, the primary determining device network monitor 20 waits for a predetermined period of time, one minute for example, for responses due to that poll message. . Responses will be generated by the function identified as block 47 in Figure 2 and may take the form of control block 128 in Figure 4H. The control block 128, which represents a primary observation report, comprises a request code 130 indicating that the survey was observed, a designation of the survey type 132 set to primary, and the node or address field 134 of the station whose ownership is being determined. Only those network monitors 24a, 24b that actually read the primary survey in its segments 14a, 14b, respectively, will respond with a primary observation report. If the polling message represented by the control block 122 does not appear on a local segment 14a, 14b, that is, the control block 122 was not transmitted by the joined bridge 16, the network monitors 24a, 24b do not see the block 122 control and thus cannot respond with a primary observation report.

Referring again to Figure 3, at step 62, a decision block as to whether any observations from the primary probe were received is found by the primary determining device, in this case, the network monitor 20. Network monitor 20 contains a function 36, shown in Figure 2, for receiving reports of primary observations. Yes

ES 2 201 073 T3 the primary observation reports represented by the control block 128 were received by the primary determining device network monitor 20, processing continues at step 66. At step 66, the device network monitor 20 primary determinant generates and sends secondary probe requests, functionally illustrated at block 38 in Figure 2. Secondary probe requests may take the form of the control block 110 shown in Figure 4E, and described above, with a designation that the target station is a secondary determining device in field 114. The control block 110 is sent to each of the network monitors 24a, 24b that transmitted reports of primary observations, via the control block 128, to the network monitor 20 of the primary determining device, such network monitors being called devices. secondary determinants.

Upon receiving the secondary probe requests, the secondary determining devices send a secondary probe that can be represented, as shown in Figure 4G, by the control block 122 with the field 126 set to secondary. The generation of a secondary probe is represented by the function indicated in block 48 of Figure 2.

Similar to the primary determining device network monitor 20, each of the secondary determining devices will wait for a predetermined period of time, one minute for example, to receive reports of secondary observations. Such secondary observation reports may take the form of control block 128 as described above, with probe type field 132 set to secondary, and will be sent to any network monitor, including primary determining device network monitor 20. , which reads the message on its local segment. Function block 40 in FIG. 2 illustrates the "observe secondary soundings and generate secondary observation reports" function.

Each secondary determining device will create a summary of the secondary observation reports that it receives and transmits that summary to the primary determining device network monitor 20. Such a summary of secondary poll results, functionally represented by block 49 in Figure 2, may be transmitted to primary determining device network monitor 20 in the form of control block 136 shown in Figure 4I. The control block 136 comprises a request code field 138 and a poll result field 140 which together indicate that the control block 136 contains the summary of the secondary observation reports. The control block 136 also includes a field 142 that indicates the number of observers for the secondary poll, and an address field 144 that contains the address of the station that received the secondary poll.

In Figure 2, the function block 42 illustrates the function of the primary determining device that receives the summaries in the form of control block 136 from the secondary determining devices. The function block 44 illustrates the determining function as performed by the primary determining device network monitor 20. In step 68 of Figure 3, the primary determining device network monitor 20 makes a decision as to whether one of the secondary determining devices did not receive secondary observation reports. If one and only one of the secondary determiners does not report secondary observations in response to its secondary probe, then that secondary determiner is the owner of the station that was surveyed. In step 70, an indication of station ownership is sent to that secondary determining device which may be in the form of control block 146 in Figure 4J comprising station owner field 148 and an address field 150 containing the address. station that was polled.

If in step 68 it is determined that none of the secondary determining devices reported secondary observations or that more than one secondary determining device did not report secondary observations, then the primary determining device network monitor 20 is unable to determine station property as shown. sample in step 78. The inability to determine station ownership is then transmitted to the host network monitor (not shown) or to the initiator of the determination request from the primary determining device network monitor 20 whose message may, for example, be represented by control block 118 in Figure 4F. Control block 118 contains a request code field 119 which means that the primary determining device cannot determine the station segment, and an address field 120 which indicates the address of the station whose ownership cannot be determined.

Referring again to Figure 3, if at decision block 62 the primary determining device network monitor 20 determines that the primary determining device network monitor 20 did not receive any primary poll observations in response to the primary poll, processing continues. in decision block 64. At decision block 64 it is determined whether the primary determining device network monitor 20 is part of a transparently bridged group. That information is known to the primary determining device network monitor 20 as it preferably maintains a list of its transparent partners as described above. If in decision block 64 the primary determining device network monitor 20 determines that it is not part of a transparently bridged group, that is, it has no transparent partners, then the primary determining device network monitor 20 determines that the station to be determined is on its local segment 12. Then, the primary determining device network monitor 20 reports that the station is operational, in step 80, to the initiator of the determination request.

If at decision block 64 it is determined that the primary determining device network monitor 20 is part of a transparently bridged group, that is, has transparent partners, then processing continues at step 72. In step 72, the primary determining device selects one of its transparent partners and then generates and sends a secondary probe request to the selected transparent partner, such secondary probe request being represented by the control block 110 in Figure 4E as described. before.

The selected transparent partner thus becomes the secondary determining device and proceeds to en5

ES 2 201 073 T3 send a secondary probe to the station as outlined above. Each of the other network monitors, including the primary determining device network monitor 20, will perform the function 40 of observing secondary probes and generating secondary observation reports. The secondary determining device will compile the secondary observation reports and transmit those results to the primary determining device network monitor 20 using, for example, the control block 136 in Figure 4I as described above.

If the control block 136 that is sent to the primary determining device network monitor 20 indicates that no reports of secondary observations were received by the secondary determining device, processing continues from decision block 74 to step 78 since the monitoring device Primary determining device network is unable to determine station ownership. That result is transmitted to the host computer-based network monitor (not shown) or to the initiator of the determination request using, for example, control block 118 in Figure 4F as described above.

If the control block 136 that is sent to the primary determining device network monitor 20 indicates that secondary observation reports were received by the secondary determining device, processing continues from decision block 74 to decision block 76. At decision block 76, the primary determining device network monitor 20 checks to see if the primary determining device network monitor 20 generated a secondary watch report. If the primary determining device network monitor 20 generated a secondary watch report, then the primary determining device network monitor 20 is the owner of the station whose ownership is being determined. That determination is communicated to the host computer-based network monitor (not shown) or the initiator of the determination request in step 80.

If in decision block 76 it is determined that the primary determining device network monitor 20 did not generate a secondary watch report, then the primary determining device network monitor 20 is unable to determine station property as shown in the step 78. It will be understood that any network monitor 20, 24a, 24b may serve as the primary determining device network monitor. Each of those network monitors 20, 24a, 24b is preferably in communication with a host computer-based network monitor (not shown) that initiates determination requests and receives responses to them. Such host-based network monitor communications are secondary to the present invention and are described herein for completeness.

Figures 4A through 4D illustrate control blocks 82, 92, 100, and 102, respectively, that can be used for communications between the network monitor 20, 24a, 24b and the host computer-based network monitor. In operation, the host computer-based network monitor may request from the network monitors 20, 24a, and 24b a list of currently unassigned stations (control block 100) and the network monitors 20, 24a, and 24b may communicate such stations. not assigned using control block 102. The control block 82 and the control block can be sent from the host computer-based network monitor to the network monitors 20, 24a, and 24b to update the station property tables on each network monitor 20, 24a, 24b. . The method of operation of the present invention is illustrated by two examples, shown in Figures 5 and 6, for a network configuration similar to that shown in Figure 1. Figure 5 is an example of station segment determination in which the station whose ownership is to be determined does not reside in the local segment of the primary determining device. Figure 6 is an example of station segment determination in which the station whose ownership is to be determined resides in the local segment of the primary determining device.

Referring to Figure 5, the message traffic between a host-based network monitor program 172 is shown, a plurality of network monitor programs forming a transparently bridged monitor system represented by the numerals 174, 176 and 178, and a station 180 whose ownership is to be determined. The message traffic in Figure 5 is shown in chronological order with earlier messages at the top and later messages at the bottom. A first segment monitor 174 communicates to host computer-based network monitor program 172 that station 180 is unassigned. In response thereto, the host computer-based network monitor program 172 requests that the first segment network monitor 174 make the determination for that station 180. The first segment network monitor sends a primary probe to station 180. The second and third segment network monitors 176, 178, respectively, report that the primary probe was observed. Then, the first segment network monitor 174 sends a secondary poll request to the second and third segment network monitors 176, 178, respectively. Second segment network monitor 176 sends a secondary poll to station 180 and receives a secondary observation report from third segment network monitor 178 in response thereto. Third segment network monitor 178 sends a secondary poll to station 180 and does not receive a secondary observation report in response to it. Second segment network monitor 176 communicates to first segment network monitor 174 that a secondary observation report was received. Third segment network monitor 178 communicates to first segment network monitor 174 that no secondary observation report was received. First segment network monitor 174, after determining that station 180 is owned by third segment network monitor 178, then sends a station ownership indication to third segment network monitor 178, followed by a station report. operational from the third segment network monitor 178 to the host computer based network monitor program 172.

Referring to Figure 6, message traffic is shown between a host-based network monitor program 182, a plurality of network monitor programs forming a transparently bridged monitor system represented by the numerals 184, 186 and 188, and a station 190 whose ownership is to be determined. So

ES 2 201 073 T3 Similar to the previous description, the message traffic of Figure 6 is shown in chronological order with the earlier messages at the top and the later messages at the bottom. A first segment network monitor 182 communicates to host computer-based network monitor program 182 that station 190 is not assigned. In response thereto, host-based network monitor program 182 requests that first-segment network monitor 182 make the determination for that station 190. First-segment network monitor 182 sends a primary poll to station 190 None of the network monitors 186, 188 in the second and third segments, respectively, report that the primary survey was observed. The first segment network monitor 182 then selects the third segment network monitor 188 and sends it a secondary poll request. Third segment network monitor 188 sends a secondary poll to station 190 and receives a secondary watch report from both network monitors 182, 186 in the first and second segments, respectively. Third segment network monitor 188 communicates to first segment network monitor 182 that two secondary observation reports were received, including one from first segment network monitor 182. The first segment network monitor 182, after determining that the station 190 resides in the first segment, then sends an operating station report to the host computer-based network monitor program 182.

Although the present invention has been described in connection with an exemplary embodiment thereof, it will be understood that many modifications and variations will be readily apparent to those of ordinary skill in the art. This description and the claims that follow are intended to cover all such modifications and variations.