JP3606064B2 - Network bandwidth management method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a network system in which a system that requires bandwidth management control and a system that does not require bandwidth management control coexist on the same network.
[0002]
[Prior art]
RSVP will be described as a typical protocol for controlling the bandwidth.
[0003]
In RSVP, first, a node that transmits data sets a route to a receiving node, and then the receiving node sends reservation information including delay time and priority information through a reverse route. This is a protocol for reserving the bandwidth of each router interposed between nodes.
[0004]
When using the above bandwidth reservation protocol in a network with multiple nodes on one network, the bandwidth reservation protocol is implemented on all nodes so that a node does not transmit data that has not been reserved for bandwidth. It is necessary to keep the application to be running.
[0005]
[Problems to be solved by the invention]
As described above, in the case of the bandwidth control method using the bandwidth reservation protocol as described above, an application that implements the bandwidth control protocol must be installed and operated in all the nodes. Therefore, a system that requires bandwidth control. In order for systems that do not require bandwidth control to coexist on a single network, there is a problem that an application that implements the bandwidth control protocol must be installed on all nodes of the system that do not require bandwidth control. Occur.
[0006]
An object of the present invention is to realize coexistence of a system that requires bandwidth control and a system that does not require bandwidth control without changing the nodes of the system that does not require bandwidth control on the same network. It is in.
[0007]
[Means for Solving the Problems]
The bandwidth management target node receives a control message for controlling the transfer rate of the own node from another node connected to the network, and controls the transfer rate of the transmission data of the own node, and a part of the communication contents of the network or There are means for intercepting all of them, and means for performing priority transmission control capable of transmitting messages with higher priority than NICs of other nodes by means of a network interface card (NIC) mounted on the node.
[0008]
In the bandwidth management manager node, a message connected to the network for intercepting part or all of the communication content of the network and a message for controlling the transfer rate to the node having the function of controlling the transfer rate of the other node. Means for transmitting.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Before describing the embodiment of the present invention, the Ethernet targeted in this embodiment will be described.
[0010]
The Ethernet is a CSMA / CD bus type LAN, and a node having transmission data checks whether or not communication is currently being performed. If the node is free, data is transmitted on the LAN. However, a plurality of nodes may make the same determination and start sending packets at the same time. In this case, packet collision occurs. When a packet collision occurs, each node detects the collision and immediately stops sending. Each node waits for a different retransmission waiting time based on a random number so as not to repeat the collision, and tries transmission again. In addition, a minimum idle time (blank time) must be waited between transfers, and the minimum idle time is basically set to the same value in any NIC. This means that in order to check whether a certain node has started the transfer, the transfer time must be refrained. In this embodiment, the logical limit value of the minimum idle time calculated from the network length and the transfer rate is referred to as ΔTa.
[0011]
As described above, since the minimum idle time of the NIC of the normal Ethernet is constant, if the minimum idle time of the NIC of a certain node is shortened, the node has a higher transmission chance than the other nodes. As a result, the node can increase the transfer rate more than other nodes. The present embodiment utilizes this Ethernet operating principle.
[0012]
An embodiment of the present invention will be described.
[0013]
A system configuration in which a monitoring network device using a plurality of cameras and sensors is used in a mixed manner with an OA network device will be briefly described with reference to FIG. That is, in this system, a monitoring system and an OA system exist together.
[0014]
The camera 40 is connected to the network 10 via the camera control device 30, the sensor 60 is connected via the sensor control device 50, the monitor 16 is connected via the monitor control device 18, and the monitor console control device 20 operated by the user includes A display 14, a mouse 24 as an input device, and a keyboard 22 are provided, and are connected to the network 10 like other controllers. The user 12 uses the monitoring console control device 20 to execute the system. A plurality of consoles 20 can be installed on the system.
[0015]
During operation of the monitoring system, an image from the camera 40 is displayed on the monitor 16 via the network 10 based on an instruction from the console 20. When the sensor 60 detects an abnormality such as a fire or an intrusion, the sensor 60 notifies the user 12 of the alarm information via the console 20. In the monitoring system, images from a plurality of cameras are simultaneously displayed on a plurality of monitors 16 via the network 10.
[0016]
Furthermore, the network 10 is also connected to an OA system device including a personal computer 60, a printer 62, and a server 64 having functions such as a mail server and a domain name server, and is operated as an OA system.
[0017]
FIG. 2 is a diagram showing a hardware configuration of the monitoring console control device 20. The network interface card 112, the CPU 102, the memory 104, the input / output controller 106 for controlling the keyboard 22 and the mouse 24, and the monitor controller 108 are connected by the bus 110. Has been.
[0018]
FIG. 3 is a diagram showing a hardware configuration of the monitor control device 18-1. A monitor controller 120, a CPU 122, a memory 124, and a network interface card 126 for controlling the monitor 16-1 for displaying an image are connected by a bus 128. It is connected.
[0019]
FIG. 4 is a diagram illustrating a hardware configuration of the camera control device 30-1 which is a bandwidth management target node according to the present exemplary embodiment. The camera controller 136, the CPU 132, the memory 134, and the camera controller 136 that control the video camera 40-1. A network interface card 130-A is connected by a bus 138.
[0020]
FIG. 5 is a diagram showing a hardware configuration of the sensor control device 50. A sensor controller 146, a CPU 142, a memory 144, and a network interface card 140 that are connected to the sensor group 60 and read an abnormality detection signal from the sensor settings and sensors are provided. They are connected by a bus 148.
[0021]
FIG. 6 is a diagram showing the configuration of software stored in the memory 104 of the monitoring console control apparatus 20. That is, a network interface card control driver 174 for controlling the network interface card 112, a message is transferred to a specified device via the network 10, a one-way communication, and a message is transferred to all devices connected to the network. A communication management module 172 that provides broadcast communication and an input / output device control driver 150 that controls the monitor 14, the keyboard 22, and the mouse 24 are stored. In addition, a GUI control module 152 and a video / alarm display module 215 are stored as modules using these software, and a plurality of transmission queues 166-A, 166-1, and reception queues having different transfer rates from the communication management module 172, respectively. Link through 170. The transfer rate management module 164 uses these transmission queues according to the type and status of the data transfer service.
[0022]
When requesting data transfer, the transfer service module 156 performs processing according to the transfer service type. The transfer service types include “bandwidth reservation transfer service” and “statistical multiplex transfer service”. Data transferred by these transfer services is called bandwidth reservation data and statistically multiplexed data, respectively. Further, the transfer rate management module 164 controls the network interface card via the network interface card control driver. The bandwidth management manager module 160 reserves, releases, and calculates the transfer rate according to the data transfer service of the communication path, and when a communication delay occurs in the network, priority transmission control for eliminating the delay, Release. These executions are realized by issuing commands to other nodes via the communication management module 172.
[0023]
FIG. 7 is a diagram showing the configuration of software stored in the memory 124 of the monitor control device 18-1. A network interface card control driver 198, a communication management module 196, and a monitor control driver 180 for controlling the monitor are stored. Also, a monitor control module 184 and a reception control module 188 for controlling received data are stored as modules that use these.
[0024]
FIG. 8 is a diagram illustrating a software configuration implemented in the memory 134 of the camera control device 30-1 that is a bandwidth management target node of this embodiment, and includes a network interface card control driver 200, a communication management module 202, and a video camera. A camera control driver 218 for controlling 40-1 is stored. Also, a camera control module 216 is stored as a module that uses these software, and cooperates via the communication management module 202 via the reception queue 204 and the transmission queue 206, respectively.
[0025]
The network interface card (NIC) 130-1 can control the following functions in addition to the functions of a normal Ethernet card via the network interface card control driver 200-1.
[0026]
(1) A function for changing the minimum idle time with the logical limit value ΔTa of the minimum idle time as a lower limit.
(2) A function to change the retry time to wait for retransmission until a collision occurs on the network
(3) A function of detecting the packet length and other information by reading information on a specific position of the packet in the all-packet reception mode and detecting the transmission end time of the packet. By using this function, transmission can be started without waiting for the logical limit value ΔTa of the minimum idle time of (1).
[0027]
FIG. 9 is a diagram showing a configuration of software stored in the memory 144 of the sensor control device 50-1, and a network interface card control driver 220, a communication management module 222, and a sensor control driver for controlling the sensor 60. 238 is stored. In addition, a sensor control module 236 and an alarm transmission module 230 are stored as modules that use these software, respectively, via a communication management module 222, a reception queue 224, and a plurality of transmission queues (226-A, ... 226-1). Work together. Also stored is a message management table 232 for storing a display message to be set for an alarm when the sensor detects an abnormality.
[0028]
FIG. 10 is a diagram showing a hardware configuration of the personal computer 61-1. A network interface card 613, a CPU 612, a memory 614, an input / output controller 616 for controlling a keyboard and a mouse, and a monitor controller 618 are connected by a bus 610. .
[0029]
FIG. 11 is a diagram showing the configuration of software stored in the memory 614 of the personal computer 60-1. The network interface card control driver 6142, various modules 6146 configured with or without using the API of the network interface card control driver, and the input / output device control driver 6148 are stored.
[0030]
FIG. 12 is a diagram showing a hardware configuration of the printer 62, and a network interface card 622, a CPU 626, a memory 624, and a printing device 628 are connected by a bus 620.
[0031]
FIG. 13 is a diagram showing the configuration of software stored in the memory 624 of the printer 64. The network interface card control driver 6242, various modules 6246 configured to use or not use the API of the network interface card control driver, and the printing apparatus control driver 6248 are stored.
[0032]
FIG. 14 is a diagram showing a hardware configuration of the server 64. A network interface card 641, a CPU 642, a memory 644, an input / output controller 646 for controlling a keyboard and a mouse, and a monitor controller 648 are connected by a bus 640.
[0033]
FIG. 15 is a diagram showing the configuration of software stored in the memory 644 of the server 64. The network interface card control driver 6442, various modules 6446 configured to use or not use the API of the network interface card control driver, and an input / output control driver 6448 are stored.
[0034]
FIG. 16 and FIG. 17 show an example of the operation form of the monitoring system in this system.
[0035]
In FIG. 16, image data is transmitted from a plurality of camera nodes (30-A, 30-B) to monitor nodes (16-A, 16-B) (502, 504). At the same time, alarm information from the connected sensors is transferred from the sensor nodes (50-A, 50-B) to the monitoring console controller 20 (506, 508).
[0036]
In FIG. 17, the state from FIG. 16 is switched, and image data is transferred from another camera node (30-C, 30-1) to the monitor node (16-C, 16-1) (510, 512). . Further, alarm information from the connected sensor is transferred from the sensor node (50-1) to the monitoring console controller 20 (514). At the same time, printing data is transferred from the personal computer 60-A to the printer 62.
[0037]
In the monitoring system, the sensor 60 detects some abnormality (fire, intrusion, etc.) during the sequential video tour in this way, and the sensor node 50 sends the alarm information to the monitoring console control device 20 where the user is located. Report. In this way, control data such as alarm information is mixed with image data and flows on the network 10. In addition, OA data including a server 64 operating as an OA system, a personal computer 61, and a printer 62 flows through the network 10. Such data is referred to as “other data” in order to distinguish from the above-described band reservation data and statistically multiplexed data.
[0038]
Next, the bandwidth of the communication path when the above operation is performed will be described with reference to FIGS.
[0039]
In FIG. 18, if the width of the total effective bandwidth 520 of the communication path is B1, the bandwidth reservation data is allocated for B2 of the bandwidth, and the bandwidth B3 is allocated for statistical multiplexed data and other data (for example, OA data). Here, the bandwidth reservation data has a large data size like video data and is applied to data that flows continuously in a fixed cycle. Therefore, if the number of videos that flow in advance and the size of one video frame are known, the bandwidth reservation data The width can also be determined. Therefore, if a bandwidth is secured based on this, the bandwidth can be used without waste during video data transfer.
[0040]
On the other hand, statistical multiplexed data and other data such as the OA system are used for data whose data size is small but whose generation timing cannot be predicted. Therefore, even if the occupied maximum bandwidth is determined from the number of nodes and the data size, it can be expected that the bandwidth actually used is much smaller than that.
[0041]
Even if the bandwidth is estimated with the width B3 shown in FIG. 18, what is actually used may be the state 526 shown in FIG. 19 or the state 528 shown in FIG.
[0042]
Fig. 20 shows that when multiple sensors are issued simultaneously due to the occurrence of a fire, etc., or when a large amount of data is transferred from an OA personal computer, data for statistical multiplex transfer service or OA data is scheduled. It shows how it occurred beyond the bandwidth. In this situation, the load on the communication path increases, and it becomes impossible to use the scheduled communication bandwidth for each of the bandwidth reservation data, the statistical multiplexed data, and the OA data, and there is a risk of data delay or data loss. is there.
[0043]
In the normal method, it is possible to suppress the transfer rate of the bandwidth reservation data transmitted by the bandwidth management target node such as the camera node, but the bandwidth management non-target node such as the sensor node or the OA system node transmits it. The transfer rate cannot be suppressed for statistical multiplexed data and other data.
[0044]
FIG. 21 shows a sequence for executing the bandwidth reservation transfer service. When there is a bandwidth reservation transfer service execution instruction from the user or a user program (6000), the transfer service module 240 issues a bandwidth reservation request 6020 to the bandwidth management manager. The bandwidth management manager returns a bandwidth reservation OK 6040 if there is a free bandwidth. The transfer service module broadcasts an image transfer request 6060, and the necessary camera node receives the request and starts image transmission. The camera node that has started image transmission continues to send images at the determined transfer rate until the next stop request is received. In this way, the bandwidth reservation data can be applied to data that is large like image data and whose transfer timing is predicted in advance and continues to be sent at regular intervals.
[0045]
FIG. 22 shows a sequence for executing the statistical multiplex transfer service.
[0046]
The sensor node to which the sensor that detected the abnormality is connected transmits an alarm report to the monitoring console node when the abnormality occurs. At this time, the statistical multiplex transfer service is used, and the use of this transfer service does not secure the bandwidth to the bandwidth manager. Thereby, data transfer can be performed immediately. As described above, the statistical multiplex transfer service is used for small data such as control data, the occurrence of which is sporadic and the occurrence timing cannot be predicted.
[0047]
Note that the OA node does not belong to the above-described bandwidth reservation transfer service or statistical multiplex transfer service, and the data generation timing and data amount depend on the user who operates the OA node. These OA nodes exist on the same network as the monitoring system node, but are not system components of the monitoring system.
[0048]
Next, the priority transmission control / priority transmission control release function will be described with reference to FIGS. The priority transmission control / priority transmission control release function maintains the quality of the bandwidth reservation transfer service even when the load on the communication path increases in the state of FIG. 20, specifically, the transfer rate of the bandwidth reservation data. Is to hold.
[0049]
First, in order to perform priority transmission control, it is necessary to detect the load on the communication path.
[0050]
In this system, the bandwidth management manager module 160 of the monitoring console control device 20 detects as a means for detecting the bandwidth used for the communication path occupied by various data. The bandwidth management manager module 160 issues a command to the network interface card 112 to switch to the all packet reception mode, so that all transfer packets on the network can be received. The means for the bandwidth management manager module 160 to detect the transfer rate using this all packet reception mode will be described below.
[0051]
FIG. 23 shows the format of the bandwidth information table 154 managed by the bandwidth control manager. The area 332 is a band identifier area for setting an identifier for identifying the reserved band, the area 334 is the reserved bandwidth, and the area 336 is one of parameters transmitted to the band control target node during priority transmission control. In the area 338, a retry time, which is one of the parameters transmitted to the band control target node, is stored.
[0052]
FIG. 24 shows a format 350 of data transmitted / received between the communication management modules of each node used in the monitoring system. This format is developed in the data portion of a message used in a lower protocol (for example, TCP / UDP / IP), and is interpreted by the communication management module.
[0053]
352 is an area for setting a transaction code TCD indicating the contents of this message, and 353 is a band identifier if this message is a band reservation message. 354 is an area for setting the address SA of the transmission source control device, 356 is an area for setting the address DA of the transmission destination control device, and 358 is an area for setting the data length L of the message main body set in the area 360. The module in each control device creates a message in this format and requests the communication management module to send it. Here, the transaction code (hereinafter abbreviated as TCD) and the address of each control device are assigned in advance. Also, it is assumed that broadcast can be specified by setting a predetermined value, for example, “0”, to the transmission destination address DA.
[0054]
On the other hand, regarding the reception of a message, when a TCD message specified on the network is broadcast by designating a TCD of a message to be received to the communication management module, or addressed to the node. The message is passed to the requesting module.
[0055]
FIG. 25 is a flowchart of transfer rate detection of the bandwidth management manager module 160.
[0056]
In the processing 1001, the NIC of the monitoring console controller 20 is set to the all process reception mode, and in the processing 1002, it is the time to observe the transfer rate described in the setting file from the input / output device of the monitoring console controller 20 in advance. The detection interval time Δt input is input, and the current time is recorded in processing 1004. When the detection end time t2 is calculated in the process 1006, a timer daemon that sends a signal to the program when a predetermined time is reached in the process 1008 is generated and the end time t2 is set. Until a signal comes in at time t2 in processing 1040, all packets flowing on the network are received from the NIC in processing 1010 and stored in the buffer. Step 1012 takes one of the packets from the buffer, and step 1014 measures the size of the packet. In this case, the size of the header part used in the lower protocol is also included. In processing 1016, the format of the data portion is checked. Here, when the received data portion is applied to the 350 communication data format described above, it is checked whether or not it can be taken out as a reasonable value, and it is detected whether or not the data belongs to the monitoring system.
[0057]
In the process 1020, the areas 352, 354, 356, and 358 of the communication data format 350 are further examined to determine whether this data is the bandwidth reservation data for the monitoring system, the statistical multiplexed data for the monitoring system, and other data, Data amounts are added in processes 1022, 1024, and 1028, respectively. Further, in process 1022, the amount of data is calculated for each band identifier unit. In step 1029, the packet is discarded, and in step 1030, it is checked whether or not the packet still remains in the buffer. If it remains, the processing returns to the processing 1012. If not, the bandwidth occupied by various data is calculated in processing 1032. This calculation is obtained by dividing the total data amount obtained by adding the respective data by the detection interval time Δt. In processing 1034, the occupied bandwidth of the various data is compared with the bandwidth information table 154, and in processing 1036, it is determined whether or not to change the transfer rate of the bandwidth reservation data.
[0058]
The approval criteria is whether or not there is a risk that the transfer amount of data other than the bandwidth reservation data will increase and there will be a delay in the transfer of the bandwidth reservation data. A process for reducing the occupied bandwidth of data other than the bandwidth reservation data by reducing the minimum idle time for the NIC or by securing the occupied bandwidth of the bandwidth reservation data by other means is performed. Thereafter, in process 1038, the respective data amounts are reset to 0, and the process returns to process 1004.
[0059]
FIG. 35 shows an approval process for changing the transfer rate of the bandwidth reservation data in process 1036. In the process 1036-1, the occupied band for each band identifier is referred to, and in the process 1036-2, it is checked whether each band exceeds the band allowable rate of 336. In the case where it exceeds, ΔT1 time which is a tuning adjustment time interval of the minimum idle time is added to the minimum idle time of the NIC of the node that is transmitting in the process 1036-3, and the retry time is returned to the default value. If not, it is checked in step 1036-4 if it is not below. If not, the process is terminated. In process 1036-5, ΔT time is subtracted from the minimum idle time of the relevant node, and in process 1036-6, it is checked whether or not the minimum idle time is smaller than the logical limit value.
[0060]
If it is smaller, ΔT2 time, which is the tuning adjustment time interval of the retry time, is subtracted from the NIC retry time of the corresponding node in processing 1036-7. If the retry time is 0 or less in the process 1036-8, the retry time is set to 0 and the highest priority control instruction command is set in the process 1036-9. As will be described later, the highest priority control is a method in which transmission is performed without waiting for the idle time and retry time described above, and processing that logically realizes priority transmission is performed.
[0061]
In the above, when it is necessary to transmit the priority control message to the bandwidth reservation target node, the message is transmitted in processing 1036-10.
[0062]
FIG. 26 shows a processing sequence when the NIC parameter change of the bandwidth reservation data is determined in order to perform the priority transmission control in the processing 1036. When the parameter change is determined in the process 266, the NIC of the camera nodes 30-A and 30-B and the monitor nodes 18-A and 18-B that transmit and receive the respective band reservation data in the process 268 is determined. A message is sent instructing a change in minimum idle time or other parameters.
[0063]
In the bandwidth management target node instructed to perform priority transmission processing, priority transmission control is performed (270, 272). However, in the state where all the bandwidths are used up in both the bandwidth management target node and the bandwidth management non-target node. In general, it is generally impossible to change the transfer rate.
[0064]
Therefore, in this embodiment, using the properties of Ethernet, the transfer rate of the non-target bandwidth management node, which is a node that does not have the transfer rate changing function, is reduced, and the bandwidth of the bandwidth management target node is substantially retained. An example is shown.
[0065]
FIG. 27 shows an example of a communication packet 350-A when the bandwidth management manager issues a priority transmission control request. The TCD number indicating “priority transmission control request” is input to 352-A, the address number of the monitoring console control apparatus that transmits this request is transmitted to 354-A, and the broadcast area is specified to DA area 356-A. Are set in the data length L area 358-A of the message body and various parameters for priority transmission control are set in the data area 360-A.
[0066]
FIG. 28 shows an example of a communication packet 350-C when the bandwidth management manager issues a minimum idle time change request. The TCD number indicating “change in minimum idle time” is input to 352-C, the address number of the monitoring console control apparatus that transmits this request is input to 354-C, and the broadcast area is specified to DA area 356-C. In the data length L area 358-C of the message body portion, the byte length is input, and in the data area 360-C, the changed minimum idle time is input.
[0067]
In addition to the TCD number indicating “minimum address time change”, 352-C also has TCD numbers such as “retry time change” and “estimate other node packet transmission end time”, and the data area 306-C includes A value corresponding to this TCD is input.
[0068]
The effect of these parameters will be specifically described with reference to FIGS.
[0069]
FIG. 29 shows an operation at the start of transmission of a general Ethernet NIC. When the NIC detects the carrier, the carrier disappears and waits for the minimum idle time that is longer than the logical limit value ΔTa of the minimum idle time, and then does not collide with the NIC waiting for another transmission. Then, after waiting for a collision avoidance time (retry time) determined randomly, it is confirmed that another carrier is not detected during that time, and then transmission is started.
[0070]
FIG. 30 illustrates an operation at the start of transmission of the NIC for the band management target node used in the embodiment of the present invention. Proposed NIC The minimum idle time can be freely changed with the logical limit value ΔTa of the minimum idle time as the lower limit. Therefore, even if the retry time is shortened and transmitted, it is transmitted more than normal NIC. The start time can be advanced. By doing so, the chance of transmission is increased as compared with a node equipped with a general NIC, and as a result, the transfer rate can be improved.
[0071]
FIG. 31 shows an operation when a collision occurs in a general Ethernet NIC and retransmission is started. When a collision occurs, a retransmission is started after waiting for a certain retry time and a retry time determined in a random manner so as not to collide with a NIC waiting for another transmission thereafter.
[0072]
FIG. 32 shows an operation when a collision occurs in the NIC for the band management target node used in the embodiment of the present invention and retransmission is started. When a collision occurs, a retransmission is started after waiting for a certain retry time and a retry time determined in a random manner so as not to collide with a NIC waiting for another transmission thereafter. At this time, by shortening the retry time, the transmission chance is increased as compared with a node equipped with a general NIC, and as a result, the transfer rate can be improved.
[0073]
FIG. 33 shows the operation when the NIC for the band management target node used in the embodiment of the present invention starts another transmission by observing another node packet and predicting its transmission end time. The NIC reads the contents of other packets in the all-packet reception mode, detects the packet length and other information by reading the information on the specific position of the packet, and detects the packet transmission end time. After this end time, the NIC waits for a retry time determined in a random manner and starts transmission. In this case, since the NIC already knows the exact time at which the carrier disappears, transmission is suddenly started without confirming that the carrier has disappeared. Therefore, transmission can be started without waiting for even the logical limit value ΔTa of the minimum idle time. If this method is used, there is virtually no chance for other general-purpose NICs to transmit, and the highest priority transmission control by the node having the proposed NIC can be realized.
[0074]
In this embodiment, change of the “minimum idle time” and “retry time”, which are parameters for changing the state of the proposed NIC, and the packet transmission end time of other nodes are calculated in advance to check for carrier loss. Depending on how you start sending without
Data can be transmitted with priority over nodes using other general NICs, but the relationship between the bandwidth occupancy rates of various data when this method is used is the number of target nodes, etc. Of course, the scale of the system is dynamically changed depending on the data length and data transmission frequency of the non-band management target nodes such as the sensor control node, the PC, and the server, and therefore cannot be predicted.
[0075]
Accordingly, the transfer rate detection flow of the bandwidth management manager module 160 is executed at any time, and the packet length change request message of FIG. 28 is transmitted at a frequency that does not place an excessive load on the network, and the bandwidth of the bandwidth reservation data is dynamically Continue to tune to ensure that
[0076]
FIG. 34 is a flowchart showing processing in which a bandwidth management target node that performs a bandwidth reservation service receives a priority transmission control request message and sets parameters as instructed in the NIC. In process 390-1, a message is received and its contents are checked. In process 390-2, it is confirmed whether or not the message is a priority transmission control request message. If not, the corresponding other process is performed in process 390-3. When the change of the idle time is instructed in the process 390-4, the idle time of the own node NIC is reset to the NIC of the own node via the network interface card control driver in the process 390-5. If the retry time is instructed in the process 390-6, the retry time of the own node NIC is reset for the NIC of the own node via the network interface card control driver in the process 390-7. If the highest-priority transmission processing instruction is issued in process 390-6, the highest-priority transmission process is set via the network interface card control driver for the local node NIC in process 390-7. In process 390-10, since neither process could be performed, the error content is entered in the error log and the process is terminated.
[0077]
【The invention's effect】
As described above, the effect of the present invention is to realize transfer control of a node having no network bandwidth management function in a network system in which nodes having a network bandwidth management function and nodes having no network bandwidth management function coexist. Network bandwidth control can be performed.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a hardware configuration of a monitoring console control apparatus operated by a user in the monitoring system according to the embodiment of the present invention.
FIG. 3 is a diagram illustrating a hardware configuration of a monitor control device in the monitoring system according to the embodiment of the present invention.
FIG. 4 is a hardware configuration diagram of a camera control device in the monitoring system according to the embodiment of the present invention.
FIG. 5 is a hardware configuration diagram of a sensor control device in the monitoring system according to the embodiment of the present invention.
FIG. 6 is a software configuration diagram of the monitoring console control device in the monitoring system according to the embodiment of the present invention.
FIG. 7 is a software configuration diagram of a monitor control device in the monitoring system according to the embodiment of the present invention.
FIG. 8 is a software configuration diagram of the camera control device in the monitoring system according to the embodiment of the present invention.
FIG. 9 is a software configuration diagram of the sensor control device in the monitoring system according to the embodiment of the present invention.
FIG. 10 is a hardware configuration diagram of a personal computer in the OA system according to the embodiment of this invention.
FIG. 11 is a software configuration diagram of a personal computer in the OA system according to the embodiment of this invention.
FIG. 12 is a hardware configuration diagram of a printer in the OA system according to the embodiment of this invention.
FIG. 13 is a software configuration diagram of a printer in the OA system according to the embodiment of this invention.
FIG. 14 is a hardware configuration diagram of a server in the OA system according to the embodiment of this invention.
FIG. 15 is a software configuration diagram of a server in the OA system according to the embodiment of this invention.
FIG. 16 is a diagram illustrating an example of an operation mode of the monitoring system in the monitoring system according to the embodiment of this invention.
FIG. 17 is a diagram illustrating an example of an operation mode of the monitoring system in the monitoring system according to the embodiment of this invention.
FIG. 18 is a diagram showing a state of distribution of communication path bands managed by the monitoring system in the monitoring system according to the embodiment of the present invention.
FIG. 19 is a diagram showing a state of distribution of communication path bands managed by the monitoring system in the monitoring system according to the embodiment of the present invention.
FIG. 20 is a diagram showing a state of distribution of communication path bands managed by the monitoring system in the monitoring system according to the embodiment of the present invention.
FIG. 21 is a diagram illustrating an example of an operation sequence of a bandwidth reservation transfer service performed by the monitoring system in the monitoring system according to the embodiment of this invention.
FIG. 22 is a diagram showing a sequence of statistical multiplex transfer service execution in the monitoring system of the embodiment of the present invention.
FIG. 23 is a diagram illustrating a format of a bandwidth information table managed by a bandwidth control manager in the monitoring system according to the embodiment of this invention.
FIG. 24 is a diagram showing a format of data transmitted between the communication management modules of each node in the monitoring system according to the embodiment of the present invention.
FIGS. 25A and 25B are a flowchart for detecting a transfer rate by a bandwidth management manager module and a flowchart for determining a change in a transfer method of bandwidth reservation data by the bandwidth management manager module in the monitoring system according to the embodiment of this invention.
FIG. 26 is a diagram showing a processing sequence when the bandwidth management manager module determines to change the bandwidth reservation data transfer rate in the monitoring system of the embodiment of the present invention.
FIG. 27 is a diagram illustrating a communication packet when the bandwidth management manager issues a priority transmission control request in the monitoring system according to the embodiment of this invention.
FIG. 28 is a diagram showing a communication packet when the bandwidth management manager in the monitoring system according to the embodiment of the present invention issues a minimum idle time change request.
FIG. 29 is a diagram illustrating an operation at the start of transmission of a general Ethernet NIC in the monitoring system according to the embodiment of this invention.
FIG. 30 is a diagram illustrating an operation at the start of transmission of a NIC for a band management target node in the monitoring system according to the embodiment of this invention;
FIG. 31 is a diagram showing an operation when a collision occurs in a general Ethernet NIC and retransmission is started in the monitoring system according to the embodiment of the present invention.
FIG. 32 illustrates an operation when a collision occurs in the NIC for the band management target node and retransmission is started in the monitoring system according to the embodiment of this invention.
FIG. 33 is a diagram showing an operation when the bandwidth management target NIC observes another node packet and predicts its transmission end time and starts transmission in the monitoring system of the embodiment of the present invention. is there.
FIG. 34 is a flowchart showing processing in which a bandwidth management target node that performs a bandwidth reservation service of the present invention receives a priority transmission control request message and sets parameters as instructed in the NIC.
FIG. 35 is a flowchart of a transfer rate detection by the bandwidth management manager module and a flowchart by which the bandwidth management manager module determines a change of the bandwidth reservation data transfer method in the monitoring system according to the embodiment of the present invention.
[Explanation of symbols]
10 ... Network, 12 ... User, 14 ... Display,
30 ... Camera control device, 40 ... Camera, 50 ... Sensor control device,
60: Sensor.

Claims (7)

In a bandwidth management system having a bandwidth management node that detects a bandwidth usage status of a network in which one or more bandwidth management target nodes and one or more bandwidth management non-target nodes are connected,
When the bandwidth management node determines that there is a possibility of delay in transferring the bandwidth reservation data output from the bandwidth management target node based on the detected bandwidth usage status, the network of the bandwidth management node A bandwidth control means for transmitting a transmission control parameter change command set in the interface card to the bandwidth management target node;
The bandwidth management target node has data transfer means for changing the transmission control parameter based on the received change command, and executing transfer of the bandwidth reservation data based on the changed transmission control parameter,
When it is determined that the delay may occur, the bandwidth reservation data output from the bandwidth management target node is transferred with higher priority than the data output from the bandwidth management non-target node. Bandwidth control system. The bandwidth control system according to claim 1, wherein
The bandwidth control means, wherein when the transfer rate of the bandwidth reservation data is less than a reserved bandwidth fluctuation allowable rate, it is determined that the bandwidth reservation data may be delayed. system. The bandwidth control system according to claim 1, wherein
If the transfer rate of the bandwidth reservation data exceeds the reserved bandwidth fluctuation allowable rate, the bandwidth control means sets the transmission control parameter set in the network interface card of the bandwidth management target node to an initial set value. A bandwidth control system for transmitting to the bandwidth management target node a change command for changing to a direction of returning to the bandwidth management target node. The bandwidth control system according to claim 1, wherein
The bandwidth control system, wherein the transmission control parameter is a minimum idle time. The bandwidth control system according to claim 1, wherein
The bandwidth control system, wherein the transmission control parameter is a retry time. In a bandwidth management system having a bandwidth management node that detects a bandwidth usage status of a network in which one or more bandwidth management target nodes and one or more bandwidth management non-target nodes are connected,
The bandwidth management target node is:
Included in the packet flowing through the network when the bandwidth management node determines that there is a possibility of delay in transferring bandwidth reservation data output from the bandwidth management target node based on the detected bandwidth usage status Calculating means for calculating the transmission end time of the packet based on the received information;
Data that starts transfer of the bandwidth reservation data after the calculated transmission end time has elapsed and before the minimum idle time that is the transmission control parameter has elapsed or before the logical limit value of the minimum idle time has elapsed. Transfer means, and
When it is determined that the delay may occur, the bandwidth reservation data output from the bandwidth management node is transferred with higher priority than the data output from the bandwidth management non-target node. Bandwidth control system. The bandwidth control system according to claim 6, wherein
The bandwidth control system is configured to transfer the bandwidth reservation data after the calculated transmission end time has elapsed and after the retry time has elapsed.

Images