JP3168254B2 - Message processing method in distributed processing - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distributed processing system in which a plurality of nodes, such as personal computers or workstations, are organically connected via a network and distributed processing is performed among the plurality of nodes. In particular, the present invention relates to a message processing method in distributed processing of the system.

[0002]

2. Description of the Related Art Inter-process communication is known as a basic technique of distributed processing. Between processes that do not share a memory space or between processes executed on individual nodes separated via a network, interprocess communication as a means for transferring various information is indispensable. In such inter-process communication, information is passed in units called messages. Inter-process communication using such a message is called "message transfer". Communication methods for message transfer are classified into synchronous communication and asynchronous communication.

[0003] Synchronous communication is a method in which a transmitting process receives from a receiving process a result of processing on a receiving side regarding a message sent from the transmitting process. On the other hand, asynchronous communication is a method in which a message sent from a transmitting process is processed by a receiving process, and the processing result is not sent to the transmitting process.

When a transmitting process transmits a message to a receiving process, if the size of the message exceeds a prescribed size of the communication means, the transmitting process converts the transmitted message into a plurality of packets. It is divided and transmitted in packet units. In addition, the plurality of divided packets are each transmitted with a message identification number added thereto. This message identification number is
It is used in the receiving buffer of the receiving node so that a received packet can be distinguished from other packets, and a plurality of received packets can be combined based on the identification number to restore a message. In the case of transmitting the processing result on the receiving side to the transmitting side in the synchronous communication, the same processing regarding the packet and the identification number is performed.

There are the following problems to be solved in message delivery as described above.

First, when assigning a message identification number, for example, a 16-bit counter is used, and the counter value is changed cyclically within the range of the maximum value that can be expressed by the counter. They are numbered as numbers.

Here, when the above message is delivered, the counter circulates, and the transmitting node assigns the same identification number as the message identification number which is still being processed for reception. To its (same) receiving node. In this case, the receiving node cannot identify the packet from the plurality of packets in the reception buffer based on the message identification number and restore the message.

[0008] Therefore, the transmitting node determines whether or not the identification number is being used in the receiving node every time the message identification number is assigned, thereby avoiding the duplicate assignment of the identification number. It has become something.

However, there is a problem that performing the duplicate determination every time the identification number is assigned is inefficient from the viewpoint of improving the throughput. Alternatively, in order to avoid such duplication, for example, when a coded absolute time at the transmitting node is applied as a message identification number, the amount of information for expressing the number is expressed in microseconds. This is about 60 bits, and redundant information must be sent constantly in message transmission, which causes a reduction in processing efficiency.

[0010]

SUMMARY OF THE INVENTION Accordingly, a first object of the present invention is to provide a message processing system in distributed processing which can assign a message identification number while efficiently avoiding duplication.

A second object of the present invention is to provide a distributed processing system which improves the reliability of message transfer.

[0012]

[Means for Solving the Problems]

(1) In a message processing method in distributed processing according to the present invention, in a message processing method in distributed processing, an identification number assigned to each of a plurality of messages to be subjected to distributed processing is numbered according to a predetermined rule. When a specific time has passed since the identification number was assigned, the duplication determination regarding the identification number to be assigned at that time is prohibited, thereby prohibiting the identification number from being assigned twice. Until time elapses, the overlap determination is not performed.

With this configuration, it is possible to prevent unnecessary duplication determination from being performed.

(2) The message processing method in the distributed processing according to the present invention is the method described in (1) above,
In addition, the specific time is calculated based on a time from when an identification number is assigned in accordance with the predetermined rule to when the next identification number is assigned.

The above-mentioned "numbering an identification number in accordance with a predetermined rule" means, for example, that a number is cyclically selected from a set of predetermined numbers and is numbered as an identification number. .

(3) The message processing method in the distributed processing according to the present invention is the method described in (1) above,
When the specific time is shorter than the time required for the duplication determination, the duplication determination is not performed, and an identification number is assigned according to a rule different from the predetermined rule.

The above-mentioned "numbering an identification number according to a rule different from a predetermined rule" means that when a number is cyclically selected from a set of predetermined numbers and numbered as an identification number. This means that a specific number is temporarily excluded from the set of numbers and numbering is performed.

(4) The message processing apparatus for distributed processing according to the present invention comprises: numbering means for numbering an identification number in accordance with a predetermined rule for each of a plurality of messages to be subjected to distributed processing; Message transmitting means for transmitting, via a transmission means, a plurality of messages whose identification numbers are numbered by the numbering means, wherein the numbering means has a specific time since the predetermined identification number is numbered. When the time has passed,
By performing a duplicate determination on the identification number to be numbered at that time, the identification number is prohibited from being duplicated, and the duplicate determination is not performed until the specific time has elapsed. It is characterized by.

With this configuration, it is possible to prevent unnecessary duplication determination from being performed.

(5) The message processing device for distributed processing according to the present invention is the device described in (4) above,
In addition, the specific time is calculated based on a time from when an identification number is assigned in accordance with the predetermined rule to when the next identification number is assigned.

The above-mentioned "numbering an identification number in accordance with a predetermined rule" means, for example, that a number is cyclically selected from a set of predetermined numbers and is numbered as an identification number. .

(6) The message processing device for distributed processing according to the present invention is the device described in (4) above,
When the specific time is shorter than the time required for the duplication determination, the duplication determination is not performed, and an identification number is assigned according to a rule different from the predetermined rule.

The above-mentioned "numbering an identification number in accordance with a rule different from a predetermined rule" means that when a number is cyclically selected from a set of predetermined numbers and numbered as an identification number. This means that a specific number is temporarily excluded from the set of numbers and numbering is performed.

(7) The distributed processing system according to the present invention
In a distributed processing system for performing distributed processing by passing messages between a plurality of nodes connected via a transmission unit, the node assigns an identification number in accordance with a predetermined rule to each of the plurality of messages provided for the distributed processing. Numbering means for numbering, a plurality of messages whose identification numbers are numbered by the numbering means, a message transmitting means for transmitting via the transmission means, and a message transmitted from the transmitting means. Message receiving means for receiving, when the specific time has passed since the predetermined identification number is numbered, the numbering means performs an inquiry process to the node of the message transmission destination via the transmission means By doing so, the identification number used in the message being processed is prohibited from being numbered, and the specific time elapses During the in it is characterized in that it does not perform the query processing for the message destination node.

[0025] With this configuration, it is possible to prevent unnecessary inquiry processing from being performed on the node of the message transmission destination.

(8) The distributed processing system according to the present invention
(7) In the system according to (7), the specific time is calculated based on a time from when an identification number is assigned according to the predetermined rule to when the next identification number is assigned. It is characterized by the following.

The above-mentioned "numbering an identification number in accordance with a predetermined rule" means, for example, that a number is cyclically selected from a set of predetermined numbers and numbered as an identification number. .

(9) The distributed processing system according to the present invention
In the system according to (7), if the specific time is shorter than the time required for the inquiry processing to the other node, the inquiry processing is not performed, and the identification is performed according to a rule different from the predetermined rule. It is characterized by numbering.

The above-mentioned "numbering an identification number in accordance with a rule different from a predetermined rule" means that when a number is cyclically selected from a set of predetermined numbers and numbered as an identification number. This means that a specific number is temporarily excluded from the set of numbers and numbering is performed.

[0030]

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

FIG. 1 is a block diagram showing a schematic configuration of a distributed processing system according to an embodiment of the present invention. The distributed processing system according to the present embodiment is roughly divided into a processing element 100 as a first computer node,
Is connected to a processing element 200 as a computer node via a network line / bus 300 as transmission means, and distributed processing is performed between the first and second computer nodes. The number of computer nodes is 2 as described above.
It is optional without being limited to one. The transmission means is the bus 30.
If the value is 0, distributed processing is performed between the processing elements 100 and 200 provided in a single computer.

The processing element 100 has a memory 107, a CPU 108, and a network / bus interface 109 as hardware resources.
It has basic software (OS) 105 for controlling these hardware resources. This OS 105
Is capable of so-called multi-process processing.
As shown in the figure, a plurality of processes, that is, processes 101, 102, and 103, are generated on the same OS 105. A process is a unit in which the OS allocates CPU resources, and means the execution state of a program corresponding to the process.

The process 101 has a message processing unit 104 for performing the above-described inter-process communication by message passing. The message processing unit 104 can transmit a message to another node process or receive a message sent from another node process via the OS 105 and the network / bus interface 109.

The processing element 200 has the same configuration as the processing element 100, and has a memory 207 as hardware resources, a CPU 208, and a network / bus interface 209.
It has basic software (OS) 205 for controlling these hardware resources. This OS 205
Is capable of so-called multi-process processing.
As shown in FIG. 2, a plurality of processes, that is, processes 201, 202, and 203, are generated on the OS 205.

The process 201 has a message processing unit 204 for performing the above-mentioned inter-process communication by message passing. The message processing unit 204 can transmit a message to a process of another node or receive a message transmitted from a process of another node via the OS 205 and the network / bus interface 209.

For example, in the message transfer performed between the process 101 and the process 201, when the process 101 transmits the message 1 and the message 2 to the process 201 as shown in FIG.
01 divides the messages 1 and 2 into a plurality of packets, and
1 (for example, the case where message 1 is divided into three packets and message 2 is divided into two packets). The message processing unit 204 of the process 201 restores the message 1 and the message 2 by combining the received packets.

Here, the message processing unit 104 (the processing unit 204 is similarly configured) will be described. FIG.
3 is a block diagram illustrating a schematic configuration of the message processing unit 104. FIG.

As shown in FIG.
Is composed of a message transmission / reception request unit 40, a counter 41, a timer monitoring unit 42, a message transmission unit 43, a message reception unit 44, a buffer management table 45, and a busy check unit 46. The timer monitoring unit 42 uses the OS1
The controller 101 controls the timer 106 in the internal device 05, and is configured as a thread generated by the process 101.
Also, a message transmitting unit 43 and a message receiving unit 44
Transmits a plurality of packets constituting a transmission message via the network / bus interface 109,
Alternatively, a plurality of packets constituting the received message sent via the interface 109 are received.

FIG. 4 is a schematic diagram showing a data structure of a transmission (reception) message.

As shown in the figure, the message D is divided into data D1, D2 and D3. Then, headers Hd1, Hd2, and Hd3 are added to each of the divided data to create packets P1, P2, and P3. The header Hd includes a message identification number, a message length, a total number of packets, a packet sequence number,
Data representing the data length, the transmitting side processing element identification (child), and the transmitting side process identification (child) are stored.

The details of each of the above data are described below.

Message length: corresponds to the total size of data D1 to D3.

Total number of packets: In this case, since the packet is divided into three packets, it becomes "3".

Packet sequence number: A serial number of a packet in the message is set.

Data length: corresponds to the size of data D.

Transmission-side processing element identification: An identifier for identifying a processing element in the distributed processing system is set.

Sender-side process identification: An identifier for identifying a process in the processing element is set.

FIG. 5 is a flowchart showing the contents of the message transmission process executed by the message transmission section 43. The message transmission unit 43 includes the message transmission request unit 4
Numbering means for numbering an identification number in accordance with a predetermined rule for each of a plurality of messages for which transmission requests have been made from 0, and a plurality of messages numbered by the numbering means by the networking / bus And a message transmitting means for transmitting via a transmitting means comprising the interface 300 and the interface 109.

First, in step S1, a message identification number assigning process is performed. As a result, an identification number is assigned to the message to be transmitted. The specific contents of the message identification numbering process will be described later.

Next, in step S2, the message assigned the identification number in step S1 is divided into a plurality of data.

Next, in step S3, a packet is created by setting header information in each data obtained by the division. The division of the message and the creation of the packet are performed according to the data structure described in FIG.

Next, in step S4, each created packet is transmitted to the destination node. Such packet transmission is performed by asynchronous communication, and the destination node stores the received packet in the buffer management table and immediately returns control.

Then, in step S5, the transmission of the message ends when the transmission of all the packets corresponding to the transmission message is completed.

Here, the numbering of message identification numbers according to the feature of the present invention will be described.

FIG. 6 is a flowchart showing the contents of the message identification number assigning process. FIGS. 7A and 7B are diagrams conceptually showing the numbering of message identification numbers.

In the figure, C1 represents a current (current) message identification number, and C2 represents a message identification number in use. The numbering rules for message identification numbers are as follows in principle. That is, 0-655
Any number is cyclically numbered from a set of numbers up to 35. That is, numbers from “0 to 65535” are sequentially numbered, and then numbered again from “0”. The current message identification number is represented by a counter 41 whose value is updated in accordance with the assignment of a new identification number.

According to the flowchart of FIG. 6, first, in step S11, the counter 41 is incremented by one. Here, the initial value of the counter 41 is set to “0”, and the current counter value changes to “1” in step S11.

Next, in step S2, the mode is determined. This mode indicates whether or not the in-use check process described later is to be performed. When the check process is performed, “check” is set. When the check process is not performed, “no check” is set. Is set. Step S
In the first mode determination in 2, the mode is the initial value, that is, "no setting", and therefore, the process proceeds to step S16.

In step S16, a time t until the counter makes one round is calculated. That is, according to the numbering rules described with reference to FIG. 7, the time t until the same identification number as the message identification number assigned at a certain time is assigned is calculated. The time t gives a specific time from when a certain identification number is assigned to when the next in-use check process is started. From the time when the identification number is assigned, this specific time is given. Until the elapse, the mode is set to "no check". That is, here, the mode is set to “no check” in step S19.
In the following step S20, an instruction is issued so that the timer monitoring unit 42 monitors the elapse of the time up to the specific time. The calculation of the time t is performed based on the time Δt from when a certain identification number is assigned to when the next identification number is assigned. That is, here, t = Δt × 65
535 is calculated.

Incidentally, Δt may be a value measured in advance or may be calculated based on the number of dynamic steps of the program.

Next, in step S21, the counter value is returned and the first numbering is completed. That is, the counter value ("1") is assigned as the message identification number.

FIG. 8 is a flowchart showing the contents of the timer monitoring process by the timer monitoring unit 42. The timer monitoring unit 42 is configured as a thread as described above, and operates in parallel with the message identification number assigning process. The message identification number numbering process and the timer monitoring process can refer to the “mode” and the “time t (until overlap)” as shared data.

First, in step S201, the timer monitoring process waits until there is a monitoring start instruction in step S20. When the start instruction is received, in step S202, the "time to overlap" which can be referred to as shared data is set. Specify the value of "t" and set the interval timer. This interval timer is realized by the timer 106 in the OS 105 described above. Then, the timer monitoring process waits until a timer run-out occurs as shown in step S203.

When the time t has elapsed, the interval timer runs out of the timer, and in response to this, the timer monitoring process shifts to step S204. In step S204, the “mode” as the shared data is changed to “check”.

Here, the contents of the message identification number assigning process immediately after the occurrence of the timer run-out will be described. First, at step S11, the current counter value is +
1 is done.

Next, in step S12, since the "mode" has been changed to "check" by the timer monitoring process in the first numbering, the process proceeds to step S13.

In step S13, the in-use check process of the message identification number is performed. In this process, the process on the message sending side is changed to the process on the receiving side (here, 2
00) (message processing unit 204)
(Not shown in the figure).

FIG. 9 is a flowchart showing the processing contents in the in-use check processing section on the receiving side.

The in-use check unit of the transmission destination receives the notification of the processing element of the transmission side, the identification of the process of the transmission side, and the current counter value of the transmission side from the message processing unit on the transmission side.

In step S131, an entry having a message identification number closest to the current counter value on the transmission side is searched for in the "buffer management table entry" in which the processing element identification and the process identification match. Details of the buffer management table will be described later.

If no entry satisfying the above search conditions is found in the subsequent step S132, the process proceeds to step S134, and “Not Found” is transmitted to the transmitting process. If an entry that satisfies the above search condition is found, the message identification number of the entry is transmitted to the transmitting process.

According to such an inquiry process, it is possible to prohibit the same identification number being used in the receiving process from being assigned in the transmitting process. However, this inquiry process
Since it involves communication processing via the network line / bus 300, it should not be performed frequently.

The description returns to the flowchart of FIG.

In step S14, step S1
3 is determined, and the process proceeds to step S16 where “Not Found” is received. If the message identification number has been received, the process proceeds to step S15, and a specific time until the counter value changes to the same value as the message identification number in the inquiry result, that is, a time t until the duplication is calculated.

The calculation of the time t is performed based on the time Δt from when a certain identification number is assigned to when the next identification number is assigned. That is, here, t = Δt ×
(Counting up until the "current counter value" reaches the "message ID number of the inquiry result")
Is calculated.

Incidentally, Δt may be a value measured in advance, or may be calculated based on the number of dynamic steps of the program.

Next, in step S17, it is determined whether or not the time t is sufficiently large. If the time t is sufficiently large, the process moves to step S19. If the time t is short (short), that is, in a state as shown in FIG. 7B (the identification number of the inquiry result is C2), step S1 is executed.
Move to 8. In step S18, after changing the current counter value to the message identification number of the inquiry result, the value is incremented by 1 and the process proceeds to step S13. As a result, in FIG. 7 (b), values subsequent to C2 are checked, and values that skip C2 and near duplicates are newly numbered, thereby avoiding duplicate numbering. become.

According to such a message identification number assigning process, when the time t has elapsed since the predetermined identification number was assigned, an inquiry process is performed to the process on the receiving side via the network line / bus 300. By doing
It is possible to realize a control operation in which the use of the identification number in use in the process 200 on the receiving side is prohibited, and the inquiry process to the process 200 on the receiving side is not performed until the time t elapses.

For this reason, it is possible to efficiently avoid duplicate numbering of message identification numbers, which may occur due to a node or network failure, for example. As a result, the message sent from the transmitting node can be reliably specified at the receiving node, and the reliability of the distributed processing system regarding message delivery can be improved.

Next, a message receiving process executed by the message processing unit 44 will be described.

FIG. 10 is a flowchart showing the contents of the message receiving process executed by the message receiving unit 44. The message receiving unit 44 receives a plurality of packets corresponding to a message in response to a request from the message transmission / reception request unit 40.

The received packet is stored in a buffer secured in the memory 107. The address of this buffer is managed by the buffer management table 45.

FIG. 11 is a diagram showing the structure of the buffer management table 45.

The buffer management table 45 is a table having a plurality of entries as elements. As shown in the figure, the entry 401 includes a transmitting-side processing element identification 402, a transmitting-side process identification 403, a message identification number 404, a total number of packets 405, and the number of received packets 4
06, and a pointer 407 to the buffer.

The pointer 407 to the buffer is a pointer to the buffer area 408 in which the buffer data is stored.

First, in step S30, it is determined whether a buffer corresponding to the message exists, and if not, a new buffer is reserved in step S31. This determination is made based on the header information Hd of the packet and the contents of the buffer management table 45.
To secure a buffer means to secure a buffer area corresponding to the message length and create a new entry in the buffer management table 45.

Next, in step S32, the data of the received packet is stored in the buffer. When storing data, the “received packet number 406” in the buffer management table entry 401 is updated.

Next, in step S33, it is determined whether all packets corresponding to the received message have been received. Here, the “total number of packets 405” in the buffer management table entry 401 is compared with the “number of received packets 406” in the entry.

When the reception of all the packets is completed, the individual received packets are combined based on the packet sequence number, whereby the received message is restored.

In the following step S34, the received message is passed to the message transmission / reception request unit 40, and in step S35, the buffer management table entry 401
Is deleted (cleared), and the message receiving process ends.

According to this embodiment described above, when the time t has elapsed since the predetermined identification number was assigned, the inquiry process is performed to the process on the receiving side via the network line / bus 300. , Prohibiting the receiving process 200 from numbering the in-use identification number, and until the time t elapses.
It is possible to realize a control operation relating to the numbering of message identification numbers, such as not performing an inquiry process for 0.

Therefore, it is possible to prevent the process on the receiving side from being unnecessarily inquired. Therefore, it is possible to provide a message processing method in distributed processing that can assign a message identification number while efficiently avoiding duplication, and to provide a distributed processing system that improves reliability in message delivery.

The present invention is not limited to the above embodiment, but can be implemented with various modifications.

[0094]

As described above, according to the present invention, it is possible to provide a message processing method in distributed processing capable of assigning a message identification number while efficiently avoiding duplication, and to improve the reliability of message delivery. Distributed processing system can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a distributed processing system according to an embodiment of the present invention.

FIG. 2 is a conceptual diagram showing message transmission in message transfer.

FIG. 3 is a block diagram showing a schematic configuration of a message processing unit 104.

FIG. 4 is a schematic diagram showing a data structure of a transmission (reception) message.

FIG. 5 is a flowchart showing the contents of a message transmission process executed by the message transmission unit 43;

FIG. 6 is a flowchart showing the contents of a message identification numbering process.

FIG. 7 is a diagram conceptually showing the numbering of message identification numbers.

FIG. 8 is a flowchart showing the contents of a timer monitoring process performed by a timer monitoring unit 42;

FIG. 9 is a flowchart showing processing contents in a busy check processing unit on the message receiving side.

FIG. 10 is a flowchart showing the contents of a message receiving process executed by the message receiving unit 44;

FIG. 11 is a diagram showing a structure of a buffer management table 45.

[Explanation of symbols]

 100 processing element 101, 102, 103 process 104 message processing unit 105 OS 106 timer 107 memory 108 CPU 109 network / bus interface 200 processing element 201, 202, 203 process 204 message processing unit 205 OS 206 Timer 207 Memory 208 CPU 209 Network / bus interface 300 Network line / bus

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Takayuki Saito 7-37-7 Akasaka, Minato-ku, Tokyo Digital Vision Laboratories Co., Ltd. (56) References JP-A-6-1873111 (JP, A JP-A-4-192742 (JP, A) JP-A-63-132553 (JP, A) JP-A-1-161949 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04L 12/56 G06F 13/00 355 H04L 29/08

Claims (9)

(57) [Claims] In a message processing method in distributed processing, identification numbers assigned to each of a plurality of messages to be subjected to distributed processing are assigned in accordance with a predetermined rule, and specified after a predetermined identification number is assigned. When the time has elapsed, prohibiting the identification number from being duplicated by performing a duplicate determination on the identification number to be numbered at that time, until the specific time has elapsed, A message processing method in distributed processing, wherein the duplication determination is not performed. 2. The method according to claim 1, wherein the specific time is calculated based on a time from when an identification number is assigned according to the predetermined rule to when the next identification number is assigned. 2. A message processing method in the distributed processing according to 1. 3. The method according to claim 1, wherein when the specific time is shorter than the time required for the duplication determination, the duplication determination is not performed, and an identification number is assigned according to a rule different from the predetermined rule. 2. The message processing method in the distributed processing according to 1. 4. A numbering means for numbering an identification number in accordance with a predetermined rule for each of a plurality of messages to be subjected to distributed processing, and a plurality of messages numbered by the numbering means by said numbering means. Message transmitting means for transmitting via a transmission means, the numbering means, when a specific time has elapsed since the predetermined identification number is numbered, the identification number to be numbered at that time A message for distributed processing, wherein the duplication determination is performed to prohibit the identification numbers from being numbered repeatedly, and the duplication determination is not performed until the specific time elapses. Processing equipment. 5. The method according to claim 4, wherein the specific time is calculated based on a time from when an identification number is assigned according to the predetermined rule until the next identification number is assigned. 13. A message processing device for distributed processing according to claim 1. 6. The method according to claim 1, wherein when the specific time is shorter than the time required for the duplication determination, the duplication determination is not performed, and an identification number is assigned according to a rule different from the predetermined rule. 5. The message processing device for distributed processing according to 4. 7. A distributed processing system for performing distributed processing by passing messages between a plurality of nodes connected via a transmission unit, wherein the node performs a predetermined process on each of the plurality of messages provided for the distributed processing. Numbering means for numbering an identification number in accordance with the rules of: message transmission means for transmitting a plurality of messages, the identification numbers of which are numbered by the numbering means, via the transmission means; and Message receiving means for receiving the transmitted message, the numbering means, when a specific time has elapsed since the predetermined identification number is numbered, the message transmission destination node via the transmission means By performing inquiry processing on the message, it is prohibited to assign the identification number used in the message being processed, Until time has elapsed, the distributed processing system, characterized in that does not perform the query processing for the message destination node. 8. The method according to claim 7, wherein the specific time is calculated based on a time from when an identification number is assigned according to the predetermined rule until the next identification number is assigned. 3. The distributed processing system according to 1. 9. When the specific time is shorter than the time required for the inquiry processing for the another node, the inquiry processing is not performed, and an identification number is assigned according to a rule different from the predetermined rule. Claim 7
3. The distributed processing system according to 1.