JP2002169738A - File delivery method - Google Patents

Abstract

(57) [Summary] [Problem] It is necessary to confirm delivery as many as the number of divided data, which causes an increase in network traffic. A file delivery method for delivering a file from a transmitting station (1) to at least one or more receiving stations (2) via a network (3), wherein the file is divided into a plurality of data blocks, and A first step (14, 1) of assigning a number and transmitting from the transmitting station 1
6), a second step (18) for receiving a retransmission request from the receiving station 2 after a plurality of data blocks for one file are transmitted in the first step, and a retransmission request in the second step. If there was, based on the order number,
A third step (22) of retransmitting the requested data block from the transmitting station 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for highly reliable and efficient file distribution for receiving and copying one or more files transmitted from one computer by one or more other computers. It is about.

[0002]

2. Description of the Related Art Conventionally, techniques in such a field include:
One described in Japanese Patent Application Laid-Open No. 6-11028 is known. The conventional file distribution method described in JP-A-6-141028 is a method of managing transmission data transmitted from a transmitting station under the initiative of a receiving station. According to this conventional technique, a transmitting station performs transmission while confirming whether or not a receiving station has received transmission data. Therefore, when large data such as a file is transmitted, delivery confirmation is performed for each data. Although several methods of confirming the delivery are described, what can be said in common is a procedure of performing a delivery confirmation of the data after transmitting each data, and retransmitting the corresponding data when there is unreceived data. .

[0003]

However, in the conventional file distribution method, it is necessary to confirm the delivery by the number of divided data, and each time the delivery is confirmed, it is necessary to wait for the response by the number of receiving stations. As a result, the traffic of the network may be increased, and if the transmission confirmation from the transmitting station is missing, the transmission confirmation may not be possible. In addition, the number of receiving stations is not described because the receiving station is of the initiative type, so that it can deal only with preset receiving stations and cannot deal with dynamically increasing or decreasing receiving stations. Was.

[0004] The present invention has been made to solve these problems, and has as its object to provide a file distribution method capable of reducing traffic from the entire network by reducing responses from receiving stations. .

[0005]

A file distribution method according to the present invention is a file distribution method for distributing a file from a transmitting station to at least one or more receiving stations via a network. A first step of transmitting one file from the transmitting station by assigning a sequence number to these data blocks, and transmitting a plurality of data blocks of one file from the receiving station in the first step. A second step of receiving a retransmission request from the transmitting station, and a third step of retransmitting the requested data block from the transmitting station based on the sequence number when the retransmission request is made in the second step. And returning to the first step when there is a file to be transmitted.
And the step of:

In the first step, it is preferable to transmit file transmission start notification data before transmitting the data block, and to transmit file transmission end notification data after transmitting the data block.

In the third step, it is preferable that retransmission start notification data is transmitted before the data block is transmitted, and retransmission end notification data is transmitted after the data block is transmitted.

It is preferable that the sequence number of all retransmitted data blocks be added to the retransmission start notification data.

Further, before starting the processing of the first step,
Transmitting the job start notification data, which is a unit of file transmission, from the reception station; and transmitting the job end notification data from the reception station when there is no file to be transmitted next in the fourth step. It is preferable to further include

[0010] The method may further include, after transmitting the job start notification data, accepting a response from the receiving station that has received the job at the transmitting station, and recognizing the receiving station that receives the job with the response. Is preferred.

Further, in the second step, it is preferable to wait for a retransmission request from the receiving station for a predetermined time.

In the second step, it is preferable to wait for retransmission requests from all receiving stations that have been confirmed to receive the job.

Further, when there is a retransmission request from a part of the receiving stations confirmed to receive the job, a third step of retransmitting the data block to some of the receiving stations, and other steps. It is preferable that the second step of receiving a retransmission request to the receiving station be processed in parallel.

[0014] In the third step, it is preferable to repeat retransmission until there is no retransmission request from the receiving station.

Further, at least one of job start notification data, file transmission start notification data, file transmission end notification data, retransmission start notification data, retransmission end notification data, and job end notification data. Preferably, one set of data is transmitted a plurality of times at predetermined time intervals.

Further, the identification number of the job scheduled to be distributed in the future,
It is preferable to include a step of distributing job information including a distribution date and time, a file name, a broadcast address and the like to a predetermined IP address.

It is preferable that the method further includes a step of periodically distributing job information including an identification number, a distribution date and time, a file name, a broadcast address, and the like of a job to be distributed in the future to a predetermined IP address.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a file distribution method according to the present invention will be described below with reference to the accompanying drawings. Embodiment 1 FIG. FIG. 1 is a block diagram showing a configuration of a system to which the file distribution method according to the first embodiment is applied. In the figure, 1 is a transmitting station for transmitting a file, 2 is a receiving station for receiving a file transmitted from the transmitting station 1, and 3 is a broadcasting station capable of broadcasting at least from the transmitting station 1 to a plurality of receiving stations 2. This is a network system capable of communicating with the transmitting station 1 from a plurality of receiving stations 2. Therefore, different infrastructures may be used for the uplink and the downlink. For example, a satellite line may be used from the transmitting station 1 to the receiving station 2, and a telephone line may be used from the receiving station 2 to the transmitting station 1.

FIG. 2 is a flowchart showing an algorithm of the file distribution method according to the first embodiment. FIG. 3 is a diagram showing the flow of data according to the file distribution method according to the first embodiment over time. FIG.
Describes only the transmitting station 1 and the receiving station 2, but there may be a plurality of receiving stations 2, and therefore, the response from the other receiving station 2 is also indicated by a large arrow in the figure.

The operation of the file distribution method according to the first embodiment will be described below with reference to FIGS. First, job information is distributed from the transmitting station 1 to the receiving station 2 (step 10). The job information includes an identification number of a job to be distributed in the future, a size of the job, a distribution date and time, a file name, a broadcast address defined for each job or file, and the like. The job information is transmitted to a broadcast address determined in advance by the transmitting station 1 and the receiving station 2 (different from the broadcast address included in the job information).

The receiving station 2 can obtain sufficient information to receive a job to be distributed in the future based on the job information. Management becomes possible.
As described above, since advanced schedule management and the like that cannot be performed after the start of a job can be performed, a failure at the time of reception, such as a reception failure due to insufficient storage capacity, can be solved in advance, and a highly reliable file distribution system can be realized. Note that the receiving station 2 can also determine whether or not the receiving station 2 should receive the message by using a keyword.

Next, the transmitting station 1 sleeps until the distribution date and time is reached (step 11). When the distribution date of the job is reached, the transmitting station 1 transmits a job start notification (step 1).
2). The job start notification includes the content of the job to be distributed, for example, a job identification number, the number of files to be distributed, the size of each file, the total size of all files, and the like. By receiving the job start notification, the receiving station 2 can grasp the size of the distributed file and the like, and can determine whether or not the job has been normally received at the end of the reception.

Upon receiving the job start notification, the receiving station 2 that receives the job transmits to the transmitting station 1 that the job is to be received by receiving the job. In the reception start notification,
A job identification number, an identifier of the receiving station 2, and the like are included.
The transmitting station 1 can manage which receiving station 2 receives the job of which identification number by receiving the reception start notification transmitted from the plurality of receiving stations 2 (step 13).

The transmitting station 1 divides the first file of the job into data blocks of a predetermined size (step 1).
4). Here, the size (upper limit) of the data block varies depending on the type of network, and is set in the initial setting so as to flexibly cope with them. Following the division of the file, the transmitting station 1 transmits a file transmission start notification (step 15). The file transmission start notification includes the identification number of the job, the number of data blocks obtained by dividing the file, the size of the file to be transmitted, the file name, and the like. Receiving the file transmission start notification, the receiving station 2 makes preparations such as generating a file with the obtained file name.

Subsequently, the transmitting station 1 transmits a data block (step 16). A plurality of data blocks obtained by dividing the file into a predetermined size are transmitted one after another. As a property of the broadcast communication, the order of the data blocks transmitted by the transmitting station 1 is not guaranteed. Therefore, the job block identification number and the data block sequence number are added to the data block. The receiving station 2 arranges the data blocks received one after another according to the sequence number and stores the data blocks. When storing the data blocks, the receiving station 2 stores the missing data blocks, if any.

When all the data blocks have been transmitted, the transmitting station 1 transmits a file transmission end notification (step 17). Receiving station 2 that received the file transmission end notification
Checks if all data blocks have been received. When a missing data block is found, a retransmission request indicating that fact is transmitted to the transmitting station 1. At this time, the receiving station 2 transmits the sequence numbers of the plurality of missing data blocks as one retransmission request so as not to wastefully increase the load between the receiving station 2 and the transmitting station 1. If the receiving station 2 has received all the data blocks, it transmits a retransmission request to the transmitting station 1 indicating that retransmission is unnecessary. As described above, since the response from the receiving station 2 is limited to only the retransmission request,
Data transmission and reception between the transmitting station 1 and the receiving station 2 can be reduced. As a result, the processing of both the transmitting station 1 and the receiving station 2 can be simply realized, and the load on the network 3 can be reduced.

After transmitting the file transmission end notification, the transmitting station 1 waits for a retransmission request (step 18).
The retransmission request queuing is terminated when the acknowledgment is received from all the receiving stations 2 that have received the reception start notification from the receiving station 2 in step 13 or when a preset queuing time elapses. Here, if the responses from all the receiving stations 2 that have received the reception start notification are received, the receiving station 2 that is receiving the currently transmitting job can be specified.
It is possible to wait for a response such as a retransmission request from those receiving stations 2. Therefore, more reliable file distribution to each receiving station 2 can be realized.

When the transmitting station 1 completes the retransmission request wait, it checks whether a retransmission request has occurred (step 19). If a retransmission request has occurred, the transmission station 1 prepares to transmit the corresponding data block. (Step 20). In this preparation, the data block generated in Step 14 may be stored and reused, or Step 1
The corresponding data block may be newly generated with the same data size as that of the data block.

After the generation of the data block for retransmission is completed, the transmitting station 1 transmits a retransmission start notification (step 21). The retransmission start notification describes the sequence numbers of all data blocks to be retransmitted from now on. Therefore, since the sequence number of the data block to be retransmitted by the transmitting station 1 is known before the data block is actually retransmitted from the transmitting station 1, the receiving station 2 transmits the data block immediately before to the transmitting station 1. The fact that the retransmission request has not reached the transmitting station 1 can be known before receiving the retransmission of the data block.

In such a case, the receiving station 2 can determine the data block to be requested to be retransmitted without waiting for the retransmission end notification and issue the retransmission request.
The transmitting station 1 can prepare for the next retransmission processing in parallel with the current retransmission processing. As a result, the time until the end of the distribution of the job can be reduced. In the receiving station 2 which has already received the file, it is known that this is the start of the retransmission phase.
In steps 1 and 22, the data transmitted from the transmitting station 1 is read and discarded.

Here, U generally used for broadcast communication
Since the DP protocol does not guarantee that data will reach the destination, the retransmission request transmitted from the receiving station 2 may not reach the transmitting station 1. The ordering of all data blocks to be transmitted in the retransmission notification is performed in consideration of the nature of the UDP protocol, and the receiving station 2 uses the retransmission data to be received (expected) in the future. It can be determined whether or not all the missing data blocks can be complemented.

Therefore, the receiving station 2 can predict the missing data block without waiting for the retransmission end notification, and can immediately transmit a retransmission request for the missing data block to the transmitting station 1. For this reason, compared to a case where the list of data blocks is not added to the retransmission start notification,
Further, efficiency and reliability are increased.

Subsequently, transmitting station 1 transmits the data block for which retransmission is requested (step 22). The data blocks requested to be retransmitted by the receiving station 2 are transmitted from the transmitting station 1 one after another. The receiving station 2 stores the data block only when an unreceived data block can be received, and discards the data block when already received data has been received.

When all the data blocks requested to be retransmitted by the receiving station 2 have been transmitted, the transmitting station 1 transmits a retransmission end notification indicating the end of the retransmission phase (step 23). Thereafter, the transmitting station 1 waits for retransmission again (step 24). The receiving station 2 that has received the retransmission end notification confirms whether all the data blocks have been received.

When a missing data block is found again, a retransmission request indicating that fact is transmitted to the transmitting station 1. At this time, the receiving station 2 transmits the sequence numbers of the plurality of missing data blocks as one retransmission request so as not to wastefully increase the load between the receiving station 2 and the transmitting station 1. If the receiving station 2 has received all the data blocks, it transmits a retransmission request to the transmitting station 1 indicating that retransmission is unnecessary.

After transmitting the retransmission end notification, transmitting station 1 waits for a retransmission request (step 24). The retransmission request queuing is performed by receiving replies from all the receiving stations 2 that have made the retransmission request among the receiving stations 2 that have received the reception start notification from the receiving station 2 in step 13, or by setting a predetermined queuing time. It ends after elapse. Transmitting station 1
When the retransmission request waiting is completed, it checks whether a retransmission request has occurred (step 25), and if a retransmission request has occurred, makes preparations for transmitting the corresponding data block (step 26). .

This preparation is performed immediately after the file transmission phase.
The data block generated in the preparation for transmitting the data block in the first retransmission phase (step 20) may be stored, and this data block may be reused. May be generated again.

In the transmitting station 1, the receiving station 2 can receive the missing data by repeating the retransmission phase shown in the processing of steps 20 to 24. The number of repetitions of the retransmission phase may be until all the receiving stations 2 completely complete the file reception, or the number may be set in advance by initial setting.

As described above, by repeating the retransmission phase, the same data block can be transmitted any number of times to the receiving station 2 that could not receive the data block by retransmission of the data block. As a result, highly reliable file distribution can be realized even in the unstable and unreliable network 2 in which data blocks are easily lost.

As a result, when retransmission notifications indicating that retransmission is unnecessary are received from all the receiving stations 2 that have transmitted the reception start notification to the transmitting station 1, the transmitting station 1 can receive the file currently being transmitted at all the receiving stations 2. It is determined whether there is a file to be transmitted next in this job (step 2).
6). If there is a file to be transmitted in this job, the process returns to step 14 to transmit the next file, and the next file transmission phase is started. When transmission of all files is completed and there is no file to be transmitted, the transmitting station 1 transmits a job end notification indicating the end of the job (step 27).

As described above, in the present embodiment, the transmitting station 1
Since the file transmission from the server and the response from the receiving station 2 are executed in file units, the number of times of transmission and reception can be reduced as compared with the conventional method of performing retransmission confirmation and delivery confirmation in data block units. Also, except in a specific case, such as when it is determined that the retransmission request from the receiving station 2 has disappeared, the receiving station 2 does not make a retransmission request until the file transmission or retransmission phase is completed. It can be further reduced. As a result, the loads on the transmitting station 1 and the receiving station 2 can be reduced, and network traffic can be minimized.

Further, since the processing proceeds in file units, the progress of the transmitting station 1 can be regarded as the progress of the job, and the receiving station management such as which receiving station 2 is receiving for each job. Can be realized relatively easily, and a highly reliable system can be realized. Furthermore, since there is no need to process a response interrupt from the receiving station 2 during transmission, efficient file transfer is possible.

It is to be noted that the job information is first transmitted from the transmitting station 1 in step 10 shown in FIG. 2 and the receiving station 2 prepares for reception, but step 10 is omitted and
If a broadcast address (an address determined for each file) is set in advance in the transmitting station 1 and the receiving station 2, it is also possible to realize file distribution using only the broadcast address.

Embodiment 2 In the first embodiment, all the processes are executed as one task. However, steps 13, 18, and 24 shown in FIG. 2 are performed by a different task from other transmission processes (steps 16 and 22). By performing the processing, the retransmission request from the receiving station 2 can be performed in parallel with the transmission processing. For this reason, when a retransmission request is received from a part of the receiving stations 2 that have been confirmed to receive the job, the data blocks for some of the receiving stations 2 are not waited for a response from all the receiving stations 2. Since the step 22 of performing retransmission and the step 18 of receiving a retransmission request to the other receiving stations 2 can be processed in parallel, it becomes more efficient.

Embodiment 3 In the first and second embodiments, FIG.
In step 10, the transmitting station 1 transmits job information first and the receiving station 2 prepares for reception. However, if the receiving station 2 is newly added, the power supply is cut off for maintenance, Therefore, when the receiving station 2 is initialized, such as when the receiving station 2 is replaced with a new receiving station 2, the job information transmitted until then cannot be received.

Therefore, the process of step 10 is set as a separate task, and even during the sleep of the transmission process of step 11, the job is periodically transmitted at a predetermined time interval until immediately before the job is started. Thereby, even if the job information of the receiving station 2 is initialized for some reason, the job information to be transmitted in the future can be obtained, and immediately after the receiving station 2 starts operating and before the receiving station 2 starts operating. It becomes possible to receive a scheduled job.

Embodiment 4 Embodiments 1 to 3 assume that the receiving station 2 can always receive control data such as a job start notification, a job end notification, a file transmission start notification, a file transmission end notification, a retransmission start notification, and a retransmission end notification. However, as described in the first embodiment, the UDP protocol generally used for broadcast communication does not guarantee that data is always transmitted to a destination, and may be lost on the way. This is true for both the receiving station 2 and the transmitting station 1. Therefore, a procedure of retransmitting a data block transmitted from the transmitting station 1 is indispensable. The transmitting station 1 also transmits control data in addition to the data blocks. Therefore, job start notification, job end notification,
By transmitting control data such as a file transmission start notification, a file transmission end notification, a retransmission start notification, and a retransmission end notification a plurality of times at predetermined time intervals, loss of the control data is prevented, and job transmission and file transmission are performed. , The transition of each phase of retransmission can be grasped by the receiving station 2. As a result, the reliability of data transmission can be improved. It should be noted that the number of transmissions of each control data and the transmission interval time for transmitting a plurality of control data vary depending on the reliability of the network 3, the distance between the transmitting station 1 and the receiving station 2, and the like.

[0048]

Since the file distribution method according to the present invention is configured as described above, the following effects can be obtained. That is, since the file transmission from the transmitting station is performed in file units, the number of times of transmission and reception can be reduced as compared with the conventional method of performing retransmission confirmation and transmission confirmation in data block units. As a result, the load on the transmitting station and the receiving station can be reduced, and network traffic can be minimized. In addition, even when a plurality of files are transmitted from a transmitting station and received by a plurality of receiving stations, a highly reliable file distribution system that can simultaneously and reliably receive the received files can be realized. Since no processing is required, it can be realized with a simpler configuration.
Further, in general, in a broadcast communication using a UDP protocol or the like, it is not guaranteed that data transmitted from a transmitting station can be reliably received by a receiving station, and that it is not guaranteed that data can be received by a receiving station in the order of transmission by the transmitting station. The probability that at least one of the same control data transmitted from the station can be received by the receiving station can be increased. As a result, the current transmission status transmitted from the transmitting station can be reliably grasped by the receiving station, and as a result, highly reliable file distribution can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a system to which a file distribution method according to a first embodiment is applied.

FIG. 2 is a flowchart illustrating an algorithm of a file distribution method according to the first embodiment.

FIG. 3 is a diagram showing a flow of data by the file distribution method according to the first embodiment over time.

[Explanation of symbols]

 1 ... transmitting station, 2 ... receiving station, 3 ... network.

Claims (13)

[Claims] 1. A file distribution method for distributing a file from a transmitting station to at least one or more receiving stations via a network, comprising: dividing the file into a plurality of data blocks; 1 from the transmitting station
A first step of transmitting files, a second step of receiving a retransmission request from the receiving station after the plurality of data blocks of one file have been transmitted in the first step, A third step of retransmitting the requested data block from the transmitting station based on the sequence number when there is a retransmission request in the step; And a fourth step of returning the process to the step (a). 2. The method according to claim 1, wherein the first step includes transmitting file transmission start notification data before transmitting the data block, and transmitting file transmission end notification data after transmitting the data block. 2. The file distribution method according to claim 1, wherein: 3. The method according to claim 3, wherein the third step transmits retransmission start notification data before transmitting the data block, and transmits retransmission end notification data after transmitting the data block. The file distribution method according to claim 1 or 2, wherein 4. The file distribution method according to claim 3, wherein sequence numbers of all retransmitted data blocks are added to the retransmission start notification data. 5. A step of transmitting job start notification data, which is a unit of file transmission, from the receiving station before the processing of the first step starts, and there is no file to be transmitted in the fourth step. Transmitting the job completion notification data from the receiving station.
The file distribution method according to any one of claims 1 to 4. 6. A step of, after transmitting the job start notification data, receiving a response from the receiving station that has received the job at the transmitting station, and recognizing the receiving station that receives the job with the response. The file distribution method according to claim 5, further comprising: 7. The file distribution method according to claim 1, wherein the second step waits for a retransmission request from the receiving station for a predetermined time. 8. The file distribution method according to claim 6, wherein the second step waits for retransmission requests from all receiving stations confirmed to receive the job. 9. The third step of retransmitting a data block to some of the receiving stations when there is a retransmission request from some of the receiving stations confirmed to receive the job; 7. The file distribution method according to claim 6, wherein the second step of receiving a retransmission request to another receiving station is performed in parallel. 10. The file distribution method according to claim 1, wherein the third step repeats retransmission until there is no retransmission request from the receiving station. . 11. The job start notification data, the file transmission start notification data, the file transmission end notification data, the retransmission start notification data, the retransmission end notification data, and the job end notification 11. The file distribution method according to claim 1, wherein at least one of the data is transmitted a plurality of times at a predetermined time interval. 12. The method according to claim 1, further comprising the step of distributing job information including an identification number, a distribution date and time, a file name, a broadcast address, and the like of a job scheduled to be distributed to a predetermined IP address. 12. The file distribution method according to any one of the eleventh to eleventh aspects. 13. The method according to claim 1, further comprising a step of periodically distributing job information including an identification number, a distribution date and time, a file name, a broadcast address, and the like of a job scheduled to be distributed to a predetermined IP address. The file distribution method according to any one of claims 1 to 11.