JP2004032474A - Network system - Google Patents

Abstract

A communication band is controlled for each communication service without using a special network connection device.
When a communication service B is provided from a transmission source computer 101-B to a destination computer 1012-B, it is detected that communication data has been discarded on the network in order to limit the bandwidth usage in the network C. Communication is performed by providing data discard detection means 107 and transmission rate changing means 106 for changing the transmission rate in response to the notification of data discard. As the communication path is made to pass through the sub-network D204 having a narrower communication band than the shared network C203, the communication band of the communication service B is changed by changing the transmission rate according to the discard of the communication data. Can be controlled to be equal to or less than that of the sub-network D204 having a narrow width.
[Selection diagram] Fig. 1

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a communication system in which a plurality of data transfer apparatuses share a network, and more particularly to a band control method capable of controlling a communication band on a network for each communication service.
[0002]
[Prior art]
In recent years, applications that support multimedia such as graphics, still images, audio, and moving images that operate on personal computers have become widespread. When transmitting and receiving multimedia information over a network using such an application, if the audio data and image data are not delivered to the receiving computer within a certain delay time, they are reproduced as human-recognizable audio and images. I can't do that. As a result, the sound is lost and the image is disturbed. This data delay is caused by the fact that the network bandwidth is used by another communication service during communication or when communication is to be started. Therefore, it is desired to provide a system that performs priority transfer control of a packet or preferentially secures a band in accordance with a request for each communication service such as an application, and is actively studied.
[0003]
As a typical method, a method of performing priority control using a Precedence parameter of a ToS (Type of Service) field as a parameter for setting a transfer priority of an IP datagram, IEEE802.3p, and the like are known.
[0004]
For example, in Japanese Patent Application Laid-Open No. 9-205461, in a network system including an IP router, priority control is performed by providing a priority control processing function. That is, the transmission source computer specifies the priority class in the ToS (Type of Service) field of the IP packet for each communication service. In the IP router, for the value of the ToS field, information indicating which priority queue is to be input, and the order of extracting IP packets from each priority queue (for example, the order in which the packets are higher in priority) are set. Keep it. The IP packet transmitted from the transmission source computer is input to a queue corresponding to the value of the ToS field in the IP router. The IP router takes out an IP packet from the priority queue according to the set order, and transfers it to the network. That is, in the IP router, priority control is realized by passing IP packets according to the priority.
[0005]
[Problems to be solved by the invention]
In the priority control method disclosed in Japanese Patent Application Laid-Open No. 9-205461, priority transmission is performed in accordance with the priority indicated by the ToS field defined in the IP header part, so that a network band desired by a communication service can be allocated. Will be possible.
[0006]
However, the priority control method disclosed in this publication has a limitation that the network system must be configured with an IP router (network connection device) having the above-described priority control processing function. In a normal network connection device that does not have the priority control processing function, high-priority communication data and low-priority communication data are treated the same.
[0007]
Therefore, when there is at least one ordinary network connection device not provided with the priority control processing function in the network system, the priority transfer is not performed. In other words, even if the data is preferentially transferred by the priority control immediately before passing through the normal network connection device, the normal network connection device does not perform the priority transfer.
[0008]
Therefore, when the communication traffic of the communication service via the normal network connection device is large, a transfer delay may occur. In the worst case, the end-to-end delay time from the transmission source computer to the destination computer may exceed the predetermined range, which may cause a situation in which the predetermined communication service cannot be realized. is there.
[0009]
In view of the above, it is an object of the present invention to provide a network system which limits a communication band for each communication service even in a network system including a network connection device having no priority control processing function.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, in a shared sub-network in which a high-priority communication service and a low-priority communication service share a band, a high-priority communication service has a higher priority than a lower-priority communication service. The communication band must be able to be used.
[0011]
For this purpose, the present invention is characterized by providing a bandwidth limiting means for limiting the bandwidth that can flow into the shared sub-network for a communication service with a low priority. That is, the communication bandwidth can be preferentially secured without any particular limitation on the bandwidth amount of the communication service with high priority. The bandwidth limiting means indicates the following transmission rate changing function, data relay function, and the like.
[0012]
1 illustrates an embodiment of the invention. In the first mode, a transmission rate changing function is provided in the bandwidth limiting means. That is, in a network system having a shared sub-network and a sub-network having a narrower communication bandwidth, a data discard detecting means for detecting occurrence of discard of communication data on the network, and transmitting when detecting discard of data. A transmission rate changing means for changing a rate is provided, and a certain communication service is made to communicate using the data discard detection means and the transmission rate changing means, thereby discarding data from the shared network to the sub-network having a narrow communication band. Enables communication without problems.
[0013]
Here, the communication band gap between the shared sub-network and the sub-network having a smaller bandwidth has an effect of causing packet loss. The data discard detection means and the transmission rate changing means decrease the transmission rate when a packet loss occurs, and increase the transmission rate when no packet loss occurs. This has the effect of converging the communication band amount of the communication service using this means to a value equal to or close to the communication band of the sub-network having a narrow communication band.
[0014]
As described above, since the communication service using the present means is controlled to the value of the communication band of the sub-network having a narrow bandwidth, it becomes impossible to flow data having the communication band or more into the shared sub-network. Therefore, the communication service that does not use the present means can freely use the remaining communication band not used by the communication service using the present means.
[0015]
Since the data discard detection means and the transmission rate change means only need to be installed in each of the transmission and reception computers that provide the communication service, the network system is configured by a network connection device having no priority control processing function. Even so, the above object can be achieved.
[0016]
In a second aspect, the bandwidth limiting means has a data relay function. That is, a data relay means for once receiving communication data transmitted from a communication service, holding the communication data for a time when a bandwidth usage amount of the communication service becomes a desired value, and transmitting the data again to the network is provided. If necessary, communication is performed via this data relay means. The desired value of the bandwidth usage is notified from the transmission source computer in the service. A desired value can be set in the data relay unit.
[0017]
According to this, the bandwidth usage of the communication service via the data relay means is limited to a predetermined value. For this reason, other communication services that do not pass through the data relay means can preferentially use the remaining communication band. Since this data relay means may be installed at least at one location in the network system, even if the network system is configured by a network connection device having no priority control processing function, it is possible to achieve the object. Can be done.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 2 shows a schematic configuration of a network system including a problem in order to explain one embodiment of the present invention. In FIG. 2, the communication service D uses the means for limiting the amount of band usage according to the present invention, and the communication service C does not use the means.
[0019]
The network system includes transmission source computers 201-C to 201-D, reception computers 202-C to 202-D, sub-networks C203 and D204, and a network connection device 210. The transmission source computer 201-D includes a transmission rate changing unit 206, and the destination computer 202-D serving as a reception computer includes a data discard detection unit 207. In FIG. 2, the communication bands of the sub-networks C203 and D204 are 100 Mbps for the sub-network C203 and 10 Mbps for the sub-network D204.
[0020]
First, the data flow of the communication service C will be described. The transmission source computer 201-C sends data to the network C203. Since the communication band of the network C203 is 100 Mbps, data 208 of a maximum of 100 Mbps is transmitted to the network C203 from the transmission source computer 201-C. Next, the data 208 transmitted from the transmission source computer 201-C reaches the boundary between the sub-network C203 and the sub-network D204. However, since the communication band of the sub-network D204 is 10 Mbps, only data up to 10 Mbps out of 100 Mbps can pass. Therefore, the difference between the communication bands of the sub-network C203 and the sub-network D204, that is, data 209 of up to 90 Mbps is discarded by the network connection device 210, and only the data that has passed through the network connection device 210 reaches the destination computer 202-C.
[0021]
The problem here is that the limit of data that can reach the destination computer 202-C is 10 Mbps data, which is the communication band of the subnetwork D, but the source computer 201-C uses That is, it consumes more communication bandwidth than necessary.
[0022]
Next, a data flow of the communication service D will be described. First, the transmission source computer 201-D starts communication by the transmission rate changing unit 206 while gradually increasing the transmission rate. The data 205 transmitted from the transmission source computer 201-D reaches the boundary between the sub-network C203 and the sub-network D204. However, since the transmission rate is low immediately after the start of communication, the data 205 transmitted from the transmission source computer 201-D flows into the subnetwork D204 without being discarded by the network connection device 210.
[0023]
When communication is continued for a while, the transmission rate of the communication service D increases to a rate where the communication band exceeds 10 Mbps. Data transmitted over 10 Mbps is discarded in the network connection device 210. In this case, the data discard detection means 207 of the destination computer 202-D detects that data discard has occurred, and notifies the transmission source computer 201-D of the occurrence of data discard. The transmission source computer 201-D that has received the notification reduces the transmission rate to an appropriate rate at which data discard does not occur. Thus, the transmission source computer 201-D can continue the communication without consuming an unnecessary communication band in the subnetwork C203.
[0024]
Next, a preferred application example of the present invention will be described. FIG. 1 shows an embodiment of the present invention, which is a network system showing an example in which a high-priority communication service can preferentially use a band as compared with a low-priority communication service.
[0025]
The network system in FIG. 1 includes transmission source computers 101-A to 101-B, destination computers 102-B to 102-C, sub-networks C203 and D204, and a network connection device 210. The transmission source computer 101-B includes a transmission rate changing unit 106, and the destination computer 102-B includes a data discard detection unit 107. The communication bands of the sub-networks C and D in FIG. 1 are 100 Mbps for the sub-network C203 and 10 Mbps for the sub-network D204, which are the same as in FIG.
[0026]
In the network system of FIG. 1, a communication service A108 is implemented between the source computer 101-A and the destination computer 102-C, and a communication service B105 is implemented between the source computer 101-B and the destination computer 102-B. Have been. As for the priorities of the two communication services, the communication service A 108 has a high priority and the communication service B 105 has a low priority. In the communication service B105 having a low priority, the communication band available in the sub-network C203 is limited by the following process. This processing is the same as that described for the communication service D in FIG.
[0027]
The transmission source computer 101-B starts communication while gradually increasing the transmission rate by the transmission rate changing unit 106. If the communication is continued for a while and the transmission rate of the communication service B increases to a rate exceeding 10 Mbps which is the band of the sub-network D204, the data transmitted over 10 Mbps is discarded in the network connection device 210. . At this time, the data discard detection means 107 of the destination computer 102-B notifies the transmission source computer 101-B that the data discard has occurred. The transmission source computer 101-B that has received the notification reduces the transmission rate to a rate at which data discard does not occur.
[0028]
In this way, for a low-priority communication service, a low-bandwidth network is passed between the source computer and the destination computer. Then, data is discarded, and the transmission rate of the low-priority communication service is controlled using the transmission rate changing means 106 and the data discard detection means 107 so that data discard does not occur. As a result, the communication band that the low-priority communication service B105 can use in the subnetwork C203 having a wide communication band can be made equal to or smaller than the communication band (here, 10 Mbps) of the subnetwork D204 having a narrow communication band.
[0029]
On the other hand, in the high-priority communication service A, there is no need to take any special measures or restrictions. As described above, the communication band that the low-priority communication service B can use in the sub-network C is limited to 10 Mbps or less. Therefore, the high-priority communication service A uses the unused communication band of the sub-network C. A certain 90 Mbps can be used freely.
[0030]
Next, a mechanism for controlling the communication band using the transmission rate changing means 206 and the data discard detecting means 207 in the subnetwork C will be described in detail.
[0031]
FIG. 3 shows a standby time storage table which is a component of the transmission rate changing means 206. The standby time storage table 301 includes an area for storing a standby time from when a data transfer module, which will be described later, transmits certain data to when the next data is transmitted. In this embodiment, the standby time is in seconds.
[0032]
FIG. 4 shows a consistent number storage table which is a component of the data discard detection means 207. The consistent number storage table 603 includes an area 601 for storing a current value of a consistent number (representing a transmission order of transmission data and a continuous number for each transmission data) added to the transmission data, and an area for storing a previous value thereof. 602.
[0033]
5 and 6 show the processing flow of the transmission rate changing means and the data discard detection means. Hereinafter, the flow of control when the communication service B performs communication from a subnetwork having a wide communication band to a subnetwork having a narrow communication band will be described with reference to the flowcharts of FIGS.
[0034]
In the transmission source computer 101-B, the data transmission module 406 is called in order for the communication service B to transmit data (400). The data transmission module 406 adds a consistent number (indicating the transmission order of transmission data and a continuous number for each transmission data) to the header part of the transmission data (401), and transmits the data passed from the communication service to the network. (402).
[0035]
Next, the data transmission module 406 subtracts 1 from the value of the standby time set in the standby time storage table 301. That is, the time until the next data transfer is performed is shortened by one second (403). As a result, the time interval at which the data transmission module 406 transmits data over the sub-network C is reduced, and the transmission rate is increased.
[0036]
After that, the data transmission module 406 adds 1 to the consistency number in the process 404, and suspends the process for the time stored in the standby time storage table 301 in the process 405.
[0037]
After the elapse of the time stored in the standby time storage table 301, the data transmission module 406 resumes the processing, and when there is a new data transmission request by the communication service B, performs data transmission by processing 401 and 402. If there is no new data transmission request, the process 400 waits until a new data transmission request is generated by the communication service B again. The data transmission module 406 repeats the above processes 400 to 405 one after another, and gradually increases the data transmission rate.
[0038]
In the transmission source computer 101-B, the transmission rate change module 407 is operating, and waits until it is notified that data has been discarded (called packet loss) (408). When a packet loss occurs, the data discard detection module 506 described later notifies that a packet loss has occurred. As a result, the transmission rate changing module 407 adds one or more values so that the value of the standby time storage table 301 increases. That is, the standby time is changed to a value longer than the current number of seconds by one second or more (409). As a result, the time interval at which the data transmission module 406 transmits data over the network increases, and the transmission rate decreases.
[0039]
On the other hand, the data discard detection module 507 in FIG. 6 waits for the arrival of data until the data is received, and when the data is received (501), acquires the consistent number from the header of the received data. Next, the acquired consistency number is stored in the present reception consistency number storage area 601 of the consistency number storage table 603 (502). Thereafter, in order to check whether or not a packet loss has occurred, the data discard detection module 507 compares the present reception consistent number 601 with a value obtained by adding 1 to the previous reception consistent number 602, and performs each processing according to the comparison result. I do.
[0040]
First, if the current reception consistent number 601 is the same as the value obtained by adding 1 to the previous reception consistent number 602, data can be received in the order in which they were transmitted. That is, it means that no packet loss has occurred. In that case, the value of the present reception consistent number 601 is set to the previous reception consistent number 602 in order to perform the same verification for the next received data (504).
[0041]
Contrary to the above, if the current reception consistency number 601 is not the same as the value obtained by adding 1 to the previous reception consistency number 602, the consistency number of the received data is skipped. That is, it means that a packet loss has occurred. In this case, the data discard detection module 507 notifies the transmission source computer 101-B that a packet loss has occurred (505), and requests that the transmission rate be reduced.
[0042]
When receiving the notification of the occurrence of the packet loss, the transmission rate change module 407 of the transmission source computer 101-B adds one or more values so that the value of the standby time storage table 301 increases, as described above, and Change the time to a value that is at least one second longer than the current number of seconds.
[0043]
Here, in the process 409 of the transmission rate change module 407, when the value added to the waiting time 301 is constant, that is, when the change value of the transmission rate is constant, the transmission rate is repeatedly increased, decreased, increased, and decreased. Will be. However, if the value to be added to the value in the standby time storage table 301 is gradually reduced according to the number of occurrences of packet loss, the reduction rate of the transmission rate can be gradually reduced. Finally, the transmission rate can be set to a value close to “the maximum transmission rate at which no packet loss occurs”.
[0044]
As described above, when the communication service B performs communication from the sub-network C having a wide communication band to the sub-network D having a narrow communication band, the communication can be performed without consuming an unnecessary communication band in the sub-network C. In other words, when the data discard detection unit 107 and the transmission rate changing unit 106 are used as in the case of the communication service B, communication can be performed in a communication band substantially equal to the communication band of the subnetwork D. This communication band is the maximum transmission rate at which no packet loss occurs. As a result, the remaining communication band can be preferentially used for another communication service using the subnetwork C, for example, the communication service A.
[0045]
Next, another embodiment will be described. FIG. 7 shows a network system for explaining another embodiment. This network system includes a source computer 701-E, a destination computer 702-E, sub-networks 703 and 704, and a data relay device 705. Communication service E is performed between the source computer 701-E and the destination computer 702-E. Note that the communication bands of the sub-networks 703 and 704 are both 100 Mbps.
[0046]
FIG. 8 is a configuration diagram of a data relay device used in the network system of FIG. The data relay device 705 includes a data receiving unit 801, a data transfer unit 803, a transfer interval determination unit 802, a transfer interval determination table 805, and a data holding unit 804. The data receiving unit 801 and the data transfer unit 803 are connected to different sub-networks, and transfer data received from the sub-network connected to the data reception unit 801 to the sub-network connected to the data transfer unit 803. I do.
[0047]
FIG. 9 shows the configuration of the transfer interval determination table 805. The transfer interval determination table 805 has a communication attribute storage area 902 for storing communication service attribute information such as the transmission address, destination address, and data type of received data. Also, a data holding time storage area 903 for storing the data holding time determined by the transfer interval determining means 802, a transfer interval log area 904 for storing the time of transfer performed so far, a communication band of the communication service desired by the user. Is stored in the area 905.
[0048]
The desired communication band is added to the head of the received data. Alternatively, a desired value may be set in the data relay device 705.
[0049]
Next, the operation of the data relay device 705 will be described. The data transmitted by the transmission source computer 701-E is received by the data receiving unit 801 of the data relay device 705. The data received by the data receiving unit 801 is passed to the transfer interval determining unit 802. The transfer interval determination unit 802 specifies a communication service based on attributes such as a source address, a destination address, and a data type.
[0050]
After specifying the communication service, the transfer interval determination unit 802 calculates how many bytes of data have been transferred within the past fixed time based on the transfer time recorded in the transfer interval log area 904 information. Then, how much data should be held so that the communication band of the specified communication service becomes the desired communication band 905 is determined. The determined time is stored in the data holding time storage area 903. Thereafter, the data is temporarily held by the data holding unit 804, and is transferred by the data transfer unit 803 after the time specified by the data holding time 903 has elapsed. When the transfer is completed, the data transfer unit 803 records the time in the transfer interval log area 904.
[0051]
In other words, the data relay device 705 calculates the communication bandwidth up to now based on the past transfer records recorded in the transfer interval log area 904, and calculates the time until the next transfer. Accordingly, the data relay device 705 can control the communication band of each communication service in the transfer destination sub-network to a value desired by the user.
[0052]
Next, an operation in which the data relay device 705 is applied to the network system of FIG. 7 will be described. First, the transmission source computer 701-E transmits data to the data relay device 705 via the sub-network 703. Next, the data relay device 705 holds the data so that the communication band of the communication service E has the value specified by the data holding time 903 of the transfer interval determination table 805, and then transfers the data to the subnetwork 704. .
[0053]
By doing so, the communication band of the communication service via the data relay device 705 is controlled to a desired value. Other communication services that do not pass through the data relay device 705, that is, communication services (not shown) that flow into the sub-network 703 and do not flow into the sub-network 704 can use the remaining communication band preferentially. Become.
[0054]
In the network system of FIG. 7, the data relay device 705 intentionally narrows down the amount of data that the communication service E passes. Thereby, the communication band of the communication service E can be made appropriate in the sub-network 704 after passing through the data relay device 705.
[0055]
However, in the sub-network 703 before passing through the data relay device 705, data is transmitted for the sake of the transmission source computer 701-E. For this reason, the communication service E may use more communication bands in the subnetwork 703 than data flowing into the subnetwork 704. In this case, since the amount of data that can pass through the data relay device 705 is determined by the transfer interval determination table 805, surplus data of the communication service E that has become surplus in the subnetwork 703 cannot pass through the data relay device 705. Destroyed.
[0056]
FIG. 10 is a configuration example of a network system showing another embodiment for solving the above problem. The network system in FIG. 10 includes a source computer 1001-G, a destination computer 1002-G, a subnetwork 1003 and a subnetwork 1004, and a data relay device 1005, as in the network system in FIG. Each of the transmission source computer 1001-G and the destination computer 1002-G includes a transmission rate changing unit 1006 and a data discard detection unit 1007, and the communication service G is provided via the data relay device 1005.
[0057]
The difference between the network systems shown in FIGS. 10 and 7 is that the transmission source computer 1001-G and the destination computer 1002-G in FIG. 10 are provided with a transmission rate changing unit 1006 and a data discard detection unit 1007, respectively. Is implemented using those means.
[0058]
In the network system of FIG. 10, the transmission source computer 1001-G does not transmit data for the sake of the transmission source computer 1001-G unlike the network system of FIG. As described with reference to FIG. 1, communication is started using the transmission rate changing unit 1006 while gradually increasing the transmission rate. The communication is continued for a while, and the communication of the communication service G continues normally within the desired communication band set in the data relay device 1005.
[0059]
However, the data of the communication service G transmitted exceeding the processing capacity of the data relay device 1005 cannot pass through the data relay device 1005 and is discarded. The data discard detection means 1007 of the destination computer 1002-G notifies the source computer 1001-G that the data discard has occurred. The notified source computer 1001-G reduces the transmission rate to a rate at which no data discard occurs.
[0060]
As described above, the presence / absence of surplus data that cannot be passed through the data relay device and is discarded is detected using the data discard detection unit 1007. When surplus data is generated, the transmission rate of the communication service G is controlled to a transmission rate at which data discard does not occur by using the transmission rate changing unit 1006. By doing so, it is possible to reliably prevent the communication service G from using the communication band of the sub-network 1003 before passing through the data relay device 1005 beyond the band regulated by the data relay device 1005.
[0061]
【The invention's effect】
According to the present invention, in the case of communication from a subnetwork having a wide communication band to a subnetwork having a narrow communication band, the used bandwidth of communication in the wide subnetwork is controlled to be equal to or less than that of the narrow subnetwork. Therefore, even if the network does not have a special device such as a priority control processing mechanism, it is possible to secure the bandwidth amount of the high-priority communication service and provide efficient communication only by providing the functions of data discard detection and transmission rate change. It has the effect of providing a network.
[0062]
Further, even in a network system configured between sub-networks having no difference in communication band, it is possible to temporarily receive communication data transmitted from a predetermined communication service and use a data relay device that sets a band usage to a desired value. This has the effect of enabling efficient communication.
[Brief description of the drawings]
FIG. 1 is a schematic diagram of a network system according to a first embodiment of the present invention.
FIG. 2 is a schematic diagram of a network system for explaining a conventional problem and a first embodiment.
FIG. 3 is a configuration diagram of a standby time storage table.
FIG. 4 is a configuration diagram of a packet information storage table.
FIG. 5 is a processing flowchart of a transmission rate changing unit.
FIG. 6 is a processing flowchart of a data discard detection unit.
FIG. 7 is a schematic diagram of a network system according to a second embodiment.
FIG. 8 is a configuration diagram of a data relay device according to a second embodiment.
FIG. 9 is a configuration diagram of a transfer interval determination table.
FIG. 10 is a schematic diagram of a network system according to a third embodiment.
[Explanation of symbols]
101: source computer, 102: destination (reception) computer, 105: data of communication service B, 106, 206: transmission rate changing means, 107, 207: data discard detection means, 108: data of communication service A, 201 ... Source computer, 202: destination computer, 203, 204: subnetwork, 205: data of communication service D, 208: data of communication service C, 209: discard data of communication service C, 210: network connection device, 301: standby Time storage table, 406: data transfer module, 407: transmission rate change module, 507: data discard detection module, 601: current consistent number storage area, 602: previous consistent number storage area, 603: consistent number storage table, 701: transmission Original computer, 702 ... Destination computer, 703, 7 4 Subnetwork, 705 Data relay device, 801 Data receiving means, 802 Transfer interval determining means, 803 Data transfer means, 804 Data holding means, 805 Transfer interval determining table, 1001 Source computer, 1002 ... destination computer, 1003, 1004 ... subnetwork, 1005 ... data relay device, 1006 ... transmission rate change means, 1007 ... data discard detection means.

Claims (6)

In a network system including a shared sub-network and another sub-network in which a high-priority communication service and a low-priority communication service share a band,
A network system, comprising: a bandwidth limiting unit configured to limit a communication bandwidth of the low-priority communication service to a communication bandwidth of each communication service that can flow into the shared subnetwork. In a network system configured between sub-networks having different communication bands and allocating a communication band to a communication service performed from a source computer to a destination computer,
In the case of a communication service for performing communication from a sub-network having a wide communication band to a sub-network having a narrow communication band, data discard detection means for detecting that the communication data has been discarded on the network at the destination computer; The computer is provided with transmission rate changing means for changing the transmission rate by being notified of the discard of the communication data. The transmission rate is increased when the communication data is not discarded, and the transmission is performed when the communication data is discarded. A network system characterized in that the rate is reduced. In claim 2,
A network system in which a communication service in which the communication data has been discarded is controlled to a communication band equal to or less than that of the sub-network having a narrow communication band. In a network system including a shared subnetwork in which a plurality of communication services share a band,
After receiving once the data sent from the source computer of the shared sub-network, hold the data for the time when the bandwidth usage of the communication service becomes a desired value, and then transfer the data to the other sub-network. A network system comprising a device. In claim 4,
The network system, wherein the desired value is transmitted together with the data by a transmission source computer that has transmitted the data, and is referred to by the data relay device. In a network system including a shared subnetwork in which a plurality of communication services share a band,
Data discard detection means for detecting the occurrence of discard of communication data on the subnetwork is provided to a destination computer on another subnetwork, and transmission rate change means for changing a transmission rate is provided to a source computer on the shared subnetwork. Provided, further, after once receiving the data sent from the transmission source computer, holding for a time when the bandwidth usage amount of the received communication service becomes a desired value, and transferring the data to the other network after the lapse of the time. Providing a data relay device in the shared sub-network,
When the bandwidth usage of the communication service by the transmission source computer exceeds the range limited by the data relay device, the transmission rate changing means receiving the data discard notification from the data discard detection means transmits the transmission rate. A network system characterized by changing data to avoid data discard.

Images