JP3892282B2 - Network configuration method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a network configuration method for dynamically dividing and reconfiguring (joining) a network.
[0002]
[Prior art]
Conventionally, in a bus-type network in which a certain data transfer band is communicated while transferring among a plurality of devices (nodes) is shared, the number of connected devices increases in principle. As the transfer bandwidth decreases and becomes insufficient, as shown in FIGS. 7 (a) and 7 (b), devices with high communication frequency belong to the same group (hereinafter also referred to as sub-network). A method of effectively utilizing a limited bandwidth by dividing a network including the devices 21 to 25 into a plurality of sub-networks and connecting the sub-networks with a relay device (hereinafter also referred to as a bridge) 26 is known. ing.
[0003]
Here, as the bus type network, for example, a serial bus (hereinafter referred to as IEEE1394 bus) defined by IEEE Std 1394 (references 1 and 2) by the American Institute of Electrical and Electronics Engineers (hereinafter abbreviated as IEEE). In order to connect a plurality of IEEE1394 buses and communicate with each other, a bridge standard (Reference 3) of P1394.1 is under development and standardization work.
[0004]
Also, as a supplement to the problem in P1394.1, there are many devices (nodes) that improve the decrease in bus utilization efficiency during network configuration (bus reset) (Japanese Patent Laid-Open No. 11-220485) Improve the problem when resources (bandwidth) are acquired (Japanese Patent Laid-Open No. 11-220485), improve the selection method of bridge manager (Japanese Patent Laid-Open No. 2000-165417), portal Several proposals have been made, such as improving the reliability of routing map settings (Japanese Patent Laid-Open No. 2000-165417) and providing a management method corresponding to bus disconnection (Japanese Patent Laid-Open No. 2000-349805). ing. The document 1 is “IEEE Std 1394-1995, Standard for a High Performance Serial Bus”, and the document 2 is described above. , “IEEE Std 1394a-2000, Standard for a High Performance Serial Bus (Amendment)”, reference 3 , “P139.1, Draft Standard for a High Performance Serial Bus Bridges”.
[0005]
However, in such a conventional network configuration method, since the network is divided into a plurality of sub-networks by the bridge, there is a cost for developing the bridge itself (in order to achieve a practical throughput in the bridge, In order to operate the entire network autonomously and efficiently, the hardware of the existing equipment (nodes) must be matched to the environment connected by the bridge. Therefore, it is necessary to deal with both sides of the software, which is extremely disadvantageous in terms of cost. At least the present IEEE 1394 bus has not realized a device having these bridge functions. In particular, in a small-scale network (generally composed of about 5 nodes), it is not practical from the viewpoint of cost to introduce an expensive bridge in order to avoid a rarely congested state.
[0006]
As a countermeasure, when the network topology is such that the route for the desired communication is not divided due to the division of the sub-network, use a relatively expensive bridge for the division into the sub-network described above. It is desirable to do it dynamically only when division is necessary.
[0007]
[Problems to be solved by the invention]
However, in the network configuration method that dynamically divides into sub-networks as needed as described above without using bridges, as shown in FIG. 8, traffic that causes congestion (communication between nodes 22 and 23) ) And the desired communication (communication between node 21 and node 25) overlap (overlapping), the division of the sub-network will also divide the route for the desired communication Therefore, there is a problem that both communication cannot be normally performed.
[0008]
The present invention has been made to solve such a problem, and provides a network configuration method capable of effectively using bandwidth by dynamically dividing / joining networks. In particular, the present invention dynamically disconnects the specific two nodes from the network and normally performs communication on the network when communication between the specific two nodes is assumed to cause a congestion state in advance. The present invention provides a network configuration method capable of performing the above.
[0009]
[Means for Solving the Problems]
First to solve the above problem The present invention is a bus-type network configuration method provided with a plurality of nodes having a repeater function for amplifying and transmitting a signal, the first node in the network being connected to the first node When the network is divided by dynamically separating the second node from the network, the first node and A third node connected to the first node; No The peater function is invalidated.
[0010]
By this method, a repeater function of amplifying and transmitting a signal is added to all devices (nodes) connected to the network, and all devices behave as repeaters so that communication between devices becomes possible. In such a bus-type network, a single bus-type network can be divided into a plurality of bus-type networks by performing control so that the repeater function is appropriately disabled.
[0011]
Also, Second to solve the above problem The present invention is a method for configuring a bus-type network provided with a plurality of nodes having a repeater function for amplifying and transmitting a signal, and comprising: a first node in the network; Above A second node connected to the first node When Other nodes except 2 of them A bypass signal transmission path connecting the first node and the second node dynamically from the network to divide the network, Of the connection ports of the first node, a connection port connected to a third node in the network and a connection port connected to the first node of connection ports of the third node The repeater function of the second node is disabled, and the second node among the connection ports connected to the fourth node in the network among the connection ports included in the second node and the connection port included in the fourth node Of the connection port connected to Disable the repeater function, and Each end By enabling the repeater function of the connection port, the signal transmission path can be used.
[0012]
By this method, a signal transmission path (alternative communication path) that bypasses the secondary network to be divided in advance is prepared, and the secondary communication is separated when the secondary network is separated due to an increase in traffic between two specific nodes. By exclusively selecting and using a route, a single bus type network is divided into a plurality while performing desired communication, and only while a specific node performs communication requiring a wide band, Since two specific nodes can be dynamically disconnected from the network, the problem that communication between other nodes becomes difficult can be solved.
[0013]
Also, Third to solve the above problem The invention The first or second invention The bus type network is configured by a serial bus defined by IEEE Std 1394.
[0014]
By this method, it is possible to dynamically disconnect a specific two node from a network and perform normal network communication without using a special device such as a bridge (relay device). Since the operation can be executed by software without the intervention of the end user, a network environment with high cost performance and good operability can be realized.
[0015]
Also, Fourth to solve the above problem The invention The third invention The software for realizing the port disable / enable function defined in IEEE Std 1394a-2000 is used to enable or disable the repeater function.
[0016]
By this method, since a complicated control protocol for cooperating with a bridge or the like is not used, reliability and compatibility with existing devices are increased.
[0017]
Also, The fifth to solve the above problem The invention The fourth invention In order to realize the disable / enable function of the port, a remote command packet defined in IEEE Std 1394a-2000 is used.
[0018]
By this method, the repeater function of this device can be validated or invalidated from a device (node) other than a device (node) having a port to be controlled by the repeater function.
[0019]
further, Sixth to solve the above problem The invention Of the second to fifth inventions In any one of the above, the node includes a plurality of nodes having the repeater function.
[0020]
By this method, a plurality of repeater function blocks (nodes) are built in one device, so that the number of apparent ports can be reduced and a plurality of signal transmission paths can be easily set.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[First Embodiment]
As shown in FIG. 1, the network configuration method according to the first embodiment of the present invention is a bus-type network configuration method in which a plurality of nodes 1 to 6 having a repeater function for amplifying and transmitting a signal are provided. When the node 5 in the network and the node 6 connected to the node 5 are dynamically separated from the network to divide the network , No 6 and Node 3 connected to node 6 No The peater function is invalidated.
[0022]
In this embodiment, as shown in FIG. 1A, the topology of a bus-type network (for example, an IEEE 1394 bus) is divided into sub-networks including a specific node (device) 5 and a node (device) 6. Shows a network configuration method in the case where the communication path between the other nodes (devices) 1 to 4 is not divided. Each device on the bus-type network has a connection port (port) with one or a plurality of other devices, and forms a network by connecting the port and the port with a cable. Each device has a function (so-called repeater function) for reproducing and relaying a signal (information) received from one of the devices and sending it to another device. With such a repeater function, the network shown in FIG. 1 functions as a bus-type network in which all devices 1 to 6 share a certain communication band. In addition, a device serving as a transmission source of information adds information (for example, an identifier (ID) or an address) for identifying a reception partner to a transmission signal (information) and transmits the transmission signal on the network. Each device that receives such a signal determines whether it is a signal (information) addressed to itself or not based on the identification information, and receives it as necessary. Thereby, communication between arbitrary two apparatuses is performed.
[0023]
Here, it is assumed that the specific node 5 and the node 6 perform communication that requires a very large bandwidth. When starting such communication, the node 6 is set so as not to operate the repeater function of the port 6a connected to the node 3, and the node 3 is also connected to the node 6 in response to this. The repeater function 3a is set not to operate. As a result, the network is divided into a main network to which the nodes 1 to 4 belong and a sub network to which the nodes 5 and 6 belong, as shown in FIG. Thereafter, no matter how much bandwidth is required to communicate between the node 5 and the node 6, the communication of the nodes 1 to 4 belonging to the main network is not affected. In this state, although the main network and the sub network are connected by a cable (indicated by a dotted line in the figure), the state is functionally the same as when the cable is disconnected. That is, the main network and the sub network function as two completely separate networks. After the communication with the node 5 is completed, the node 6 again restores the repeater function of the port 6a connected to the node 3, and similarly the node 3 also sets the repeater function of the port 3a connected to the node 6 to the valid state. By returning, the initial state shown in FIG.
[0024]
Note that in a state where communication is performed using the network bandwidth at the nodes 5 and 6, it is difficult for the nodes 5 and 6 to communicate with the other nodes 1 to 4. Even if 3 and node 6 are connected by a bridge, this situation does not change. Therefore, it can be said that there is no great difference in terms of functionality even if the network is divided or connected using a bridge.
[0025]
Here, with reference to FIG. 2 and FIG. 3, the invalidation and validation operation of the above-described repeater function of the node 6 in the network using the IEEE 1394 bus will be described. FIG. 2 shows a register configuration of the PHY (physical layer) controller. Here, FIG. 2A is obtained by quoting FIG. 7-1 of the above-mentioned reference 2, and FIG. 2B is obtained by quoting FIG. 7-2 of the above-mentioned reference 2. FIG. 3 is a quotation from FIG. 8-8 of the reference 2.
[0026]
In the network using the IEEE 1394 bus, the port repeater function can be disabled (disabled) / enabled (enabled) by software by operating a register of a PHY (physical layer) controller. Here, when a port (for example, 3a and 6a shown in FIG. 1) is operated, first, 0 is written in the Page_Select field of the register 6 of the PHY controller, and the target port number is written in the Port_Select field. Thereafter, by setting 1 in the Disabled field of the register 7, the port is disabled, and by clearing it to 0, the port is enabled.
[0027]
Further, in the IEEE 1394 bus, by using a remote command packet that is a kind of PHY (physical layer) packet, it is possible to disable / disable the above ports 3a and 6a from devices other than the device (node) having the target port. Enable operations can be performed. When such a remote command packet is used, when the node 5 in FIG. 1 is the main node and starts communication with the node 6, the node 5 sends the remote command packet to the node 6 to 6 can be disabled / enabled. That is, the network can be divided and reconfigured (joined) from a node other than the node having the port that is the target of the repeater function validation and invalidation operations.
[0028]
Specifically, when a disable operation is performed on the target ports (for example, 3a and 6a shown in FIG. 1), the identifier (physical ID) of the node having the operation target port in the phy_ID field shown in FIG. A port number to be operated is set in the port field, and 1 is set in the cmnd field, and a remote command packet is transmitted. When performing an enable operation on the target port, the identifier (physical ID) of the node having the operation target port in the phy_ID field shown in FIG. 3, the port number of the operation target in the port field, and the cmnd field 5 is set and a remote command packet is transmitted.
[0029]
As described above, the network configuration method according to the first embodiment of the present invention is a bus-type network configuration method provided with a plurality of nodes 1 to 6 having a repeater function for amplifying and transmitting a signal. When the node 5 in the network and the node 6 connected to the node 5 are dynamically separated from the network to divide the network , No 6 and Node 3 connected to node 6 No Since the Peter function is disabled, when the bus type network is divided into a plurality of networks including two specific nodes (devices), the communication path between the other nodes is not divided. One node can be dynamically disconnected from the network and the network can communicate normally.
[0030]
[Second Embodiment]
FIG. 4 shows a network configuration method according to the second embodiment of the present invention. The bus-type network of this embodiment is configured using an IEEE 1394 bus according to the first embodiment (shown in FIGS. 2 and 3), and has a repeat function at each port of the nodes 1-7. The enable / disable control uses the disable / enable function for ports specified in IEEE Std 1394a-2000, and the disable / enable function for target ports from a node other than the node that performs communication. Since a remote command packet is used when setting, a description on this is omitted.
[0031]
This embodiment is different from the first embodiment in the alternative communication connecting the node 4 in the network and any of the other nodes 1, 2, 3, 5, 6 except the node 7 connected to the node 4 When the path 30 is provided and the node 4 and the node 7 are dynamically separated from the network to divide the network, Node 4, node 3 connected to node 4, node 7, and node 6 connected to node 7 respectively have Connection 4b, 3b, 7b and 6b The repeater function is disabled, the repeater functions of the connection ports 3a and 6a of the nodes 3 and 6 connected to both ends of the alternative communication path 30 are enabled, and the alternative communication path 30 can be used. According to this method, when the topology of the bus-type network is such that the communication path between other nodes is divided by dividing the sub-network including two specific nodes (devices), It is also possible to dynamically disconnect two specific nodes from the network and to perform normal network communication.
[0032]
In FIG. 4, devices (nodes) 1 to 7 on a bus-type network (for example, using an IEEE 1394 bus) are each provided with a connection port (port) with one or more other devices, A network is formed by connecting ports to each other by a cable. Each of the devices 1 to 7 has a function (so-called repeater function) for reproducing and relaying a signal (information) received from any of the devices and sending it to another device. With such a repeater function, the network shown in FIG. 4 functions as a bus-type network in which all devices 1 to 7 share a certain communication band. In addition, a device serving as a transmission source of information adds information (for example, an identifier (ID) or an address) for identifying a reception partner to a transmission signal (information) and transmits the transmission signal on the network. Each device that receives such a signal determines whether it is a signal (information) addressed to itself or not based on the identification information, and receives it as necessary. Thereby, communication between arbitrary two apparatuses is performed.
[0033]
Here, if it is preliminarily assumed that the node 4 and the node 7 perform communication that requires a very large bandwidth, the node 3 and the node 7 are empty so as to bypass the nodes 4 and 7. The port 3 a and the empty port 6 a of the node 6 are connected and used as the alternative communication path 30. In the initial state, the disabled state is set so that the repeater functions of the both-end ports 3a and 6a of the alternative communication path 30 are not enabled. This is because the IEEE 1394 bus does not allow a connection form that forms a loop in the bus.
[0034]
In such a configuration, when communication that requires a very large bandwidth is started between the node 4 and the node 7 and communication is also performed between the node 1 and the node 5 at the same time, as shown in FIG. The node 4 invalidates the repeater function of the port 4b connected to the node 3, and the node 3 also invalidates the repeater function of the port 3b connected to the node 4. In addition, the node 3 enables the repeater function of the port 3a connected to the node 6 in order to use the alternative communication path 30 when the communication with the node 4 is cut off. Similarly, the node 7 invalidates the repeater function of the port 7b connected to the node 6, and the node 6 invalidates the repeater function of the port 6b connected to the node 7 and receives an alternative communication path. 30 is used, the repeater function of the port 6a connected to the node 3 is enabled.
[0035]
By setting the repeater function as described above, the node 4 and the node 7 can be divided as local sub-networks without interrupting the communication between the node 1 and the node 5, and the bandwidth between the node 4 and the node 7 You can perform transfers that use as much as possible. In this state, although the main network and the sub network are connected by a cable, they are functionally the same as when the cable is disconnected. The main network and the sub network are completely different from each other. Since it functions as a network, no matter how large bandwidth communication is performed between the node 4 and the node 7, the communication between the node 1 and the node 5 belonging to the main network is not affected.
[0036]
Thereafter, when the communication with the node 7 is completed in the node 4, the repeater function of the port 4b connected to the node 3 is made valid again. As a result, the node 3 first disables the repeater function of the port 3 a connected to the node 6, and subsequently enables the repeater function of the port 3 b connected to the node 4. Similarly, the node 7 enables the repeater function of the port 7b connected to the node 6. As a result, the node 6 first disables the repeater function of the port 6 a connected to the node 3, and subsequently enables the repeater function of the port 6 b connected to the node 7. By such setting of the repeater function, the state shown in FIG.
[0037]
[Third Embodiment]
FIG. 5 shows a network configuration method according to the third embodiment of the present invention. The bus-type network of this embodiment is configured using an IEEE 1394 bus according to the first embodiment (shown in FIGS. 2 and 3), and has a repeat function at each port of the nodes 1-7. The enable / disable control uses the disable / enable function for ports specified in IEEE Std 1394a-2000, and the disable / enable function for target ports from a node other than the node that performs communication. Since a remote command packet is used when setting, a description on this is omitted.
[0038]
This embodiment is different from the second embodiment in that an alternative communication path 30 used between the node 4 and the node 7 for performing communication using the entire band is set between the node 1 and the node 5. Is different.
[0039]
In FIG. 5, the node 4 invalidates the repeater function of the port 4 c connected to the node 3, and in response to this, the node 3 also invalidates the repeater function of the port 3 c connected to the node 4. Further, in the node 3, the connection with the node 4 is cut off, so that the alternative communication path 30 is used, so that the repeater of the port 1 c connected to the node 5 in the node 1 using the remote command packet described above. Enable the function. Similarly, the node 7 invalidates the repeater function of the port 7c connected to the node 6, and in response to this, the node 6 also invalidates the repeater function of the port 6c connected to the node 7, and the alternative communication path. 30 is used, the repeater function of the port 5c connected to the node 1 is enabled at the node 5 by using the remote command packet described above.
[0040]
Thereafter, when the communication with the node 7 is completed in the node 4, the repeater function of the port 4c connected to the node 3 is made valid again. As a result, the node 3 first disables the repeater function of the port 1c connected to the node 5 in the node 1, and subsequently enables the repeater function of the port 3c connected to the node 4 in the own node 3. Similarly, at node 7, The repeater function of the port 7c connected to the node 6 is made valid. As a result, at node 6, Disable the repeater function of port 5c connected to node 1 at node 5 6 enables the repeater function of the port 6 c connected to the node 7. As a result, the state shown in FIG.
[0041]
[Fourth Embodiment]
FIG. 6 shows a network configuration method according to the fourth embodiment of the present invention. The bus-type network of this embodiment is configured using an IEEE 1394 bus according to the first embodiment (shown in FIGS. 2 and 3), and has a repeat function at each port of the nodes 1-7. The enable / disable control uses the disable / enable function for ports specified in IEEE Std 1394a-2000, and the disable / enable function for target ports from a node other than the node that performs communication. Since a remote command packet is used when setting, a description thereof will be omitted.
[0042]
This embodiment is different from the second embodiment in that nodes (devices) 10 and 20 include a plurality of nodes 3, 4, 6, and 7 having a repeater function. According to this method, it is possible to obtain an effect of facilitating connection and connection by reducing the number of apparent ports.
[0043]
In FIG. 6, the devices 10 and 20 communicate with each other using a wider bandwidth than the other devices 1, 2, and 5. In addition to the node (device) 4 connected to the original functional block of the device, the device 10 includes a node (device) 3 that is a repeater function block. In addition to the node (device) 7 connected to the original functional block of the device, the device 20 includes a node (device) 6 that is a repeater functional block. Here, if each of the devices 10 and 20 is regarded as one node, the connection of the nodes 1, 2, 5, 10, and 20 forming the bus-type network is in accordance with the first to third embodiments. I understand. Here, since each of the nodes 1 to 7 is substantially the same as that of the third embodiment, the description thereof is omitted.
[0044]
As described above, the device 10 has a built-in repeater function block (device 3), so that it apparently has a port A for cable connection with the other device 2 and a cable connection with the device 20 (node 7). A port B and a port b for connecting the alternative communication path 30 between the device 20 (that is, a port for connecting the device 3 and the device 6 with a cable) b are provided. In addition, the device 20 has a repeater function block (device 6) built in, so that it apparently has a port A 'for cable connection with another device 5 and a cable connection with the device 10 (node 4). A port B ′ and a port b ′ for connecting an alternative communication path 30 between the device 10 (that is, a port for connecting the device 6 and the device 3 with a cable) b ′ are provided. Here, by bundling the cables between the ports BB ′ and bb ′ into one, the devices 10 and 20 can be easily connected without complicated connection.
[0045]
【The invention's effect】
Book According to the invention, by dividing and reconfiguring (joining) the network, it is possible to configure a network in which an increase in local traffic does not affect communication between other devices.
[0046]
Book According to the present invention, even when the network is divided, communication between other devices is possible by dynamically using the alternative communication path. In addition, since a complicated control protocol for cooperating with a bridge or the like is not used, the reliability is high and the compatibility with existing devices is also high.
[0047]
Book According to the invention, when an IEEE 1394 bus is used, the network can be divided and reconfigured without introducing a special device such as a bridge, and all operations are performed without software intervention by an end user. Is feasible. All can be applied to the current IEEE 1394 standard and existing hardware. Therefore, it is possible to improve the cost performance and realize an easy-to-use network environment.
[0048]
Book According to the invention, as described above, when the IEEE 1394 bus is used, it is possible to improve the cost performance and realize an easy-to-use network environment.
[0049]
Book According to the invention, as described above, when the IEEE 1394 bus is used, it is possible to improve the cost performance and realize an easy-to-use network environment.
[0050]
Book According to the invention, it is possible to reduce the number of connectors and cables to be connected by incorporating an additional repeater function in the device, and it is possible to provide a network environment that can be easily set (wired) for the user.
[0051]
As described above, according to the present invention, it is possible to provide a network configuration method having an excellent effect that the network can be dynamically divided and reconfigured to effectively use the bandwidth.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a network configuration method according to a first embodiment of the present invention.
FIG. 2 is a diagram illustrating network division and joining operations according to the first embodiment of the present invention.
FIG. 3 is a diagram showing a format of a remote command packet used for network division and joining operations according to the first embodiment of the present invention.
FIG. 4 is a diagram illustrating a network configuration method according to a second embodiment of the present invention.
FIG. 5 is a diagram illustrating a network configuration method according to a third embodiment of the present invention.
FIG. 6 is a diagram illustrating a network configuration method according to a fourth embodiment of the present invention.
FIG. 7 is a diagram for explaining a conventional network configuration method (division method using a bridge).
FIG. 8 is a diagram illustrating a network configuration method that does not depend on a bridge;
[Explanation of symbols]
1-6, 21-25 nodes (equipment)
26 Bridge (relay device)
3a-3c, 4b, 4c, 5c, 6a-6c, 7b, 7c, A, A ', B, B', b, b 'ports
30 Alternative communication path (signal transmission path)

Claims (5)

A method of configuring a bus-type network provided with a plurality of nodes having a repeater function for amplifying and transmitting a signal,
A signal transmission path for bypass connecting the two of the first node and the second node and the other nodes except that is connected to the first node in the network is provided, wherein the first node And a connection port connected to a third node in the network among connection ports of the first node when the network is divided by dynamically separating the second node from the network; and Of the connection ports of the third node, the repeater function of the connection port connected to the first node is disabled, and the connection node of the second node has a fourth node in the network. disable the connected repeater connection ports connected to the second node of the connection port and connection ports said fourth node has, the signal transduction pathway both Enable the repeater functions of the connection ports, the network configuration method which is characterized in that available the signal transduction pathway. 2. The network configuration method according to claim 1, wherein the bus type network is configured by a serial bus defined by IEEE Std 1394. 3. The network configuration method according to claim 2 , wherein software for realizing a disable / enable function of a port defined in IEEE Std 1394a-2000 is used when the repeater function is enabled or disabled. . 4. The network configuration method according to claim 3 , wherein a remote command packet defined in the IEEE Std 1394a-2000 is used to realize the disable / enable function of the port. Wherein any of the plurality of nodes, the network configuration method according to any one of claims 1 to 4, characterized in that a plurality internal nodes having the repeater function.

Images