JP2000232477A - Data communication system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform the communication of excellent data in a type where the packets are exchanged via a transmission circuit of a wide band by transmitting a control packet after the transmission interval adjustment designation information is added into a protocol control information. SOLUTION: When a control packet that designated the adjustment of the transmission interval is received, a transmission interval adjustment designation input part 76 reads the transmission interval designation information out of the protocol control information on the control packet and outputs the transmission interval that is shown by the designation information to a transmission interval adjustment part 75. The part 75 changes the value of a variable dTout (holding the transmission interval) into a newly received transmission interval. The part 75 also waits until the current time is equal to the time that is defined by adding the transmission interval dTout to the packet transmission time T. In other words, the part 75 repeats the same treatment to the next data packet when the interval dTout elapsed after a data packet is finally transmitted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data communication system for transmitting data between a plurality of data terminal devices, and more particularly to a data communication system for dividing data into packets of a predetermined length for transmission. .

More specifically, the present invention relates to a data communication system of the type in which packets are exchanged via a broadband transmission line, and in particular, for performing flow control for avoiding overflow of a received packet in a receiving buffer of a receiving device. Related to a data communication system. The flow control is
Generally, OSI (Open Systems Inte
rconnection: interconnection between open systems)
The protocol processing is performed by the transport layer of the fourth layer or the data link layer of the second layer (described later) of the reference model. In this specification, however, the flow control performed by the transport layer is particularly handled.

[0003]

2. Description of the Related Art Two or more data communication devices (DTE: Da)
Research and development relating to “data communication” for exchanging data between the terminals (Ta Terminal Equipment) have been actively conducted. The main significance of such data communication lies in sharing computer resources and sharing and distributing information.

Here, the data communication device may be a general-purpose computer system into which communication software is introduced, in addition to a dedicated communication terminal. Each data terminal device is a line termination device (DCE: Data Circuit Terrier).
It is connected to a communication medium via a “minal equipment”.

As a transmission line for connecting the data communication devices, a LAN (Local Area) is used.
Network (LAN) to LAN (Local Area Network), which are constructed by connecting LANs with dedicated lines.
Area Network), General Public Line (PST)
N), there is a wide variety of things, such as a wide area such as the Internet, and the "Internet" which has become a huge network worldwide as a result of interconnection between servers.

[0006] If the transmission line is an analog line such as PSTN, DCE is a modem (Modulator / Dele).
modulator). Also, ISDN (Int
Egrated Services Digital
Network), the DCE is a TA (Terminal Adapter) and a DSU (Digital Service).
ce Unit). In the case of a LAN, DCE is a LAN adapter (for example, an Ethernet (registered trademark) card or a token ring card).
It is.

When a specific data communication device exchanges data, if a specific transmission path is occupied, the other data communication device may be occupied by another data communication device (for example, while data or messages are interrupted). The communication device cannot communicate. Therefore, packet communication has been devised as a communication method for realizing a plurality of communications by a plurality of data terminal devices. In packet communication, data on a transmission line is a type of data communication that is handled in a predetermined length called a “packet”. That is, since the data to be transmitted is intermittent, the transmission line can be shared, and the transmission line is not occupied for a long time by the communication between the specific data communication devices. Packets are roughly classified into “data packets” for data transmission and “control packets” for the purpose of transmitting only control information.

[0008] Data communication via such a transmission line is made possible by each data communication device performing communication according to a predetermined communication protocol. The standard model of the communication protocol is OSI (Open System).
s Interconnection: Open System Interconnection Reference Model. The OSI reference model is
It is composed of seven layers of protocols: a physical layer, a data link layer, a network layer, a transport layer, a session layer, a presentation layer, and an application layer (well-known).

[0009] Among them, the role of the transport layer protocol as the fourth layer is to improve the quality of communication information.

FIG. 10 schematically illustrates data communication by the packet communication system. As shown in the figure, each of the data communication devices on the transmission side and the reception side connected via a transmission medium includes an upper layer application, a transport layer protocol entity, and a packet transfer service.

Data transfer via a transmission line is performed in lower layers,
That is, it is realized by a packet transfer service configured by a protocol of the third layer or lower in the OSI reference model.
For this purpose, the transport layer protocol entity on the transmitting side divides terminal data of an arbitrary length into pieces of predetermined length defined in the lower layer packet transfer service, and furthermore, converts protocol control information into a header (or trailer). ) Is added to the divided data and packetized. The protocol control information includes information indicating the destination data terminal device (ie, address), a serial number of the divided data, and the like. Further, the transport layer protocol entity on the receiving side removes the protocol control information from each received packet and assembles the original terminal data from each divided data (see FIG. 11).

[0012] The transport layer protocol entity is responsible for making the upper layer applications on the transmitting and receiving sides transparent. Here, "transparent" means that the data communication devices can exchange data of any length handled by the upper layer application between the data communication devices without error without being restricted by the packet format data transfer.

[0013] The transport layer protocol entity may also be used to avoid the situation where the receiving buffer becomes full and the received packet overflows.
Perform flow control.

FIG. 12 schematically shows how the flow control is performed. The data communication device on the transmitting side sequentially sends out data packets on the transmission line at predetermined transmission intervals. On the other hand, the data communication device on the receiving side temporarily stores the received data packet in the reception buffer, and then extracts the packet from the reception buffer at a predetermined extraction interval and processes the packet.

Here, since the packet extraction interval is larger than the arrival interval (in other words, the transmission interval is too short), if the amount of staying received packets exceeds the length of the receiving buffer, packet overflow occurs. Also,
Conversely, if the transmission interval is too long compared to the extraction interval, the efficiency of data communication will decrease. The flow control means adjusting a packet transmission interval on the transmission side so as to keep a balance between a packet arrival interval and a packet extraction interval.

As representative examples of the flow control, there are two methods, a "rate-based flow control method" and a "credit-based flow control method".

The rate-based flow control method is a method of performing flow control when an imbalance between a data packet arrival interval and a data packet extraction interval from a reception buffer is detected. That is, if the extraction interval is longer than the arrival interval, packet transmission on the transmitting side is suppressed and the packet arrival interval is extended. Conversely, when the extraction interval can be made shorter than the arrival interval, packet transmission on the transmitting side is promoted to shorten the packet arrival interval.

In general, the imbalance between the arrival interval and the extraction interval of data packets is detected by observing the amount of packets stored in the packet reception buffer in the data communication device on the receiving side. Based on the detection result, suppression or promotion of data packet transmission is determined,
A flow control packet including an instruction to suppress or promote packet transmission is transferred from the receiving side to the transmitting side. The transmitting side adjusts the packet transmission interval according to the instruction by the flow control packet. However, there is no need to transfer a flow control packet while the balance between the extraction interval and the arrival interval is maintained.

Japanese Patent Application Laid-Open No. Hei 5-207033 discloses a modification of the rate-based flow control system. In the method disclosed in the publication, a CPU (Central P
(Processing Unit) load is measured.
When the CPU load on the receiving side increases with the reception of the packet and reaches a predetermined value, an instruction is issued to the transmitting side to suppress the packet transmission, and the packet arrival interval is extended.

The credit-based flow control method is a method of performing flow control according to "credit", that is, the free space of a packet reception buffer. The data communication device on the receiving side notifies its credit to the transmitting side sequentially. The data communication device on the transmission side limits the number of continuous transmissions of data packets to credits or less, and subtracts the credits for the transmission packets so as to prevent continuous transmission of packets exceeding the size of the packet reception buffer. In this method, it is necessary to continuously transfer the flow control packet regardless of the state of balance between the extraction interval and the arrival interval.

FIG. 13 schematically shows the configuration of a data communication system based on the credit-based flow control method.

In FIG. 13, on the receiving side, the credit collection unit instructs the extraction of the next received packet in response to the completion of the process of assembling the divided data of each data packet into the original terminal data. Upon receipt of the completion notice, the credit is detected in conjunction with these operations. Then, the calculated credit is sent to the transmitting side as a flow control packet via the credit output unit and the transmission buffer. The transmitting side adjusts the subsequent packet transmission amount according to the received credit.

As a specific example of the credit-based flow control method, TCP (Transmission)
Control Window Protocol (Control Protocol). In this method, in the protocol information added to the packet,
The serial number (sequence number) of the packet is written. This serial number indicates the order in which the divided data included in the packet is cut out from the head of the terminal data.
On the receiving side, the serial number of the packet taken out from the packet receiving buffer is indicated by the delivery confirmation (Ackn.
(OldageNumber) is notified by a flow control packet. The transmitting side transmits data packets up to the serial number of a value obtained by adding the window (that is, the size of the packet reception buffer) to the serial number of the delivery confirmation notified by the flow control packet.

FIG. 14 schematically illustrates the format of the TCP header. As shown in the figure, it is understood that various protocol control information is written in the TCP header, but a field for writing a serial number (Sequence Number) and an acknowledgment (Acknowledgment Number) is prepared.

By the way, when comparing the rate-based flow control method and the credit-based flow control method, the present inventor judges that the former is more advantageous in a broadband network. This is mainly due to the following 2
Rely on the point. That is,

(1) In the credit-based flow control method, the amount of packets that can be transferred within the round-trip propagation delay time between the transmitting and receiving devices (ie, the theoretical transmission band) is limited by the size of the packet receiving buffer.

(2) According to the rate-based flow control method, there is no need to transfer a flow control packet while the balance between the extraction interval and the arrival interval is maintained. On the other hand, in the credit-based flow control method, it is necessary to frequently transmit flow control packets to supply credits so that credit transmission is not stopped due to depletion of credits. That is, the latter has a larger communication load for transmitting and receiving packets.

However, TCs currently in practical use
P, SSCOP (ServiceSpecific C
connection Oriented Protocol
ol), such as general-purpose transport layer protocols,
None employ rate-based flow control, relying on a credit-based flow control scheme. This is because the purpose of these general-purpose transport layer protocols is to realize transparent data transfer between various types of transmitting and receiving devices, and therefore, the hardware and operating system of the transmitting and receiving devices that serve as a platform for implementing a protocol entity. This is because there is a demand that a special mechanism should not be provided in the system (OS). That is, the following restrictions exist in the design of the transport layer protocol.

(1) In general, the packet reception buffer
Implemented by protocol entities below the transport layer. For this reason, it is not always possible to observe the state of stay of a specific transport layer packet in the packet reception buffer using the transport layer protocol. (2) The transport layer packet is usually multiplexed into a packet of a protocol layer lower than the transport layer. For this reason, it is difficult to observe the arrival interval of a specific transport layer packet to the packet reception buffer. (3) The method of measuring the CPU load as disclosed in Japanese Patent Application Laid-Open No. H5-207023 is not always feasible under any hardware and platform.

Under the above-mentioned restrictions, it is almost impossible to detect the imbalance between the arrival interval of the data packet to the receiving device and the interval of fetching the packet from the receiving buffer. Therefore, the rate-based flow control method, which should be dominant in a broadband network, could not be adopted in practice.

[0031]

SUMMARY OF THE INVENTION An object of the present invention is to provide an excellent data communication system of a type in which packets are exchanged via a wideband transmission line.

It is a further object of the present invention to provide an excellent data communication system capable of suitably performing flow control for avoiding overflow of a received packet in a reception buffer of a receiving device.

A further object of the present invention is to provide an excellent flow control which can suitably perform flow control based on detection of an imbalance between a data packet arrival interval and a packet fetch interval from a reception buffer at a receiving side. It is to provide a data communication system.

It is a further object of the present invention to balance / imbalance the interval between arrival of data packets at the receiving device and the interval at which packets are fetched from the receiving buffer while accepting the above-described implementation constraints. It is an object of the present invention to provide a data communication system which enables detection and adopts a rate-based flow control system.

[0035]

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above problems, and a first aspect of the present invention is to provide a packet format including divided data obtained by dividing terminal data and its protocol control information. A data communication method for exchanging data between data communication devices connected via a transmission line, wherein a data communication device on a transmitting side transmits transmission interval information indicating a transmission interval of a data packet transmitted by itself to a protocol. The data packet is transmitted in the control information, and the data communication device on the receiving side transmits the control packet by including the transmission interval adjustment instruction information for instructing the adjustment of the transmission interval of the data packet in the protocol control information. A data communication method.

Here, the data communication device on the transmitting side includes:
The transmission interval adjustment instruction information included in the protocol control information of the received control packet is read, the transmission interval of the data packet is determined based on the transmission interval adjustment instruction information, and the transmission interval information indicating the determined transmission interval May be written in the protocol control information of the data packet, and the data packet may be transmitted to the data communication device on the receiving side according to the determined transmission interval.

The data communication device on the receiving side measures a data packet extraction interval from the reception buffer, reads transmission interval information in the protocol control information of the data packet, and indicates the transmission interval information. Comparing the transmission interval with the measured extraction interval to detect an imbalance between the two, determining an instruction to adjust the data packet transmission interval based on the detection, and indicating the determined transmission interval adjustment instruction. The transmission interval adjustment instruction information may be written into the protocol control information of the control packet, and the control packet may be transmitted to the data communication device on the transmission side.

If the received data packet does not exist when the data packet is taken out from the receiving buffer, the data communication device on the receiving side receives the data packet and outputs the data packet from the receiving buffer. It waits until the data can be taken out, takes out the data packet from the reception buffer, records the completion time every time the data packet is taken out from the reception buffer, and records the completion time of each data packet. The packet extraction interval.

If the received data packet does not exist at the time when the data communication device on the receiving side starts to fetch the data packet from the reception buffer, the data communication device stops the fetching of the data packet and sets the data packet.
Repeatedly starting packet extraction, if there is a received packet in the reception buffer, extract the data packet, and start the packet extraction for the data packet first and the last packet extraction. A difference between the times at which the extraction starts may be determined, and the difference may be subtracted from the interval between the previous data packet extraction completion time and the interval between the data packet extraction completion times to obtain a packet extraction interval.

The data communication apparatus on the receiving side compares the data packet transmission interval with the data packet extraction interval, and when the transmission interval becomes shorter than the extraction interval, sets the data packet transmission interval. A transmission interval adjustment instruction to be extended may be determined, and a transmission interval adjustment instruction to shorten the data packet transmission interval when the transmission interval becomes longer than the extraction interval may be determined.

The data communication apparatus on the receiving side compares the transmission interval of the data packet with the retrieval interval of the data packet, and determines that the transmission interval is shorter than the retrieval interval. If the current retrieval interval is shorter than the retrieval interval, the current retrieval interval is compared with the previous retrieval interval, and the transmission interval is set only if the current retrieval interval is longer than the previous retrieval interval. If the transmission interval adjustment instruction to expand is determined, and the transmission interval is longer than the extraction interval and the transmission interval is longer than the extraction interval in the previous comparison, the current extraction interval and the previous extraction are further determined. A transmission interval that compares the intervals and shortens the transmission interval only when the current retrieval interval is shorter than the previous retrieval interval. It is also possible to determine the section instructions.

When the transmission-side data communication apparatus starts transmitting a series of data packets, the transmission-side data communication apparatus transmits the data packets at a predetermined transmission interval, and is controlled by the reception-side data communication apparatus. After receiving the packet, the data packet may be transmitted according to the transmission interval determined based on the transmission interval adjustment instruction information written in the protocol control information of the control packet.

Further, the data communication apparatus on the transmitting side may transmit the data packet at a predetermined time after the transmission of a series of data packets is started or after the transmission interval of the data packets is changed in response to the reception of the control packet. The transmission interval of the data packets may be gradually reduced every time the number of data packets is transmitted.

Further, the data communication apparatus on the transmitting side may transmit the data packet at a predetermined time after the transmission of a series of data packets is started or after the transmission interval of the data packets is changed in response to the reception of the control packet. The transmission interval of data packets may be gradually shortened each time a data packet of the total size is transmitted.

Further, the data communication apparatus on the transmitting side has a predetermined time after the transmission of a series of data packets is started or after the transmission interval of the data packets is changed in accordance with the reception of the control packet. The transmission interval of the data packet may be gradually reduced every time.

The transmission interval information written in the protocol control information of the data packet may be an interval of the transmission time of the data packet.

The transmission interval information written in the protocol control information of the data packet may be the number of data packets transmitted per unit time.

The transmission interval information written in the protocol control information of the data packet is the total size of the data packet transmitted per unit time or the total size of the divided data included in the data packet. Is also good.

The transmission interval adjustment instruction information written in the protocol control information of the control packet may be a transmission time interval of the data packet.

The transmission interval adjustment instruction information written in the protocol control information of the control packet may be the number of data packets transmitted per unit time.

Further, the transmission interval adjustment instruction information written in the protocol control information of the control packet includes the total size of data packets transmitted per unit time, or
It may be the total size of the divided data included in the data packet.

Further, the transmission interval adjustment instruction information written in the protocol control information of the control packet may have a value indicating one of shortening, extending, and maintaining the transmission interval of the data packet.

Further, the transmission interval adjustment instruction information written in the protocol control information of the control packet includes the data packet transmission interval indicated by the transmission interval information written in the protocol control information of the data packet, and the data transmission interval from the reception buffer. It may be a value indicating a difference from the measured value of the packet extraction interval.

[0054]

In order to realize the present invention, it is necessary for the transport layer protocol to directly detect the balance / imbalance between the arrival interval of the data packet to the reception buffer and the packet extraction interval from the reception buffer. is there. this is,
It is possible to measure each of the arrival interval and the take-out interval, and to compare each measurement result.

First, the extraction interval can be measured by the transport layer protocol entity on the receiving side counting time each time a packet is extracted from the packet reception buffer, and calculating the difference in the time measurement. As a prerequisite, a clock mechanism must be installed on the implementation platform of the protocol entity.
Generally, the transport layer protocol detects packet loss based on timing of a packet reception waiting time in order to maintain transparency of data transfer even if a packet loss due to a transmission error or the like occurs. That is, there is no problem since the clocking mechanism itself is conventionally provided.

Further, the interval for taking out packets from the reception buffer is shorter than the arrival interval for data packets,
A received packet may not always exist in the reception buffer. In such a case, the time from when the transport layer protocol entity becomes ready to retrieve the packet until the packet arrives and can be retrieved from the receive buffer (ie, the waiting time) is measured and subtracted from the retrieval interval. In addition, it is possible to detect the limit of the take-out interval that can be shortened.

On the other hand, regarding the arrival interval of packets,
Due to the design constraints described in the [Prior Art] section,
It cannot be measured directly by the transport layer protocol entity. For this reason, conventionally, the packet arrival interval and the packet extraction interval are directly compared and the rate-based
Flow control could not be realized (described above). Therefore, in the present invention, instead of directly measuring the arrival intervals of data packets, the arrival intervals are roughly estimated by another method.

The data packet arrival interval means that a series of data packets transmitted from the transmitting data communication device at a predetermined interval arrives at the receiving buffer of the receiving data communication device via the transmission line. The interval (Fig. 1
2).

In Ethernet (ETHERNET: a registered trademark of XEROX Corporation of the United States), which has been widely used as a conventional narrow band network technology, when a plurality of data communication apparatuses sharing the same transmission line try to transmit packets simultaneously. A so-called "collision" occurs. For this reason, the time required from the start of transmission of a packet by the transmitting side to the completion thereof is undefined. In contrast,
In a broadband transmission line, it is becoming common to equip a packet switch so that there is no collision between packet signals, and the time required from the start of packet transmission on the transmitting side to the arrival of a packet on the receiving side is increasing. Definitive. Therefore, the transmission interval of the data packet, which can be grasped by the transport layer protocol entity on the transmission side, is almost unchanged from the time when the data packet is actually transmitted from the transmission side device and reaches the packet reception buffer on the reception side. Will be maintained. In other words, the data packet transmission interval determined by the transport layer protocol entity on the transmitting side is almost equal to the data packet arrival interval in the packet receiving buffer on the receiving side.

A data packet is generally composed of divided data obtained by dividing the original terminal data, and protocol control information (header or trailer) necessary for assembling the divided data (known).

Therefore, according to the present invention, the transport layer protocol entity on the transmitting side writes the packet transmission interval (transmission interval information) in the protocol control information, so that the transport layer protocol entity on the receiving side transmits the packet. , A packet transmission interval almost equal to the packet arrival interval is notified. On the other hand, the transport layer protocol entity on the receiving side detects the balance / imbalance between the received packet transmission interval and the packet extraction interval from the reception buffer, and calculates an adjustment amount of the transmission interval. Then, a control packet in which the transmission interval adjustment instruction information for instructing the adjustment of the transmission interval is written in the protocol control information is sent back to the transmission side. The transport layer protocol entity on the transmitting side adjusts the transmission interval in accordance with the instruction to balance the transmission interval (ie, arrival interval) of the packet with the extraction interval.

In short, according to the present invention, it is possible to detect the equilibrium / imbalance between the arrival interval of a data packet to a receiving device and the packet fetching interval from a reception buffer while accepting restrictions on implementation. This makes it possible to adopt the rate-based flow control method.

According to the present invention, there is no need to add any special mechanism to the platform for implementing the transport layer protocol entity, that is, the hardware and operating system of the data communication device for both transmission and reception, and the rate-based flow control is not required. The flow control according to the method can be realized.

In the credit-based flow control system, the limit of the theoretical transmission band is limited by the size of the reception buffer, but the present invention eliminates such a limit.

Further, according to the present invention, the frequency of transmission / reception of flow control packets is minimized (only when an imbalance occurs between the transmission interval of data packets and the retrieval interval from the reception buffer). Layer protocol
The communication load associated with the processing of the entity is reduced, leading to an improvement in data transfer throughput.

Still other objects, features and advantages of the present invention are:
It will become apparent from the following more detailed description based on the embodiments of the present invention and the accompanying drawings.

[0067]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 schematically shows a data communication system 100 provided for implementing the present invention. The data communication system 100 includes a network 10 as a data transmission medium, and two or more data communication devices (DTEs) 50A, B ... connected to the network.
One certain data communication device can transmit data to another specific data communication device.

The network 10 in this embodiment is preferably a broadband transmission line equipped with a packet switch so that there is no collision between packet signals. In such an environment of the broadband network 10, the time required from the start of packet transmission on the transmitting side to the arrival of the packet on the receiving side is deterministic, and the data packet transmission interval determined by the transport layer protocol entity on the transmitting side is , Approximately equal to the data packet arrival interval in the packet receiving buffer on the receiving side.

On the network 10, each of the data communication devices 50a... Can perform communication according to a predetermined communication protocol. The standard model of the communication protocol is OSI (Open Systems Inter
connection: open system interconnection) reference model. The OSI reference model is composed of seven layers: a physical layer, a data link layer, a network layer, a transport layer, a session layer, a presentation layer, and an application layer (described above).

Each of the data communication devices 50A, 50B,... Is generally provided with a line termination device (DCE: Data Circuit).
Terminal Equipment) 51A, B
Are connected to the network 10 via. FIG.
Although the DCE is illustrated as being separate from the data communication device, the DCE may be built in each data communication device in the form of an adapter card such as a network interface card (NIC). .

Data communication devices 50a, 50b, ... are dedicated communication terminals and general-purpose computer systems (workstations, personal computers, etc.) on which communication protocol layers are mounted.
Computer or the like). Data communication device 50
Is a CPU (Central Processing U)
nit) 61, a memory 62, a display 63a as a user interface, a keyboard 63b and a mouse 63c, a hard disk drive (HDD) 64a as an external storage device, a floppy (registered trademark) disk drive (FDD) 64b and CD-
A ROM drive 64c and the like are provided.

In order to configure the data communication device 50, many electric circuits and the like other than those shown in FIG. 1 are required. However, since these are well known to those skilled in the art and do not constitute the gist of the present invention, they are omitted in this specification. Also, it should be noted that only a part of the connections between the respective hardware blocks in the drawings is illustrated in order to avoid complicating the drawings (for example, the CPU 61
Generally, each peripheral device is connected via a bus instead of a local connection.)

FIG. 2 shows the configuration of the data communication device 50.
FIG. 2 shows a block diagram illustrated from the viewpoint of function realization. The data communication device on the transmitting side is arranged on the left side in the figure, and the data communication device on the receiving side is arranged on the right side in the figure. The two data communication devices are interconnected via a network 10 as a transmission line, and exchange data packets and control packets. Also, in FIG. 2, modules that need to be added or changed for the present invention are represented by bold blocks, but it should be understood that these are a part of components of the transport layer protocol entity. . Hereinafter, each block will be described.

The data division processing section 71 on the transmission side and the data assembly processing section 84 on the reception side both constitute a part of a transport layer protocol entity. However, since these may be conventional ones and do not require any particular improvement for realizing the present invention, they will not be described further in this specification.

Each of the transmission buffers 74 and 87 and the reception buffers 77 and 81 is a component of a packet transfer service which is a lower layer of the transport layer. In this lower layer packet transfer service, a packet output from the transport layer protocol entity is transmitted via transmission buffers 74 and 87 in synchronization with the clock of the transmission line. The reception buffers 77 and 81 receive the packet in synchronization with the clock of the transmission line and output the packet to the upper transport layer protocol entity.

In the data communication device on the transmission side, the transmission interval adjustment instruction input unit 76 extracts the control packet sent from the reception side from the reception buffer 77 of the packet transfer service, which is the lower layer, and The transmission interval adjustment instruction information held in the protocol control information is read.

FIG. 3 schematically shows a format of a control packet transmitted from the data communication device on the receiving side. The protocol control information of the control packet has a field in which the “transmission interval adjustment instruction information” to be written by the receiving-side transport layer protocol entity to the transmitting-side transport layer protocol entity is prepared. Please check the points. The transmission interval adjustment instruction information includes information on an instruction to adjust the transmission interval of the data packet on the transmitting side.

The transmission interval adjusting section 75 extracts the transmission interval adjustment instruction information from the transmission interval adjustment instructing section 76 and determines the transmission interval of the data packet based on the information.

The transmission interval writing section 72 writes transmission interval information indicating the transmission interval of the data packet given from the transmission interval adjusting section 75 in the protocol control information of the transmission packet.

The packet output unit 73 adjusts to transmit a data packet at every determined transmission interval.

FIG. 4 schematically shows the format of a data packet transmitted from the data communication apparatus on the transmitting side. The protocol control information of the data packet has a field in which “transmission interval information” to be written by the transmitting transport layer protocol entity to the receiving transport layer protocol entity is prepared. I want to be confirmed.

On the other hand, in the data communication device on the receiving side,
The extraction interval measuring unit 88 is a functional module that measures an extraction interval of a data packet from the reception buffer 81. That is, upon receiving the assembly completion notification from the data assembly processing unit 84, the extraction interval measurement unit 88 instructs the packet input unit 82 to extract a data packet. Every time the extraction completion notification is received from the packet input unit 82, the elapsed time from the time when the extraction completion notification is received for the data packet extracted earlier is obtained, and this is used as the extraction interval by the interval adjustment determination unit 85. Output.

When receiving an instruction to extract a data packet from the extraction interval measuring unit 88, the packet input unit 82 extracts a data packet from the reception buffer 81 of the packet transfer service, which is a lower layer. The packet input unit 82 also stores received data in the reception buffer 81.
If the packet does not exist, the process waits until the subsequent data packet arrives, and outputs a completion notification to the extraction interval measuring unit 88 when the extraction of the data packet is completed.

The transmission interval reading section 83 reads the transmission interval information held in the protocol control information (see FIG. 4) of the received packet, and determines the transmission interval indicated by the transmission interval information. The information is passed to the adjustment determining unit 85.

The interval adjustment judging unit 85 compares the extraction interval provided from the extraction interval measuring unit 88 with the transmission interval provided from the transmission interval reading unit 83, and
It is determined whether it is necessary to adjust the packet transmission interval. When it is determined that the transmission interval of the data packet needs to be adjusted, a transmission interval adjustment instruction is output to the transmission interval adjustment instruction output unit 86.

Upon receiving the transmission interval adjustment instruction, the transmission interval adjustment instructing unit 86 creates a control packet (see FIG. 3) for instructing the adjustment of the transmission interval of the data packet, and transmits the lower layer packet transfer. Output to the transmission buffer 87 of the service.

In the above description, the transmission time interval of the data packet is used as the transmission interval and the transmission interval adjustment instruction information, but the present invention is not limited to this.
For example, as transmission interval information, the number of data packets transmitted per unit time, the total size of data packets transmitted per unit time,
It is also possible to use the total size of the divided data included in the packet.

Similarly, the transmission interval adjustment instruction information includes
In addition to the data packet transmission time interval described above, the number of data packets transmitted per unit time, or the total size of data packets transmitted per unit time and the divided data included in the data packets Can be used. Further, as the transmission interval adjustment instruction information, an instruction in the form of "reducing", "extending", or "maintaining" the transmission interval of the data packet may be expressed by flag setting / releasing. .

Next, a procedure for transmitting and receiving a packet by the transport layer protocol entity according to the present embodiment will be described in detail.

FIG. 5 is a flowchart showing an operation procedure in the data communication device on the receiving side. However,
In the figure, T is a variable that holds the packet fetching completion time, and Told is a variable that holds the previous packet fetching completion time. Each variable T and Told is
Used by the take-out interval measurement unit 88. Also, S
Is a variable that holds the state of equilibrium between the previous packet arrival interval and the extraction interval, dTin is a variable that holds the extraction interval, dTold is a variable that holds the previous extraction interval, and dTout is a variable that holds the transmission interval. Variable. Each variable S, dTin, dTold, and dTo
ut is used by the interval adjustment determining unit 85. Hereinafter, each step will be described.

When the transfer of a series of divided data constituting the terminal data is started, each variable is initialized (step S11). That is, the take-out interval measurement unit 88 sets the value of the variable Told to the current time, the interval adjustment determination unit 85 sets the value of the variable S to Under, that is, an underflow state, and sets the variable dTold to zero.

Then, the data communication device on the receiving side repeatedly executes the processing procedure described below until transfer of a series of divided data is completed.

First, the extraction interval measuring unit 88 instructs the packet input unit 82 to extract data packets from the reception buffer 81, and waits for a packet extraction completion notification (step S12). At this time, the variable T holds the current time.

Next, when the completion of data packet extraction is notified from the packet input section 82, the current time at this point is obtained and stored in a variable T, and the current data packet extraction completion time T Subtract the packet extraction completion time Told to determine the data packet extraction interval. The obtained take-out interval is output to the interval adjustment determining unit 85.

The packet extracted by the packet input unit 82 is separated into protocol control information and divided data, and the divided data is output to the data assembling unit. The transmission interval reading unit 83 outputs the transmission interval indicated by the transmission interval information in the protocol control information to the interval adjustment determining unit 85.

The interval adjustment judging unit 85 holds the take-out interval given from the take-out interval measuring unit 88 as a variable dTin, and holds the transmission interval given from the transmission interval reading unit 83 as a variable dTout (step S13). .

Next, the interval adjustment judging section 85 sets the variable dT
in and dTout are compared in magnitude (step S1).
4).

If the variable dTin is larger, that is, if the extraction interval is longer than the data packet transmission interval, the branch “Ye” in decision block S14 is made.
s ", and refers to the value of the variable S holding the previous equilibrium state (step S21).

When the value of S is other than Over, that is, in the underflow state or the balanced state, the interval adjustment determining unit 85 determines that it is necessary to extend the data packet transmission interval, The value "Over" is substituted for the variable S (step S24).

On the other hand, when the value of S is Over, that is, when the overflow has already occurred in the previous time, the control packet for instructing the data communication device on the transmission side to extend the data transmission interval is redundantly transmitted. So that the previous removal interval dTold and the current removal interval dTin
Are compared with each other (step S22).

If the current take-out interval dTin is larger, that is, if the present take-out interval is longer than the previous take-out interval, the interval adjustment determining section 85 needs to extend the data packet transmission interval. Judge,
The transmission interval adjustment instruction information is output to the transmission interval adjustment instruction output unit 86 and transferred to the transmission side as a control packet (step S23). The transmission interval adjustment instruction information is
The value of the packet extraction interval dTin may be used. If the current take-out interval dTin is smaller, that is, the current take-out interval is shorter than the previous take-out interval dTin,
The interval adjustment determination unit 85 determines that it is not necessary to extend the transmission interval of the data packet, skips the next step S23, and does not dare to transmit the control packet.

As a result of the decision block S14 and the next decision block S15, if the variable dTin is smaller,
That is, if the extraction interval is shorter than the data packet transmission interval, the branch of the decision block S15 is performed.
The process proceeds to "Yes", and refers to the value of the variable S holding the previous equilibrium state (step S31).

When the value of S is other than Under, that is, in the overflow state or the balanced state, the interval adjustment determining unit 85 determines that it is necessary to reduce the data packet transmission interval, and sets the variable The value "Under" for S
Is substituted (step S34).

On the other hand, if the value of S is Under, that is, if an underflow has already occurred in the previous time, a control packet instructing the data communication device on the transmitting side to shorten the data transmission interval should be transmitted redundantly. So that the previous removal interval dTold and the current removal interval dTi
n is compared in magnitude (step S32).

If the current take-out interval dTin is smaller, that is, if the present take-out interval is shorter than the previous take-out interval, the interval adjustment judging section 85 determines that it is necessary to shorten the data packet transmission interval. Judge,
The transmission interval adjustment instruction information is output to the transmission interval adjustment instruction output unit 86 and transferred to the transmission side as a control packet (step S33). The transmission interval adjustment instruction information is
The value of the packet extraction interval dTin may be used. If the current take-out interval dTin is larger, that is, if the present take-out interval is longer than the previous take-out interval,
The interval adjustment determination unit 85 determines that it is not necessary to reduce the transmission interval of the data packet, skips the next step S33, and does not dare to transmit the control packet.

In steps S23 and S33, the transmission interval adjusting time output unit 86 writes the transmission interval adjustment instruction information received from the interval adjustment determining unit 85 into the protocol control information, and creates a control packet. The control packet is passed to the transmission buffer 87 of the packet transfer service, which is the lower layer, and sent out to the transmission side.

When the completion of the data assembling process is notified after the above-described processing procedure, the extraction interval measuring section 88 saves the value of the current packet extraction completion time T to a variable Told, and also outputs the current packet extraction interval dTin. Is saved in the variable dTold (step S16). And
Returning to step S12, the packet input unit 82 is instructed to take out the next data packet from the reception buffer.

If the received packet does not exist in the reception buffer 81 at the time when the retrieval of the data packet is started, the standby state is set until the subsequent data packet is received and can be retrieved. When the data communication device on the receiving side operates according to the processing procedure shown in FIG. 5, the transmission interval instructed by the receiving side may include the waiting time. In other words, there is a possibility that the receiving side instructs the transmitting side to take out an interval longer than the actually achievable extracting interval. Therefore, it is preferable that the data communication device on the transmission side includes a mechanism that operates autonomously so as to gradually reduce the transmission interval.

In the transmitting-side data communication apparatus described below, the determination is made after the transmission of a series of divided data constituting the terminal data is started or based on the transmission interval adjustment instruction information in the protocol control information of the control packet. After the data packet transmission interval is changed to the set value, every time a predetermined number of data packets are transmitted, the data packet transmission interval is gradually reduced. However,
The timing for autonomously adjusting the data packet transmission interval is not limited to such a form. Alternatively, the transmission interval may be autonomously adjusted, for example, every time a data packet of a predetermined total size is transmitted, or each time a predetermined time elapses.

FIG. 6 shows, in the form of a flowchart, an operation procedure in the data communication device on the transmitting side. However,
In the figure, T is a variable that holds the packet transmission time, dTout is a variable that holds the transmission interval, and N
Is a variable that holds the number of transmission interval continuations. Each variable T,
dTout and N are variables used in the transmission interval adjusting unit 75. Further, dTinit is the initial transmission interval, Nmax is the maximum number of transmission interval continuations, and dTdec is the transmission interval shortening time. These are preset constant values, and are used by the transmission interval adjusting unit 75. Hereinafter, each step will be described.

At the time when a series of divided data constituting the terminal data is transferred, the transmission interval adjusting unit 75 initializes the value of the transmission interval dTout to the initial transmission interval dTinit and sets the value of the transmission interval continuation number N to the initial transmission interval dTinit. Initialize to zero (step S41).

The data communication device on the transmitting side repeatedly executes the processing procedure described below until a series of divided data transfer is completed.

First, the transmission interval adjusting section 75 outputs the value of the current transmission interval held in the variable dTout to the transmission interval writing section 72. The transmission interval writing unit 72 writes the given value of dTout in the transmission interval information field in the protocol control information of the transmission packet, and outputs the value to the packet output unit 73 (Step S).
42).

The transmission interval adjusting section 75 instructs the packet output section 73 to transmit a data packet, and obtains the current time and stores it in a variable T. The packet output unit 73 outputs the data packet passed from the transmission interval writing unit 72 to the transmission buffer 74 of the packet transfer service, which is the lower layer, in response to the data packet transmission instruction. After outputting the data packet, the transmission interval adjuster 75 increments the transmission interval continuation number N by 1 (step S42).

Next, the transmission interval adjusting unit 75 compares the transmission interval continuation number N with the maximum transmission interval continuation number Nmax (step S43). If N does not exceed Nmax, the next step S44 is skipped, and if N exceeds Nmax, the process proceeds to the next step S44 and the transmission interval dTo
In addition to subtracting a predetermined transmission interval reduction time dTdec from ut, the transmission interval continuation number N is reset to zero. That is, in step S44, the transmission interval adjusting unit 75 shortens the transmission interval by the transmission interval reduction time dTdec every time the data packet transmission at the same transmission interval is performed Nmax times. In this way, the transmission interval is adjusted autonomously.

When the control packet instructing the adjustment of the transmission interval is received (step S45), the transmission interval adjustment instruction input unit 76 reads the transmission interval adjustment instruction information in the protocol control information of the control packet, and reads the information. Is output to the transmission interval adjusting unit 75.
The transmission interval adjuster 75 changes the value of the variable dTout to the newly received transmission interval and resets N to zero (step S46).

The transmission interval adjusting unit 75 waits until the current time reaches a time obtained by adding the transmission interval dTout to the packet transmission time T (step S47). That is, the transmission interval dTout since the last transmission of the packet
After elapses, the process returns to step S42, and the same processing as described above is repeatedly performed on the packet.

FIG. 7 schematically shows the configuration of a receiving-side data communication device according to another embodiment. In this embodiment, when the packet input unit 82 attempts to extract a data packet from the reception buffer 81, the reception buffer 81
If there are no data packets in the
Instead of waiting for the arrival of the packet, an unsuccessful extraction notification is immediately output to the extraction interval measuring unit 88.

When the packet input unit 82 returns not the extraction completion notification but the extraction complete notification instead of the extraction completion notification, the extraction interval measuring unit 88 again issues a data packet extraction instruction.

In the embodiment shown in FIG. 7, since the interval for taking out data packets from the receiving buffer is shorter than the interval for arriving data packets, the received data buffer always exists in the receiving buffer 81. If not, a particularly suitable operation is performed. In other words, the transport layer protocol entity on the receiving side
The time from when the packet can be extracted to when the data packet arrives and can be extracted from the reception buffer 81 can be excluded from the measured value of the data packet extraction interval. This can determine the shortest achievable data packet retrieval interval.

Next, the operation realized by the transport layer protocol entity according to the embodiment shown in FIG. 7 will be described.

FIG. 8 is a flowchart showing an operation procedure in the receiving data communication apparatus according to this embodiment. However, in the figure, T is a variable holding the packet extraction completion time, Told is a variable holding the previous packet extraction completion time, and T0 and T1 are variables holding the packet extraction start time. The variables T and Told, T0, T1 are used by the extraction interval measuring unit 88. S is a variable that holds the state of equilibrium between the previous packet arrival interval and the extraction interval, dTin is a variable that holds the extraction interval,
dTold is a variable that holds the previous retrieval interval,
dTout is a variable that holds the transmission interval. The variables S, dTin, dTold, and dTout are used by the interval adjustment determining unit 85. Hereinafter, each step will be described.

At the time when the transfer of a series of divided data constituting the terminal data is started, each variable is initialized (step S51). That is, the take-out interval measurement unit 88 sets the value of the variable Told to the current time, the interval adjustment determination unit 85 sets the value of the variable S to Under, that is, an underflow state, and sets the variable dTold to zero.

The data communication device on the receiving side repeatedly executes the processing procedure described below until transfer of a series of divided data is completed.

First, when instructing the packet input section 82 to extract a packet from the reception buffer, the extraction interval measuring section 88 holds the current time in variables T0 and T1. Then, it waits for a notification that the packet has been taken out or not (step S53). The time at which the notification of the completion of the packet extraction or the notification of the completion is received is stored in the variable T.

If the notification from the packet input unit 82 is a notification of unsuccessful take-out, the process proceeds to branch "No" in the decision block S54, where the value of the variable T1 is updated to the notification reception time T, and It instructs the unit 82 to take out the data packet again (step S59). The extraction interval measurement unit 88 repeats this processing procedure while receiving the extraction completion notification from the packet input unit 82.

On the other hand, if the notification from the packet input unit 82 is the completion notification of the extraction, the judgment block S5
The process advances to the next step S55 from the branch “Yes” of Step 4.

In step S55, a difference between the extraction completion time is obtained by subtracting the previous data packet extraction completion time Told from the current data packet extraction completion time T, and the difference between the variable T1 and the variable T0 is obtained from the difference value. The extraction interval dTin is obtained by subtracting the difference. Here, the difference between the variable T1 and the variable T0 is equal to the waiting time, that is, the time from the start of the extraction of the data packet to the time when the data packet is received by the reception buffer 81 and can be extracted. The obtained take-out interval dTin is output to the interval adjustment determining unit 85.

The packet extracted by the packet input unit 82 is separated into protocol control information and divided data, and the divided data is output to the data assembling processing unit. The transmission interval reading unit 83 outputs the transmission interval indicated by the transmission interval information in the protocol control information to the interval adjustment determining unit 85.

The interval adjustment judging section 85 holds the take-out interval given from the take-out interval measuring section 88 as a variable dTin, and holds the transmission interval given from the transmission interval reading section 83 as a variable dTout (step S55). .

Next, the interval adjustment judging section 85 sets the variable dT
in and dTout are compared in magnitude (step S5).
6).

If the variable dTin is larger, that is, if the extraction interval is longer than the data packet transmission interval, the branch “Ye” in the decision block S56 is made.
s ", and refers to the value of the variable S holding the previous equilibrium state (step S61).

When the value of S is other than Over, that is, in an underflow state or a balanced state, the interval adjustment determining unit 85 determines that it is necessary to extend the data packet transmission interval, The value "Over" is substituted for the variable S (step S64).

On the other hand, when the value of S is Over, that is, when the overflow has already occurred in the previous time, the control packet for instructing the data communication device on the transmission side to extend the data transmission interval is duplicated. So that the previous removal interval dTold and the current removal interval dTin
Are compared with each other (step S62).

If the current take-out interval dTin is larger, that is, if the present take-out interval is longer than the previous take-out interval, the interval adjustment judging section 85 determines that it is necessary to extend the data packet transmission interval. Judge,
The transmission interval adjustment instruction information is output to the transmission interval adjustment instruction output unit 86 and transferred to the transmission side as a control packet (step S63). The transmission interval adjustment instruction information is
The value of the packet extraction interval dTin may be used. If the current take-out interval dTin is smaller, that is, the current take-out interval is shorter than the previous take-out interval dTin,
The interval adjustment determining unit 85 determines that it is not necessary to extend the transmission interval of the data packet, skips the next step S63, and does not dare to transmit the control packet.

As a result of the decision block S56 and the next decision block S57, if the variable dTin is smaller,
That is, when the extraction interval is shorter than the data packet transmission interval, the branch of the decision block S57 is performed.
The process proceeds to "Yes", and refers to the value of the variable S that holds the previous equilibrium state (step S71).

When the value of S is other than Under, that is, in the overflow state or the balanced state, the interval adjustment determining unit 85 determines that it is necessary to shorten the data packet transmission interval, and The value "Under" for S
Is substituted (step S74).

On the other hand, if the value of S is Under, that is, if the underflow has already occurred in the previous time, a control packet for instructing the transmitting data communication apparatus to shorten the data transmission interval must be transmitted in duplicate. So that the previous removal interval dTold and the current removal interval dTi
n is compared in magnitude (step S72).

If the current take-out interval dTin is smaller, that is, if the present take-out interval is shorter than the previous take-out interval, the interval adjustment determining section 85 determines that it is necessary to shorten the data packet transmission interval. Judge,
The transmission interval adjustment instruction information is output to the transmission interval adjustment instruction output unit 86 and transferred to the transmission side as a control packet (step S73). The transmission interval adjustment instruction information is
The value of the packet extraction interval dTin may be used. If the current take-out interval dTin is larger, that is, if the present take-out interval is longer than the previous take-out interval,
The interval adjustment determining unit 85 determines that it is not necessary to reduce the data packet transmission interval, skips the next step S73, and does not dare to transmit the control packet.

In steps S63 and S73, the transmission interval adjustment instruction output unit 86 writes the transmission interval adjustment instruction information received from the interval adjustment determination unit 85 into the protocol control information, and creates a control packet. The control packet is passed to the transmission buffer 87 of the packet transfer service, which is the lower layer, and sent out to the transmission side.

When the completion of the data assembling process is notified after the above-described processing procedure, the extraction interval measuring section 88 saves the value of the current packet extraction completion time T to a variable Told, and simultaneously extracts the current packet extraction interval dTin. Is saved in the variable dTold (step S58). And
The process returns to step S52, and instructs the packet input unit 82 to take out the next packet.

FIG. 9 is a flowchart showing an operation procedure in the receiving data communication apparatus according to this embodiment. However, in the figure, T is a variable holding the packet transmission time, and dTout is a variable holding the transmission interval. The variables T and dTout are variables used in the transmission interval adjusting unit 75. Hereinafter, each step will be described.

When a series of divided data constituting the terminal data starts to be transferred, the transmission interval adjusting section 75 initializes the value of the transmission interval dTout to the initial transmission interval (step S81).

The data communication device on the transmitting side repeatedly executes the processing procedure described below until a series of divided data transfer is completed.

First, the transmission interval adjusting section 75 outputs the value of the current transmission interval held in the variable dTout to the transmission interval writing section 72. The transmission interval writing unit 72 writes the given value of dTout in the transmission interval information field in the protocol control information of the data packet to be transmitted, and outputs the value to the packet output unit 73. (Step S82).

The transmission interval adjusting section 75 instructs the packet output section 73 to transmit a data packet, obtains the current time, and stores it in a variable T. In response to the data packet transmission instruction, the packet output unit 73 outputs the data packet passed from the transmission interval writing unit 72 to the transmission buffer 74 of the packet transfer service, which is the lower layer (step S82).

When the control packet instructing the adjustment of the transmission interval is received (step S83), the transmission interval adjustment instruction input section 76 reads the transmission interval adjustment instruction information in the protocol control information of the control packet, and receives the information. The indicated transmission interval is output to the transmission interval adjusting unit 75. The transmission interval adjusting unit 75 changes the value of the variable dTout to the newly received transmission interval (Step S84).

The transmission interval adjusting section 75 waits until the current time reaches a time obtained by adding the transmission interval dTout to the packet transmission time T (step S85). That is, the transmission interval d since the last transmission of the data packet
After the elapse of Tout, the process returns to step S82,
The same processing as described above is repeatedly performed on the next data packet.

In the embodiment shown in FIG. 7, at the transmission interval instructed by the receiving side, it is determined that the data packet is received by the receiving buffer 81 of the receiving side and can be taken out after the data packet is started to be taken out. This does not include the waiting time until the start time (see, for example, step S55 in FIG. 8). For this reason, the transmitting side only needs to continue transmitting data packets at the transmission interval as instructed, and there is no need to adjust the transmission interval autonomously.

<< Supplement >> The present invention has been described in detail with reference to the specific embodiments. However, it is obvious that those skilled in the art can modify or substitute the embodiment without departing from the spirit of the present invention. That is, the present invention has been disclosed by way of example, and should not be construed as limiting. In order to determine the gist of the present invention, the claims described at the beginning should be considered.

[0152]

As described above in detail, according to the present invention,
Provided is an excellent data communication method and an excellent data communication device capable of appropriately performing flow control based on detection of an imbalance between a data packet arrival interval and a packet extraction interval from a reception buffer at a receiving side. Can be.

Further, according to the present invention, it is possible to detect the equilibrium / imbalance between the arrival interval of a data packet to a receiving device and the interval of taking out a packet from a receiving buffer while accepting restrictions on mounting. Possible and rate based
A data communication method and a data communication device employing a flow control method can be provided.

According to the present invention, there is no need to add a special mechanism to the hardware and operating system of both the transmitting and receiving data communication devices which are platforms for implementing the transport layer protocol entity.
Flow control by a rate-based flow control method can be realized.

In the credit-based flow control method, the limit of the theoretical transmission band is limited by the size of the reception buffer, but the present invention eliminates such a limit.

Further, according to the present invention, the frequency of transmission / reception of flow control packets is minimized (only when an imbalance occurs between the transmission interval of data packets and the retrieval interval from the reception buffer). Layer protocol
The communication load associated with the processing of the entity is reduced, leading to an improvement in data transfer throughput.

[Brief description of the drawings]

FIG. 1 is a diagram schematically showing a data communication system 100 provided for implementing the present invention.

FIG. 2 is a block diagram illustrating the configuration of a data communication device 50 from the viewpoint of realizing functions.

FIG. 3 is a diagram schematically illustrating a format of a control packet according to the embodiment;

FIG. 4 is a diagram schematically illustrating a format of a data packet according to the embodiment;

FIG. 5 is a diagram showing an operation in a receiving-side data communication device in the form of a flowchart.

FIG. 6 is a diagram showing an operation procedure in a transmission-side data communication device in the form of a flowchart.

FIG. 7 is a diagram schematically illustrating a configuration of a receiving-side data communication device according to another embodiment.

FIG. 8 is a diagram illustrating an operation in a receiving-side data communication device according to another embodiment in the form of a flowchart.

FIG. 9 is a diagram showing, in the form of a flowchart, an operation procedure in a transmission-side data communication device according to another embodiment.

FIG. 10 is a diagram (conventional example) schematically showing data communication by a packet communication method.

FIG. 11 is a diagram (conventional example) schematically showing data communication by a packet communication method.

FIG. 12 is a diagram (conventional example) schematically showing a flow control performed by a transport layer protocol entity.

FIG. 13 is a diagram (conventional example) schematically showing a configuration of a data communication system based on a credit-based flow control method.

FIG. 14 is a diagram showing a format (conventional example) of a TCP header.

[Explanation of symbols]

Reference Signs List 10 network, 50 data communication device, 51 line terminating device, 61 CPU, 62 memory, 63 user interface, 64 external storage device, 71 data division processing unit, 72 transmission interval writing unit, 73: packet output unit, 74: transmission buffer, 75: transmission interval adjustment unit, 76: transmission interval adjustment instruction input unit, 77: reception buffer, 81: reception buffer, 82: packet input unit, 8
3 ... transmission interval reading unit, 84 ... data assembly processing unit, 8
5: interval adjustment determining unit, 86: transmission interval adjustment instruction output unit
87: transmission buffer, 88: take-out interval measuring unit, 10
0: Data communication system.

Claims (19)

[Claims] 1. A data communication system for exchanging data in a packet format including divided data obtained by dividing terminal data and protocol control information between data communication devices connected via a transmission line, comprising: The data communication device transmits the data packet by including transmission interval information indicating the transmission interval of the data packet transmitted by itself in the protocol control information, and the data communication device on the receiving side transmits the data packet. A data communication method, comprising transmitting control packets by including transmission interval adjustment instruction information for instructing interval adjustment in protocol control information. 2. The transmission-side data communication apparatus reads transmission interval adjustment instruction information included in protocol control information of a received control packet, and determines a data packet transmission interval based on the transmission interval adjustment instruction information. Determining, writing transmission interval information indicating the determined transmission interval in the protocol control information of the data packet, and transmitting the data packet to the data communication device on the receiving side according to the determined transmission interval. The data communication method according to claim 1, wherein 3. The data communication device on the receiving side measures an interval for extracting a data packet from a receiving buffer,
The transmission interval information in the protocol control information of the data packet is read, and the transmission interval indicated by the transmission interval information is compared with the measured extraction interval to detect imbalance between the two. An instruction for adjusting a packet transmission interval is determined, transmission interval adjustment instruction information indicating the determined transmission interval adjustment instruction is written in protocol control information of a control packet, and the control packet is transmitted to the data communication device on the transmission side. 2. The data communication method according to claim 1, wherein the data communication method is transmitted. 4. The data communication device on the receiving side, if a received data packet does not exist at the time of starting to retrieve a data packet from a receiving buffer, the data packet is received and the data packet is received from the receiving buffer. It waits until the data can be taken out, takes out the data packet from the reception buffer, records the completion time every time the data packet is taken out from the reception buffer, and records the completion time of each data packet. 4. The packet extraction interval according to claim 1, wherein
The data communication method according to any of the above. 5. The data communication device on the receiving side, if the received data packet does not exist at the time when the data packet is fetched from the receiving buffer, stops fetching the data packet and sets the data packet. Repeatedly starting packet extraction, if there is a received packet in the reception buffer, extract the data packet, and start the packet extraction for the data packet first and the last packet extraction. 2. A packet fetching interval, wherein a difference between times at which fetching is started is obtained, and the difference is subtracted from the interval between the previous data packet fetching completion time and the interval between the data packet fetching completion times to obtain a packet fetching interval. 3. The data communication method according to any one of 3. 6. The data communication device on the receiving side, comprising:
The packet transmission interval is compared with the data packet extraction interval, and when the transmission interval becomes shorter than the extraction interval, a transmission interval adjustment instruction to extend the data packet transmission interval is determined. The transmission interval adjustment instruction for shortening the data packet transmission interval is determined at the time when the length of the transmission packet becomes longer.
The data communication method according to any one of 3, 4, and 5. 7. The data communication device on the receiving side, comprising:
Comparing the packet transmission interval with the data packet retrieval interval, if the transmission interval is shorter than the retrieval interval, and if the transmission interval was shorter than the retrieval interval in the previous comparison, the current retrieval is further performed. Compare the interval with the previous extraction interval, determine the transmission interval adjustment instruction to extend the transmission interval only when the current extraction interval is longer than the previous extraction interval, and If the transmission interval is longer than the interval and the transmission interval is longer than the extraction interval in the previous comparison, the current extraction interval is compared with the previous extraction interval, and the current extraction interval is compared with the previous extraction interval. The transmission interval adjustment instruction for shortening the transmission interval is determined only when the transmission interval is shorter than the transmission interval. Or data communication system according to any one of 6. 8. The transmission-side data communication device, when starting transmission of a series of data packets, transmits data packets at a predetermined transmission interval, and controls the reception-side data communication device. After receiving the packet,
The data packet is transmitted according to a transmission interval determined based on transmission interval adjustment instruction information written in protocol control information of the control packet. 8. The data communication method according to any one of 6 and 7. 9. The data communication apparatus on the transmitting side, after starting transmission of a series of data packets, or after changing a transmission interval of data packets in response to reception of a control packet, 9. The data communication method according to claim 8, wherein the transmission interval of the data packets is gradually reduced every time the number of data packets is transmitted. 10. A data communication apparatus according to claim 1, wherein said transmission-side data communication apparatus starts transmission of a series of data packets, or changes a data packet transmission interval in response to reception of a control packet. 9. The data communication method according to claim 8, wherein the transmission interval of the data packet is gradually reduced each time a data packet having the total size of the data packet is transmitted. 11. A data communication apparatus according to claim 1, wherein said transmission-side data communication apparatus starts transmission of a series of data packets, or changes a data packet transmission interval in response to reception of a control packet. 9. The data communication method according to claim 8, wherein the transmission interval of the data packet is gradually reduced each time the time elapses. 12. The transmission interval information written in protocol control information of a data packet is an interval between transmission times of the data packet.
2. The data communication method according to any one of 1. 13. The data communication apparatus according to claim 1, wherein the transmission interval information written in the protocol control information of the data packet is the number of data packets transmitted per unit time. Data communication system. 14. The transmission interval information written in the protocol control information of a data packet may include a total size of data packets transmitted per unit time, or a data packet size.
12. The data communication system according to claim 1, wherein the data communication system has a total size of divided data included in a packet. 15. The data communication system according to claim 1, wherein the transmission interval adjustment instruction information written in the protocol control information of the control packet is an interval between transmission times of data packets. . 16. The transmission interval adjustment instruction information written in protocol control information of a control packet is the number of data packets transmitted per unit time. Data communication system. 17. The transmission interval adjustment instruction information written in the protocol control information of the control packet is a total size of data packets transmitted per unit time or a total size of divided data included in the data packet. 15. The data communication method according to claim 1, wherein: 18. The transmission interval adjustment instruction information written in the protocol control information of the control packet is a value indicating one of shortening, extending, and maintaining the transmission interval of the data packet. The data communication method according to claim 1. 19. The transmission interval adjustment instruction information written in the protocol control information of the control packet includes: a data packet transmission interval indicated by the transmission interval information written in the protocol control information of the data packet; 15. The data communication method according to claim 1, wherein the data communication method is a value indicating a difference from a measured value of a packet extraction interval.