JP2007088908A - Data communication system, transmission side computer, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide technologies for updating standby conditions, in association with a network system which automatically sets the standby conditions in packet transfer, in accordance with the situation of packet retransmission occurrence so as to be adaptive to variations in various factors related to the standby conditions. <P>SOLUTION: A packet transmission control section 105 sets on standby in accordance with standby conditions calculated based on error information generated in last transmission while a packet decomposing transmission data is transmitted at a fixed transmission rate, and when error information is further received, an error generating packet is retransmitted. When there is no record of packet retransmission, a standby condition updating section 111 shortens a standby time included in the standby conditions stored in a standby condition storage section 109. When there is a record of packet retransmission, a standby condition clear section 112 deletes the updated conditions. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a network system that automatically sets standby conditions for packet transfer, and relates to a technique for revising standby conditions by adapting to changes in various factors related to standby conditions.

  Japanese Patent No. 3495678 relates to a network system configured using the function of broadcast communication in response to complicated factors such as the network configuration, hardware capability, operating status, and data transmission conditions. A technique for easily setting conditions for performing the fastest data transmission while preventing packet leakage is disclosed.

  For example, packet leakage occurs when the limit of the processing capability is exceeded in the processing of the arithmetic unit of the receiving computer. If the transmission rate is 10 (Mbps) and the computing unit has a processing capacity of about 10 (Mips), the processing capacity will reach its limit when about 800 packets of 1.5 (Kbytes) are transmitted. Packet leakage occurs. Then, after that, it returns to a normal state in about 0.6 (sec).

  Packet leakage also occurs when the limit of the cache processing capacity to the disk device is exceeded. For example, when the transmission rate is 10 (Mbps) and the disk device 27 has a processing capability of about 1 (Mbyte / sec), processing is performed when about 680 packets having a size of 1.5 (Kbytes) are transmitted. Packet capacities occur due to capacity limitations. After that, the normal state is restored in about 1 (sec).

  The transmission side computer disclosed in the above-mentioned patent gazette is based on a standby condition (waiting condition while transmitting a packet based on error information (FIG. 17) received from the reception side computer as a result of broadcast communication). FIG. 18) is calculated and operates so as to wait according to the standby condition while transmitting a packet obtained by decomposing the transmission data.

  FIG. 19 is a diagram illustrating a standby condition of standby number 1. When the number of transfer packets is equal to WP (1), and when it is equal to an integral multiple thereof, it represents that a waiting time of WT (1) occurs.

  FIG. 20 is a diagram showing a standby condition of standby number 2. This indicates that when the number of transfer packets is equal to WP (2), a waiting time of WT (2) occurs. Although not shown, the waiting time of WT (2) also occurs when it is equal to an integral multiple of WP (2).

  FIG. 21 is a diagram illustrating a state of packet transfer related to the standby condition of standby number 1. While waiting does not occur, packets are transferred at a constant transmission rate, and change at a constant gradient. However, when the standby condition of standby number 1 is met, standby occurs. At the timing shown in FIG. 19, the packet is not transferred while the standby time elapses.

  FIG. 22 is a diagram illustrating a state of packet transfer related to the standby condition of standby number 2. Transition is performed in the same manner as in FIG. 21, and when the standby condition of standby number 2 is met, standby occurs. At the timing shown in FIG. 20, the packet is not transferred while the standby time elapses.

  In this way, by controlling packet transfer, it is possible to eliminate the cause of the failure in packet transfer and to restore the normal state. Therefore, even when data is continuously transmitted, packet leakage does not occur and packet re-transfer is not required, so that the immediacy of data used in the network system is ensured. Further, since the transmission speed is not delayed more than necessary, the fastest transmission state can be secured.

  However, complicated factors such as network configuration, hardware capabilities, operating conditions, and data transmission conditions vary. For example, there is a case where the receiving side computer in which an error has occurred is replaced with hardware having a processing unit with good processing performance, or the case where the processing capacity of the receiving side computer is improved. In this way, when the factor improves, in the conventional technology, the standby condition is not reviewed, so the standby condition for the operation of the transmitting computer is reduced despite the potential standby condition being relaxed. , Remain intact and miss the opportunity to speed up transmission.

Conversely, various factors may worsen. This is the case when a computer with poor performance is added to the receiving side. In such a case, it is preferable to reset the standby condition again rather than adding the standby condition.
Japanese Patent No. 3495678

  Therefore, the present invention solves such problems, and when the potential standby conditions are alleviated by improving various factors, the standby time included in the standby conditions is automatically shortened step by step, which is useless. The purpose is to reduce waiting time and speed up transmission.

  It is another object of the present invention to clear the set standby condition and reset the standby condition according to the current situation when the potential standby condition becomes severe due to deterioration of various factors.

A data communication system according to the present invention includes:
A data communication system comprising a sending computer and a receiving computer connected via a network,
The transmission computer includes a transmission data storage unit that stores transmission data;
A packet decomposing unit that decomposes transmission data into packets;
A packet transmission control unit for transmitting the decomposed packets to the receiving computer at a constant transmission rate via the network;
The receiver computer receives a packet from the transmitter computer via the network, and a packet reception control unit;
A packet assembly unit that generates received data by assembling received packets, and generates error information including information that identifies the number of missing packets for errors that occurred in the assembly of received packets;
A received data storage unit for storing the generated received data;
An error information transmitting unit that transmits the error information generated to the transmitting computer via the network;
The transmission computer further includes an error information collection unit that receives error information from the reception computer via the network;
A waiting condition calculation unit that identifies the number of missing packets based on the received error information, obtains a waiting time for identifying a waiting period based on the number of missing packets, and calculates a waiting condition including the waiting time; Have
The transmission data storage unit stores the next transmission data,
The packet decomposition unit decomposes the next transmission data into packets,
The packet transmission control unit waits according to the standby time included in the calculated standby condition while transmitting the packet in which the next transmission data is decomposed at a constant transmission rate, and further receives error information. , Resend the packet in error,
The transmitting computer further includes a packet retransmission record storage unit for storing a packet retransmission record,
A standby condition updating unit that shortens a standby time included in the standby condition stored in the standby condition storage unit when the number of packet retransmission records is less than a predetermined number;

  The standby condition update unit shortens the standby time included in the standby condition stored in the standby condition storage unit when no packet retransmission is recorded.

  The standby condition update unit shortens the standby time included in the standby condition stored in the standby condition storage unit when the number of packet retransmissions retransmitted within a predetermined period is less than a predetermined number. .

  Further, the standby condition update unit subtracts a predetermined shortening time from the standby time.

  The standby condition update unit may multiply the standby time by a predetermined reduction rate.

  In addition, the transmission side computer further includes a standby condition clear unit that deletes the standby condition stored in the standby condition storage unit when there are more than a predetermined number of packet retransmission records.

  In addition, the transmission side computer further includes a standby condition relaxation unit that extends a standby time included in the standby condition stored in the standby condition storage unit when the number of packet retransmission records is larger than a predetermined number. And

  In addition, the standby condition calculation unit specifies a period longer than a period required for transmitting the number of missing packets specified from the error information as a standby period.

The transmitting computer according to the present invention is:
A transmitting computer that transmits data to a receiving computer connected via a network,
A transmission data storage unit for storing transmission data;
A packet decomposing unit that decomposes transmission data into packets;
A packet transmission control unit for transmitting the decomposed packets to the receiving computer at a constant transmission rate via the network;
An error information collection unit that receives error information from the receiving computer via the network;
A waiting condition calculation unit that identifies the number of missing packets based on the received error information, obtains a waiting time for identifying a waiting period based on the number of missing packets, and calculates a waiting condition including the waiting time; Have
The transmission data storage unit stores the next transmission data,
The packet decomposition unit decomposes the next transmission data into packets,
The packet transmission control unit waits according to the standby time included in the calculated standby condition while transmitting a packet in which the next transmission data is decomposed at a constant transmission rate, and further, when error information is received, Resend the packet in error,
The transmitting computer further includes a packet retransmission recording storage unit for storing a packet retransmission record;
A standby condition updating unit that shortens a standby time included in the standby condition stored in the standby condition storage unit when the number of packet retransmission records is less than a predetermined number;

The program according to the present invention is:
A program having a transmission data storage unit for storing transmission data, and causing a computer to be a transmission computer to transmit data to a reception computer connected via a network to execute the following procedure (1) Procedure for decomposing transmission data into packets (2) Procedure for transmitting the decomposed packets to the receiving computer at a constant transmission rate via the network (3) Via the network, the procedure described above Procedure for receiving error information from the receiving computer (4) Identify the number of missing packets based on the received error information, obtain the waiting time to identify the waiting period based on the number of missing packets, and wait Procedure for calculating standby conditions including time (5) The next transmission data is decomposed into packets, and the decomposed packets are transmitted at a constant transmission rate. (6) Procedure for resending a packet in which an error has occurred when error information is received (7) Procedure for storing a packet retransmission record (8) ) A procedure for shortening the waiting time included in the waiting condition when the number of packet retransmission records is less than the predetermined number.

  According to the present invention, the standby time included in the standby condition is automatically reduced step by step, so that the operational standby condition can be brought close to the relaxed potential standby condition to reduce unnecessary standby. it can.

  In addition, when the retransmission occurs, the standby condition for operation is cleared, so that the standby condition suitable for the current situation can be set again.

Embodiment 1 FIG.
FIG. 1 is a diagram illustrating a configuration example of a network system. 1 is a transmission side computer, 2 to 6 are reception side computers, 7 is a LAN, 11 is a LAN controller, 12 is a channel control bus, 13 is a calculation unit, 14 is a main storage unit, 15 is 16 is a disk controller, 17 is a disk device, 21 is a LAN controller, 22 is a channel control bus, 23 is an arithmetic unit, 24 is a main storage unit, 25 is a system bus, and 26 is A disk controller 27 is a disk device.

  FIG. 2 is a diagram illustrating a module configuration of the transmission-side computer in the first embodiment. 101 is a system configuration value storage unit, 102 is a transmission data storage unit, 103 is a packet division unit, 104 is a broadcast address storage unit, 105 is a packet transmission control unit, 106 is an error information collection unit, and 107 is , A transmission side error information storage unit, 108 is a standby condition setting unit, 109 is a standby condition storage unit, 110 is a retransmission record storage unit, 111 is a standby condition update unit, and 112 is a standby condition clearing unit. The system configuration value storage unit 101 is configured to include an area for storing a transmission rate and a packet size, and has a function of storing data transmission conditions.

  In the present invention, elements of a retransmission recording storage unit 110, a standby condition update unit 111, and a standby condition clear unit 112 are added as compared with the prior art, and the operations of the packet transmission control unit 105 and the error information collection unit 106 are different. The packet transmission control unit 105 operates to retransmit the packet when an error occurs, and the error information collection unit 106 receives the error information from each receiving computer as an independent task, and receives the error information as needed. It operates to notify the packet transmission control unit 105.

  FIG. 3 is a diagram showing a flow of overall processing of the transmission side computer in the first embodiment. When a data transmission instruction is received by an application program (not shown) of the transmission side computer 1, the error information collection processing task is activated (S302), and the data transmission processing (S303) is started. The error information collection processing task operates independently during the data transmission processing. When the data transmission process (S303) ends, the end of the error information collection process task is checked (S304), and after the task ends, a standby condition setting process (S305) is performed.

  The error information collection processing task will be described later.

  Data transmission processing (S303) in the transmission side computer 1 will be described. FIG. 4 is a diagram showing a flow of data transmission processing in the first embodiment. This flow includes processing (S404, S405) that waits when the standby condition is met. However, since the standby condition is not generated (initial state) in the first data transmission, the processing that waits. Is not done.

  Data to be transmitted is stored in the transmission data storage unit 102. For example, it is stored in a file format. The packet division unit 103 reads transmission data from the transmission data storage unit 102 and divides it into packets. The divided packets are supplied to the packet transmission control unit 105 in order. At this time, the packet division unit 103 divides according to the packet size stored in the system configuration value storage unit 101.

  The packet transmission control unit 105 initializes the packet number (PNo), which is an internal variable, to 1 before transmitting the first packet (S401). The first packet is acquired (S402), and a packet number is assigned to the packet (403).

  In S404, it is determined whether the standby condition is met. However, since the standby condition itself has not yet been generated, the packet is transmitted without waiting (S406).

  If all the packets obtained by dividing the transmission data have been transmitted, the process ends (S407). If there is a packet that has not been transmitted yet, the packet number is incremented (S409) and the processes from S402 to S406 are repeated unless there is notification of error information (S408). At this time, the packet transmission control unit 105 repeats packet transmission according to the transmission rate stored in the system configuration value storage unit 101. In addition, the broadcast address stored in the broadcast address storage unit 104 is used to transmit a packet group to a plurality of receiving side computers.

  In this way, data in the initial state is transmitted to the receiving computer via the LAN 7.

  Next, the operation of the receiving computer 2 will be described. FIG. 5 is a diagram illustrating modules of the reception-side computer in the first embodiment. 201 is a packet reception control unit, 202 is a packet assembly unit, 203 is a reception data storage unit, 204 is a reception side error information storage unit, 205 is a transmission side computer address storage unit, and 206 is an error information transmission unit. It is.

  The packet reception control unit 201 receives a packet via the LAN 7. The packet reception control unit 201 sends the received packet to the packet assembly unit 202.

  The packet assembly unit 202 assembles a packet group and generates reception data. It also outputs error information about errors that occur in the assembly of received packets. In this embodiment, error information includes an error occurrence packet number (EPNo) which is the number of the first packet in the packet group in which leakage has occurred and an error packet number (NSPAN) which is the number of packets in the packet group in which leakage has occurred. include.

  The generated reception data is stored in the reception data storage unit 203. The output error information is stored in the receiving-side error information storage unit 204.

  FIG. 6 is a diagram illustrating a configuration example of the reception-side error information storage unit in the first embodiment. The above-mentioned error occurrence packet number (EPNo) and the number of error packets (NSPAN) are stored in association with each other.

  FIG. 7 is a diagram illustrating a configuration example of a received packet and a missing packet. In this example, the transmission side computer 1 transmits a packet group in order from the packet (701) of the packet number 1, and the reception side computer 2 receives the packet group (701 to 706) of the EPNo-1 from the packet number 1. Further, a packet group after the packet of the packet number EPNo + NSPAN is received. That is, the packet group (707 to 712) from the packet number EPNo to EPNo + NSPAN-1 is lost and packet leakage occurs.

  In this example, the number of packets continuously received from the beginning can be specified by the error occurrence packet number (EPNo) included in the error information. For this purpose, it is also effective to include other information in the error information. That is, by including other information in the error information in order to specify the number of packets continuously received from the beginning, the same effect as in the present embodiment can be obtained. For example, the number of packets received continuously from the beginning may be stored. In this example, EPNo-1 is stored. In this case, the number of continuously received packets may be counted.

  Further, the number of lost packets can be specified by the number of error packets (NSPAN) included in the error information. For this purpose, it is also effective to include other information in the error information. That is, by including other information in the error information in order to specify the number of lost packets, the same effect as in the present embodiment can be obtained. For example, the first packet number and the last packet number of the missing packet group may be stored, and the number of missing packets may be calculated based on the difference between them. In this example, EPNo + NSPAN-1 and EPNo are stored, and 1 is added to these differences to obtain the number of missing packets. Or you may memorize | store the packet number of the last packet of the packet group received continuously from the beginning, and the packet number of the packet received next. In this example, EPNo-1 and EPNo + NSPAN are stored, and 1 is subtracted from these differences to obtain the number of missing packets.

  The error information transmission unit 206 transmits the error information stored in the reception side error information storage unit 204 with the transmission side computer address stored in the transmission side computer address storage unit 205 as a destination.

  Next, the operation of the error information collection processing task in the transmission side computer 1 will be described. The transmission side computer 1 starts an error information collection processing task in S302 of FIG. FIG. 8 is a diagram showing a flow of error information collection processing in the first embodiment.

  After the data transmission process is completed, the confirmation of error information reception is repeated until a predetermined wait time elapses (that is, until a time-out occurs) (S801 to S803). When the error information collection unit 106 receives the error information (S803), the error information is notified to the packet transmission control unit 105 (S804). For example, the error information collection unit 106 writes the error information in the common storage area of the transmission side computer 1, and the packet transmission control unit 105 reads the common storage area as needed to perform the notification operation. Thereafter, the received error information is stored in the transmission-side error information storage unit 107 (S805).

  When the error information collection unit 106 receives the error information again, the received error information is added to the transmission side error information storage unit 107.

  If time-out occurs (S802), the error information in the transmission-side error information storage unit 107 is sorted in the order of the error occurrence packet number (EPNo), and the process ends.

  As shown in FIG. 17, in the transmission side error information storage unit, the error occurrence packet number (EPNo) included in the received error information is associated with the number of error packets (NSPAN), and the error has occurred. The address (RADR) of the receiving computer is stored. These error information are assigned error numbers. The received error information is sorted in the order of error occurrence packet numbers (EPNo).

  As shown in FIG. 3, when the error information collection task ends, the transmission side computer 1 performs a standby condition setting process (S305). FIG. 9 is a diagram showing a flow of standby condition setting processing in the first embodiment.

  First, the error number X which is an internal variable is initialized to the error number of the first error information stored in the transmission-side error information storage unit 107 (S901). In this example, it is initialized to 1.

  The standby condition setting unit 108 reads error information corresponding to the error number X from the transmission-side error information storage unit 107 (S902).

The standby condition setting unit 108 obtains the number of standby start packets (WP) based on the error occurrence packet number (EPNo) included in the read error information (S903). The number of waiting start packets (WP) is the number of packets that serve as a condition for starting waiting in the next data transmission. The number of standby start packets (WP) is obtained by the following equation.
WP (X) = EP (X) -MP
Here, MP is the number of packets that can be reserved. MP is 0 or more, and the larger the value is, the less packet leakage occurs. The MP may be unique in the system or variable.

  That is, the standby start time is when the number of packets received continuously from the beginning or a smaller number of packets is transmitted.

  The calculated standby start packet number (WP) is stored in the standby condition storage unit 109 (S904).

Further, the standby condition setting unit 108 obtains a standby time (WT) based on the number of error packets (NSPAN) included in the read error information (S905). The waiting time (WT) is a time for waiting in the next data transmission. The waiting time (WT) is obtained by the following equation.
WT (X) = NSPAN (X) × MTUT
Here, the MTU transmission time (MTUT) is a value obtained by the packet size / transmission rate. By multiplying NSPAN (X) by the MTU transmission time (MTUT), it is possible to obtain the time required to transmit the lost packet, that is, the time when the packet could not be received. It is also effective to provide a margin for this waiting time. For example, when this waiting time is multiplied by a value of 1 or more or a positive value is added, packet leakage is less likely to occur.

  That is, the waiting time (WT) is a time required for transmitting the number of packets that are missing or a longer time.

  The calculated standby time (WT) is stored in the standby condition storage unit 109 (S906).

  If there is error information that has not yet been processed, the error number is incremented, and the processing from S902 to S906 is repeated for the next error information.

  When the processing is completed for all error information (S907), the standby conditions are sorted in the order of the number of standby start packets (S909). Here, the sorting is performed in order to execute in order from the standby condition with the smallest number of standby start packets. In this embodiment, since the error information in the transmission-side error information storage unit 107 is already sorted by the error occurrence packet number (S806), each standby condition is not necessarily the number of standby start packets, even though the process of S909 is not necessarily performed. Generated and stored in order.

  A configuration example of the standby condition storage unit is as shown in FIG. The waiting conditions are assigned a waiting number, the waiting start packet number (WP) is associated with the waiting time (WT), and are sorted and stored in the order of the waiting start packet number (WP).

  An operation in which the transmission side computer 1 transmits the next transmission data will be described. The transmission side computer 1 operates again according to the flow chart of the data transmission process shown in FIG.

  Data to be transmitted is stored in the transmission data storage unit 102. This data need not be the same as the previously transmitted data. The processing from S401 to S403 is the same as in the previous data transmission. Since the transmission rate and the packet size are stored in the system configuration value storage unit 101, the transmission conditions are the same as those in the previous data transmission.

  In S404, the packet transmission control unit 105 reads the standby condition stored in the standby condition storage unit 109 and determines whether to start standby. Specifically, when the packet number (PNo) of the packet to be transmitted is equal to any one of the waiting start packet numbers (WP), it is determined that the waiting is started. Further, it is determined that the standby is started also when the packet number (PNo) is equal to any integral multiple of the number of standby start packets (WP).

  The packet transmission control unit 105 reads the waiting time (WT) stored corresponding to the number of waiting start packets (WP) that is adapted as a condition for starting waiting, and continues waiting until the time elapses. That is, the packet transmission control unit 105 determines a waiting period based on the waiting time (WT), and continues the waiting during that period. Even when the number of waiting start packets (WP) is equal to an integer multiple, the waiting time (WT) stored corresponding to the number of waiting start packets (WP) is read and the same operation is performed. To do.

  When the waiting time (WT) elapses, the pending packet is transmitted (S406). Thereafter, the packet transmission is repeated following the processing of S407 and S409.

  By repeating the above-described processing, a standby state suitable for the standby condition can be inserted while data is transmitted at a constant transmission rate. The manner in which data transmission is performed while inserting a standby state will be described with reference to the drawings.

  The standby condition of standby number 1 is as shown in FIG. When the number of transfer packets is equal to WP (1), and when it is equal to an integral multiple thereof, it represents that a waiting time of WT (1) occurs.

  The standby condition of standby number 2 is as shown in FIG. This indicates that when the number of transfer packets is equal to WP (2), a waiting time of WT (2) occurs. Although not shown, the waiting time of WT (2) also occurs when it is equal to an integral multiple of WP (2).

  The state of packet transfer related to the standby condition of standby number 1 is as shown in FIG. While waiting does not occur, packets are transferred at a constant transmission rate, and change at a constant gradient. However, when the standby condition of standby number 1 is met, standby occurs. At the timing shown in FIG. 19, the packet is not transferred while the standby time elapses.

  The state of packet transfer related to the standby condition of standby number 2 is as shown in FIG. Transition is performed in the same manner as in FIG. 21, and when the standby condition of standby number 2 is met, standby occurs. At the timing shown in FIG. 20, the packet is not transferred while the standby time elapses.

  In this way, by controlling packet transfer, it is possible to eliminate the cause of failure in packet transfer and to restore the normal state. Therefore, even when data is continuously transmitted, packet leakage does not occur and packet re-transfer is not required, so that the immediacy of data used in the network system is ensured. Further, since the transmission speed is not delayed more than necessary, the fastest transmission state can be secured.

  The above-mentioned standby conditions reflect complicated factors such as network configuration, hardware capability, operation status, and data transmission conditions (transmission rate, packet size, transmission data amount). It is not always effective when the factors change. For example, when the hardware capability is improved, it is assumed that no error occurs even if the standby time is shortened.

  In the following, a description will be given of a standby condition update process in which the standby time of the standby conditions is shortened step by step until an error occurs. The standby condition update process operates as an independent task.

  This waiting condition update process is performed when the packet transmission control unit 105 receives notification of error information in the data transmission process shown in S303 of FIG. 3 and retransmits the packet and records the retransmission. It is assumed. Packet retransmission will be described with reference to FIG.

  As described above, in the data transmission process, packets are sequentially transmitted until all packets are transmitted, continuously or while waiting. In the meantime, when notification of error information is received (S408), the error occurrence packet number included in the error information is set to PNo (S410), and the retransmission record is written in the retransmission record storage unit 110 (S411). The packet is retransmitted (S402 to S406). The retransmission record includes a retransmission time and an error occurrence packet number. As the retransmission time, the current time is acquired from the clock unit in the computer, and the current time is used. This time can also include the date and time.

  In addition, even after sending all packets, until a predetermined error monitoring period elapses, whether or not error information is notified is monitored, and if error information notification is received, it is included as described above. The error occurrence packet number is set to PNo (S410), the retransmission record is written in the retransmission record storage unit 110 (S411), and the packet is retransmitted (S402 to S406).

  In this way, when an error occurs during the data transmission process, the packet is retransmitted and a retransmission record is held. Based on this process, the standby condition update process will be described.

  FIG. 10 is a diagram showing a flow of standby condition update processing in the first embodiment. This process is repeated with a waiting time. In the process of waiting for the review of the waiting condition (S1001), the process waits for the waiting time to elapse. Then, the retransmission record is read from the retransmission record storage unit 110 (S1002). If there is no re-recording, the process returns to waiting for the next waiting time (S1001).

  Note that during the processing of S302 to S305 in FIG. 3, since the standby condition is used, the error information processing task is operating or the data transmission processing is operating in the processing of waiting for the standby condition review (S1001). Or if the waiting condition setting process is running, if any of these processes is running, wait until all of those processes are not running It is desirable to do.

  If there is a retransmission record (S1003), the following processing is repeated for each standby condition stored in the standby condition storage unit 109 (S1004). The standby time included in the standby condition is read (S1005), the standby time is shortened (S1006), and the reduced standby time is written back to the standby time of the standby condition (S1007).

  As a method for shortening the waiting time, for example, a predetermined shortening time is subtracted from the waiting time. If the predetermined shortening time is 5 msec, 5 msec is subtracted from 640 msec which is WT (1), and new waiting time WT (1) is set to 635 msec. Similarly, 5 msec is subtracted from 3200 msec which is WT (2), and new waiting time WT (2) is set to 3195 msec. Alternatively, the waiting time is multiplied by a predetermined reduction rate. If the predetermined shortening rate is 99%, 640 msec, which is WT (1), is multiplied by 0.99, and a new standby time WT (1) is set to 634 msec. Similarly, WT (2), 3200 msec, is multiplied by 0.99 to give a new standby time WT (1) of 3168 msec. Moreover, you may set shortening time and a shortening ratio for every range of waiting time. For example, the range is divided into 0 to 500 msec, 500 to 1000 msec, 1000 to 2000 msec, and the shortening time is set to 5 msec, 10 msec, and 20 msec, or set to 0.99, 0.98, and 0.97, respectively. To do. As the waiting time indicated by the range is larger, in addition to the method of extending the shortening time as described above, a method of conversely reducing the time can be considered. Further, as the waiting time indicated by the range becomes larger, a method of conversely enlarging is conceivable in addition to the method of reducing the shortening rate as described above.

  If there is no re-recording, the standby condition is repeatedly shortened every time the standby condition review wait time elapses.

  By the above processing, when the waiting condition is relaxed due to improvement of factors related to waiting conditions, such as replacement of the receiving computer with hardware with excellent performance, the new situation will be adapted to the new situation. The operation standby condition stored in the standby condition storage unit can be relaxed.

  On the other hand, if the standby condition factor is getting worse due to deterioration of the factors related to the standby condition, such as a hardware receiving computer with inferior performance being added to the network, or the standby condition storage unit by the above processing If the operation standby condition stored in the table is too relaxed, an error occurs and retransmission is performed. In such a case, it is desirable to set the standby condition again.

  Next, processing for clearing the standby condition when an error occurs and retransmission is performed will be described. The waiting condition clear process operates as an independent task. That is, it operates in parallel with them regardless of the processing operations of FIG. 3 and the operations of other processing tasks.

  FIG. 11 is a diagram showing a flow of standby condition clear processing in the first embodiment. This process is also repeated with a waiting time. In the process of waiting for the review of the waiting condition (S1101), the process waits for the waiting time to elapse. Then, the retransmission record is read from the retransmission record storage unit 110 (S1102). If there is no re-recording, the process returns to waiting for the next waiting time (S1101).

  In the process of waiting for a review of the standby condition (S1101), it is checked whether the error information processing task is operating, the data transmission process is operating, or the standby condition setting process is operating. When any one of them is operating, it is desirable to wait until all of the processes are not operating, as in S1001 described above.

  If there is a retransmission record (S1103), all standby conditions stored in the standby condition storage unit 109 are deleted (S1104). All the retransmission records stored in the retransmission record storage unit 110 are also deleted.

  In this way, by deleting the standby condition, it is initialized, error information is collected again as described above, and a standby condition suitable for the current situation is set based on the error information.

Embodiment 2. FIG.
In the above-described embodiment, the standby condition update process and the standby condition clear process are separate tasks, but may be operated as an integrated task including the respective processes.

  FIG. 12 is a diagram illustrating a flow of the standby condition update process and the standby condition clear process in the second embodiment. S1201 to S1203 are the same as S1001 to S1003 or S1101 to S1103. S1204 to S1208 are the same as S1004 to S1008. S1209 and S1210 are the same as S1104 and S1105.

Embodiment 3 FIG.
In the above-described embodiment, the standby condition is cleared when an error occurs and retransmission is performed, but the standby condition may be relaxed. Specifically, the waiting time is extended.

  FIG. 13 is a diagram illustrating a module configuration of the transmission-side computer according to the third embodiment. Instead of the standby condition clearing unit 112, a standby condition relaxing unit 113 is provided.

  Instead of the waiting condition clear processing task, the waiting condition relaxation processing task is operated. This task works independently as well.

  FIG. 14 is a diagram showing a flow of standby condition relaxation processing in the third embodiment. S1401 to S1403 are the same as S1101 to S1103.

  If there is a retransmission record (S1403), the following processing is repeated for each standby condition stored in the standby condition storage unit 109 (S1404). The standby time included in the standby condition is read (S1405), the standby time is extended (S1406), and the extended standby time is written back to the standby time of the standby condition (S1407).

  As a method for extending the standby time, for example, a predetermined extension time is added from the standby time. If the predetermined extension time is 5 msec, 5 msec is added to 635 msec which is WT (1), and a new standby time WT (1) is set to 640 msec. Similarly, 5 msec is added to 3195 msec which is WT (2), and new waiting time WT (1) is set to 3200 msec. Alternatively, the waiting time is multiplied by a predetermined extension rate. If the predetermined extension rate is 101%, WT (1) 634 msec is multiplied by 1.01, and a new standby time WT (1) is set to 640 msec. Similarly, by multiplying 3168 msec, which is WT (2), by 1.01, a new standby time WT (1) is set to 3200 msec. Moreover, you may set extension time and extension ratio for every range of waiting time. For example, the range is divided into 0 to 500 msec, 500 to 1000 msec, 1000 to 2000 msec, and the extended time is set to 5 msec, 10 msec, and 20 msec, or set to 1.01, 1.02, and 1.03, respectively. To do. As the waiting time indicated by the range increases, in addition to the method of extending the extension time as described above, a method of reducing the extension time can be considered. Further, as the waiting time indicated by the range increases, in addition to the method of expanding the extension ratio as described above, a method of conversely reducing is conceivable.

  In this way, if there is retransmission recording, the standby time is extended, so that the recurrence of errors can be prevented. If an error occurs as a result of shortening the waiting time in the waiting condition update process, the waiting time is extended in the waiting condition relaxation process. Can be prevented.

  Therefore, it is effective to make the extension time the same as the shortening time of the waiting time shortening process (S1006 in FIG. 10) in the waiting condition update process. It is also effective to set the extension rate to the reciprocal of the reduction rate of the waiting time reduction process.

Embodiment 4 FIG.
In the above-described embodiment, the standby condition update process and the standby condition relaxation process are separate tasks, but may be operated as an integrated task including the respective processes.

  FIGS. 15 and 16 are flowcharts showing the standby condition update process and the standby condition relaxation process in the fourth embodiment. S1501 to S1503 are the same as S1001 to S1003 or S1401 to S1403. S1504 to S1508 are the same as S1004 to S1008. S1509 to S1514 are the same as S1404 to S1409.

Embodiment 5. FIG.
In the above-described embodiment, it is determined whether or not there is retransmission recording in the standby condition update process (S1003). If there is no retransmission recording, the standby time is shortened, but whether or not the retransmission recording is a predetermined number or more. It may be determined that the waiting time may be shortened if the predetermined number is not reached. Note that determining whether or not there is retransmission recording is synonymous with determining whether or not retransmission recording is 1 or more.

  Further, it is determined whether or not there is retransmission recording in the standby condition clearing process (S1103), and when there is retransmission recording, the standby condition is cleared, but it is determined whether or not the retransmission recording is equal to or more than a predetermined number, and the predetermined number In the above case, the standby condition may be cleared and the retransmission record may be cleared.

  The same applies to the determination (S1203) in the task to be integrally processed as in the second embodiment.

  Also, in the case of the standby condition relaxation processing determination in the third embodiment (S1403), it is determined whether or not the number of retransmission recordings is equal to or greater than a predetermined number. It is also effective to do.

  The same applies to the determination (S1503) in the task to be processed integrally as in the fourth embodiment.

Embodiment 6 FIG.
As in the fifth embodiment, when determining whether or not the number of retransmission records is equal to or greater than a predetermined number, it is also possible to determine whether or not the number of retransmission records generated within a predetermined period from the current time is equal to or greater than the predetermined number. It is valid.

  The transmission side computer and the reception side computer are computers, and each element can execute processing by a program. Further, the program can be stored in a storage medium so that the computer can read the program from the storage medium.

It is a figure which shows the structural example of a network system. FIG. 3 is a diagram illustrating a module configuration of a transmission-side computer in the first embodiment. FIG. 3 is a diagram illustrating a flow of overall processing of a transmission side computer in the first embodiment. FIG. 10 is a diagram showing a flow of data transmission processing in the first embodiment. FIG. 3 is a diagram showing modules of a receiving computer in the first embodiment. 6 is a diagram illustrating a configuration example of a reception-side error information storage unit in Embodiment 1. FIG. It is a figure which shows the structural example of the received packet and the missing packet. FIG. 10 is a diagram illustrating a flow of error information collection processing in the first embodiment. FIG. 10 is a diagram showing a flow of standby condition setting processing in the first embodiment. FIG. 6 is a diagram showing a flow of standby condition update processing in the first embodiment. FIG. 10 is a diagram showing a flow of standby condition clear processing in the first embodiment. It is a figure which shows the flow of the standby condition update process in Embodiment 2, and a standby condition clear process. FIG. 10 is a diagram showing a module configuration of a transmission side computer in a third embodiment. FIG. 10 is a diagram showing a flow of standby condition relaxation processing in the third embodiment. It is a figure which shows the flow (1/2) of the standby condition update process and standby condition relaxation process in Embodiment 4. It is a figure which shows the flow (2/2) of the standby condition update process and standby condition relaxation process in Embodiment 4. It is a figure which shows the structural example of the transmission side error information storage part in a prior art. It is a figure which shows the structural example of the standby condition memory | storage part in a prior art. It is a figure showing the standby conditions of the standby number 1 in a prior art. It is a figure showing the standby conditions of the standby number 2 in a prior art. It is a figure showing the mode of the packet transfer regarding the standby condition of the standby number 1 in a prior art. It is a figure showing the mode of the packet transfer regarding the standby condition of the standby number 2 in a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Transmission side computer, 2-6 Reception side computer, 7 LAN, 11 LAN controller, 12 channel control bus, 13 arithmetic part, 14 main memory part, 15 system bus, 16 disk controller, 17 disk apparatus, 21 LAN controller, 22 Channel control bus, 23 arithmetic unit, 24 main storage unit, 25 system bus, 26 disk controller, 27 disk unit, 101 system configuration value storage unit, 102 transmission data storage unit, 103 packet division unit, 104 broadcast address storage unit, 105 Packet transmission control unit, 106 error information collection unit, 107 transmission side error information storage unit, 108 standby condition setting unit, 109 standby condition storage unit, 110 retransmission record storage unit, 111 standby condition update unit, 112 standby condition clear unit, 113 Waiting section Relaxation unit, 201 packet reception control unit, 202 packet assembly unit, 203 reception data storage unit, 204 reception-side error information storage unit, 205 transmission-side computer address storage unit, 206 error information transmitting unit.

Claims (10)

A data communication system comprising a sending computer and a receiving computer connected via a network,
The transmission computer includes a transmission data storage unit that stores transmission data;
A packet decomposing unit that decomposes transmission data into packets;
A packet transmission control unit for transmitting the decomposed packets to the receiving computer at a constant transmission rate via the network;
The receiver computer receives a packet from the transmitter computer via the network, and a packet reception control unit;
A packet assembly unit that generates received data by assembling received packets, and generates error information including information that identifies the number of missing packets for errors generated in the assembly of received packets;
A received data storage unit for storing the generated received data;
An error information transmitting unit that transmits the error information generated to the transmitting computer via the network;
The sending computer further includes an error information collection unit that receives error information from the receiving computer via the network;
A waiting condition calculation unit that identifies the number of missing packets based on the received error information, obtains a waiting time for identifying a waiting period based on the number of missing packets, and calculates a waiting condition including the waiting time; Have
The transmission data storage unit stores the next transmission data,
The packet decomposition unit decomposes the next transmission data into packets,
The packet transmission control unit waits according to the standby time included in the calculated standby condition while transmitting the packet in which the next transmission data is decomposed at a constant transmission rate, and further receives error information. , Resend the packet in error,
The transmitting computer further includes a packet retransmission recording storage unit for storing a packet retransmission record;
A data communication system comprising: a standby condition update unit that reduces a standby time included in a standby condition stored in a standby condition storage unit when the number of packet retransmission records is less than a predetermined number.   The data communication system according to claim 1, wherein the standby condition update unit shortens the standby time included in the standby condition stored in the standby condition storage unit when no packet retransmission is recorded.   The standby condition update unit shortens the standby time included in the standby condition stored in the standby condition storage unit when there are fewer than a predetermined number of packet retransmission records retransmitted within a predetermined period. The data communication system according to 1.   The data communication system according to claim 1, wherein the standby condition update unit subtracts a predetermined shortening time from the standby time.   The data communication system according to claim 1, wherein the standby condition update unit multiplies the standby time by a predetermined reduction rate.   The said transmission side computer further has a standby condition clear part which deletes the standby condition memorize | stored in the standby condition memory | storage part, when there are more packet retransmission records than predetermined number. Data communication system.   The transmission computer further includes a standby condition relaxation unit that extends a standby time included in the standby condition stored in the standby condition storage unit when the number of packet retransmission records is greater than a predetermined number. The data communication system according to claim 1.   The data communication according to claim 1, wherein the standby condition calculation unit specifies a period longer than a period required for transmitting the number of missing packets specified from the error information as a standby period. system. A transmitting computer that transmits data to a receiving computer connected via a network,
A transmission data storage unit for storing transmission data;
A packet decomposing unit that decomposes transmission data into packets;
A packet transmission control unit for transmitting the decomposed packets to the receiving computer at a constant transmission rate via the network;
An error information collection unit that receives error information from the receiving computer via the network;
A waiting condition calculation unit that identifies the number of missing packets based on the received error information, obtains a waiting time for identifying a waiting period based on the number of missing packets, and calculates a waiting condition including the waiting time; Have
The transmission data storage unit stores the next transmission data,
The packet decomposition unit decomposes the next transmission data into packets,
The packet transmission control unit waits according to the standby time included in the calculated standby condition while transmitting a packet in which the next transmission data is decomposed at a constant transmission rate, and further, when error information is received, Resend the packet in error,
The transmitting computer further includes a packet retransmission recording storage unit for storing a packet retransmission record;
A transmission-side computer, comprising: a standby condition update unit that reduces a standby time included in a standby condition stored in a standby condition storage unit when the number of packet retransmission records is less than a predetermined number. A program having a transmission data storage unit for storing transmission data, and causing a computer to be a transmission computer to transmit data to a reception computer connected via a network to execute the following procedure (1) transmission data (2) Procedure for transmitting the decomposed packet to the receiving computer at a constant transmission rate via the network (3) Error information from the receiving computer via the network (4) Based on the received error information, the number of missing packets is identified, and based on the number of missing packets, a waiting time for determining a waiting period is obtained, and a waiting condition including the waiting time is calculated. (5) Decompose the next transmission data into packets, and calculate the waiting time while transmitting the decomposed packets at a constant transmission rate (6) A procedure for resending a packet in which an error has occurred when receiving error information (7) A procedure for storing a packet retransmission record (8) A packet retransmission record is predetermined A procedure to reduce the waiting time included in the waiting condition when it is less than the number.

Images