JP2002252635A - Data communication system, data relay device and data relay method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make the most suitable route setting referring dynamic processing performance of each router comprising a data communication network. SOLUTION: A sending host A1 sends a diagnosis packet which diagnoses dynamic processing performance of each router 3 to 6 on a data distribution route 11. Each router 3 to 6 receives sequentially the diagnosis packet, writes the processing performance of the router in the diagnosis packet, and sends the diagnosis packet to the next router. Finally, the receiving host B2 receives the diagnosis packet and creates a diagnosis result packet showing the dynamic processing performance of each router 3 to 6 from the diagnosis packet. Each router 3 to 6 receives sequentially the diagnosis result packet, verifies the processing performance of the next router described in the diagnosis result packet, and if the processing performance of the next router is insufficient, for instance, changes the router 6 on the data distribution route 11 to a router 10 on another route different from the data distribution route 11 and sets a new data distribution route 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a diagnostic packet payload from a sending host to a receiving host in an IP network in which a plurality of routers are arranged between a sending host and a receiving host and a data delivery route is determined by the router. The IP address of each router existing in the middle of the route and the dynamic state information and attribute information are recorded and transferred in the (data section), and each of the middle of the route collected in the payload of the diagnostic packet arriving at the receiving host is recorded. The IP address of the router and the dynamic status information and attribute information are copied to the payload of the diagnostic result packet returned by the receiving host toward the transmitting host, and are recorded inside the diagnostic result packet returned from the receiving host to the transmitting host. The IP address and dynamic status information of each router that is used refers to attribute information when a router in the middle forwards a packet It refers to the IP address and dynamic state information and attribute information of the router existing downstream from the own router on the route from the sending host to the receiving host, and the value referred to is determined in advance and set in the own router. Compared with a certain threshold value, if the value is equal to or greater than the threshold value, data generated by dynamic changes in route information and attribute information during data transfer by changing the router that exists on the downstream route and relays packets The present invention relates to a route control method and a route information collection method for constructing an IP network system that prevents transfer delay and data loss and determines an optimal data delivery route.

[0002]

2. Description of the Related Art Conventionally, in an IP network in which a plurality of routers are arranged between a sending host and a receiving host, a data delivery route is determined by the router, and data is delivered while being relayed, a data delivery route is determined. Then, each router exchanges packets based on a routing protocol, so that one router determines the next router to relay data.

Prior art 1. "Third Edition Network Construction by TCP / IP Vol.I -Principle, Protocol, Architecture-" (Douglas Comer
According to Chapter 16 of the book, Jun Murai and Hiroyuki Kusumoto, published by Kyoritsu Shuppan Co., Ltd., routing control information protocol (RIP),
The HELLO protocol, open show test path first protocol (OSPF) is shown.

[0004] In RIP, a pair of the number of routers existing on a route to a receiving host, called a hop count, and its own IP address is determined by a RIP packet between routers located on the network. By exchanging between routers, route information is constructed inside each router. Assuming that the greater the number of routers on the route to the receiving host, the greater the network distance, and when transferring data, the router selects the next router so that the hop count to the receiving host is smaller. In FIG. 41, the host A 1001 and the host B 10
The two routes that can be used between the two routes are a first route 1008 composed of router a1003-router b1004-router c1005-router d1006 and a second route 1009 composed of router a1003-router e1007-router d1006. A route exists. When each router exchanges RIP packets based on RIP on such a network and creates routing control information inside the router, in the first route 1008, four hosts are connected between the host A 1001 and the host B 1002. The router, that is, the number of hops is 4, and the second route 1009 indicates that the host A
There are three routers between the host 1001 and the host B 1002, that is, the number of hops is three. As described above, since the RIP uses a route with a smaller number of hops, data is transferred using the second route 1009 in the example of FIG.

In the HELLO protocol, data is transferred using a path distance based on a network delay, instead of a hop count indicating the number of routers present on a path to a receiving host used in RIP. HELLO
By adding the delay information calculated from the time stamp when the protocol packet is exchanged by each router, the distance to the receiving host is expressed using the delay,
The router establishes or changes the delivery route while selecting a route with less delay. Since each component in the network model in FIG. 42 is the same as the component in FIG. 41, the description is omitted here. HELLO based on the HELLO protocol on such a network
When each router exchanges protocol packets and creates routing control information inside the router, the first route 10
08, the total delay between the host A 1001 and the host B 1002 becomes 0.3 ms, and the second path 10
At 09, the total delay between host A 1001 and host B 1002 is 0.5 ms. HELL as described above
In the example of FIG. 42, data is transferred using the first path 1008 because the O protocol uses a path with a smaller transfer delay.

In the OSPF protocol, a manager of each router located on a network defines a value called a cost in advance based on static information such as the speed of a connection interface and the bandwidth of a physical transmission path. By exchanging predefined cost information between each router, route information is built in the own router, and when transferring data, the cost is lower based on the exchanged cost information. Forward to the next router like. Each component in the network model of FIG. 43 is the same as the component of FIG.
The description here is omitted. When each router exchanges OSPF protocol packets based on the OSPF protocol on such a network and creates routing control information inside the router, the total cost between the host A 1001 and the host B 1002 in the first route 1008 Is 4
In the second route 1009, the host A10
The total cost between 01 and host B 1002 is 9.
As described above, in the OSPF protocol, a path having a lower cost is used. Therefore, in the example of FIG. 43, data is transferred using the first path 1008.

Conventional technique 2. FIG. 44 shows “Japanese Unexamined Patent Application Publication No. 8-89.
09 is a configuration diagram of a network system status diagnosis / monitoring device, which is indicated by “09. Network system status diagnosis / monitoring device”.

[0008] Here, as a target network system, a network A 1101, a network B 1102, and a network C 1103.
A system based on three wide area networks is assumed. For such a network system, a monitoring device 1111 performs diagnosis and monitoring of the system.

For this purpose, in the prior art 2, a test packet is sent over a network. For example, the device 111
2 to 1117 (devices 1112, 1114, and 1116 are relay devices), and the live / dead state of the surrounding network,
When monitoring the load state, the monitoring device 1111
The packet which returns while circulating through the arrows 1112 to 1117 is transmitted.

When each device receives a packet, it
Information on the own device such as the load, reception of the packet, transmission time and the like are written in the packet and transmitted to the next device.
It is also returned to the monitoring device 1111.

The monitoring device 1111 collects and stores information of packets returned from each device. When the information is collected, the monitoring device can detect the state of each device from the information such as the CPU load written in the packet by each device.

The behavior of the packet can be known from the reception and transmission times of the packet written by each device, and the state such as the load on the network can be estimated.

Further, when the packet is not returned normally, the location of the failure can be detected from the returned status. Since the circulating route of the test packet can be set arbitrarily, according to the prior art 2, the above-described diagnosis and monitoring can be performed at an arbitrary portion of the network system, and the test packet can be collected in a wider range of information. Accordingly, it is possible to provide an apparatus that realizes state diagnosis and monitoring of a wide range of network systems not limited to failure detection.

Prior art 3. FIG.
467, a network fault control method in a packet switching network.

The packet switching devices A1201 to D1204 are respectively provided with transmission lines 1211 to 1212.
18 and are interconnected. For such a packet-switched network, a test packet circulating in the network that does not follow the communication packet control method specified in the network is provided, and a packet indirectly transmitted by transferring the test packet from each packet switching device. All the lines between the switching device and the designated packet switching device and the operation of the packet switching device within the designated section are checked, and the test result is notified to the source of the test packet to enable efficient routing control.

Now, consider the case where the packet switching apparatus A 1201 is the test request source.

In the test request packet, the packet switching device A 1201 is set as a test request source, the packet switching device D 1204 is set as a test section designation area, and cyclic order information.

The route to the packet switching device D 1204 is as follows: packet switching device A 1201 → packet switching device D
1204, packet switching apparatus A1201 → packet switching apparatus B1202 → packet switching apparatus D1204, and packet switching apparatus A1201 → packet switching apparatus C12
03 → packet switching apparatus B1202 → packet switching apparatus D1204. In this case, the test result of the traveling destination is determined, and when an overlapping test occurs, the test at the traveling station is skipped.

The packet switching apparatus A 1201 confirms the operation of the transmission lines 1211 to 1215 and stores the test results. Next, the destination packet switching device D1 of the first route
The test request packet is transferred to the transmission line 121.
4, 1215, 1217 and 1218 are checked, and the test result is notified to the packet switching apparatus A.

If all the lines (transmission lines 1214 and 1215) to the packet switching device D or the switching device have failed and the test request packet cannot be accepted, the test request packet is not forwarded but stored to that effect. Then, move to the next route test.

In this manner, the test request packet is transferred to the destination packet switching apparatus B1202 of the second route and further to the destination packet switching apparatus C1203 of the third route.
The status of each packet switching station is grasped.

The packet switching apparatus A 1201 receives the test result notification packet of each circulating destination, and
By recognizing the situation of the entire network starting from 1201, the routing control is created and changed.

[0023]

In the prior art 1, a data delivery path formed by a sending host, a receiving host, and a plurality of routers located therebetween is based on a routing protocol exchanged between the routers. It is determined using packets. When RIP is used as a routing protocol, a data delivery route formed between a sending host and a receiving host is determined so that the number of routers existing on the route is smaller. No consideration is given to the bandwidth or transfer speed of the physical transmission path, and no consideration is given to the data delivery processing capacity of each router or the amount of data delivered from or delivered to other paths. Therefore, for example, the load on the delivery process at the router in the middle of the determined data delivery route is extremely high, and despite the delay in the data transfer process within the router, such a route is used. In some cases, resulting in a delay in data transfer between the sending host and the receiving host. Also, if there is a delay in the data transfer process in the router,
The data waiting to be processed is temporarily stored in a buffer provided inside the router. Since only a finite amount of transferred data can be stored in this buffer, if the buffer overflows, the transferred data will be lost.

When the HELLO protocol is used as the routing protocol, a data transfer path between the sending host and the receiving host is determined by using a transfer delay, and data is transferred using a path that reduces the transfer delay. However, packets exchanged by each router based on the HELLO protocol include only information on the transfer delay time, and each router determines a route based only on the transfer delay time. No important status information or attribute information is considered at all. Therefore,
For example, dynamic status information and attribute information closely related to the router processing, such as the remaining buffer capacity and CPU performance of the router and the amount of data delivered from another route or delivered to another route, are used for data transfer. Nothing will be considered in the routing. Therefore, when the load on the router in the middle of the data transfer path is high, the data transfer path is not changed,
If a delay occurs due to data transfer processing in the router or a buffer overflows in the router, data loss will occur.

When the OSPF protocol is used as a routing protocol, a data transfer path between a sending host and a receiving host is determined using a cost previously determined by an administrator of each router. It is determined based on static information such as the speed of the connection interface of each router and the bandwidth of the physical transmission path. The administrator of each router determines the cost in consideration of the processing performance of the router, but how much data is actually transferred to the router connected to the network, and how much There are many points that cannot be determined in a static state, such as whether the data amount needs to be transferred. However, such dynamic state information and attribute information cannot be reflected in router route determination during router operation. Therefore, the data transfer path is not changed when the load on the data transfer path is high, and data loss occurs when the data transfer processing in the router is delayed or the buffer overflows in the router. Become.

In the prior art 2, diagnosis / monitoring can be performed at an arbitrary part of the network system, and the diagnosis / monitoring is used for more extensive information collection, whereby the state diagnosis / monitoring of the network system is not limited to fault detection. Although it is possible to provide a device that realizes the above, according to the present technology, as a result of the diagnosis and monitoring, it is possible to avoid a case where a failure has occurred in a certain part of the network or a certain part of the communication load is high. Is not shown, diagnosis and monitoring results are accumulated in a monitoring device that performs centralized management, and by presenting the results to the operator, the operator can easily grasp the state of the network. This is a technique intended to instruct a monitoring process. Therefore, when a failure occurs in the network or an overloaded communication occurs, the data transfer router does not automatically change the data transfer path. Further, in order to prevent overload communication from occurring on the network, it is not possible to set another data transfer path when communication over a certain load occurs.
Therefore, although it is possible to measure the load in the middle of the data transfer path, the data transfer path is not changed when the load is high, and the data is transferred when the data transfer processing in the router delays or the buffer overflows in the router. Loss will occur.

In the prior art 3, a test packet circulating in the network which does not follow the communication packet control method defined in the network is provided, and the test packet is transferred from each packet switching device to thereby perform indirect packet switching. It is possible to perform efficient routing control by checking the operation of all the lines between the device and the designated packet switching device and the packet switching device within the designated section and notifying the test result to the test packet generation source. The only events that can be detected by the transfer of test packets are failures occurring on the packet switching equipment or failures on the lines arranged between the packet switching equipments. Dynamic status information such as the load and resource amount, and the packet flow rate on the lines arranged between the packet switching equipment It can not be the diagnosis and notification of attribute information. Therefore, if the communication load on the packet switching device or line in the middle of the data transfer path is high, the data transfer path is not changed, and data is lost due to delay due to data transfer processing in the router or buffer overflow in the router. Will occur. In addition, although it is said that efficient routing control can be performed by using the present technology, when a failure actually occurs on a packet switching device or a line arranged between the packet switching devices, how the routing control information is transmitted. It does not disclose whether it is created or modified.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and comprises dynamic status information during data transfer of each router arranged on a data transfer path between a sending host and a receiving host. And attribute information, and refer to each router for the dynamic status information and attribute information collected from each router. This is a technique that, when the value is higher than a threshold, causes the own router to select another downstream route. As a result, when the data transfer route is determined only by the static state information and attribute information of the conventional technology, the data transfer delay due to the data transfer load of the router during the actual communication and the buffer overflow in the router. It is an object of the present invention to improve the reliability of communication using the present network system by preventing data loss due to the above.

[0029]

A data communication system according to the present invention comprises a data transmitting device, a data receiving device, and a plurality of data relay devices disposed between the data transmitting device and the data receiving device. A data communication system comprising: a plurality of data relay devices, wherein the plurality of data relay devices include a plurality of intra-route data relay devices included in a data relay route set between the data transmitting device and the data receiving device. One of the data transmission device and the data reception device generates processing performance notification information indicating the data processing performance of each intra-route data relay device, and transmits the generated processing performance notification information to the data transmission device. And transmitting the processing performance notification between the plurality of data relay devices in the route according to the data relay route. Information, the data relay devices in each path receive the transferred processing performance notification information, and are set as their own data relay destinations in the data relay path based on the received processing performance notification information. It is characterized in that it is determined whether or not the data processing performance of the set relay destination data relay device conforms to a predetermined level, and the own data relay destination is selected based on the determination result.

[0030] At least one or more off-route data relay devices that are not included in the data relay route are connected to any of the plurality of in-route data relay devices. The in-path data relay device, based on a result of the determination on the data processing performance of the set relay destination data relay device, sets its own data relay destination to any of the out-of-path data relay devices. It is characterized by changing to.

One of the data transmitting device and the data receiving device investigates the data processing performance of each of the data relay devices in the path transmitted from the other of the data transmitting device and the data receiving device. Receiving the processing performance investigation information to be processed, and generating the processing performance notification information based on the received processing performance investigation information.

The other of the data transmitting device and the data receiving device generates the processing performance research information and directs the generated processing performance research information to one of the data transmitting device and the data receiving device. Transmitting the processing performance investigation information sequentially among the plurality of intra-path data relay apparatuses according to the data relay path, and each of the intra-path data relay apparatuses receives the transferred processing performance investigation information, Describes its own data processing performance in the received processing performance investigation information, transfers the processing performance investigation information describing its own data processing performance to another data relay device in the path, and transmits the data transmission device and the data One of the receiving devices is sequentially transferred between the plurality of intra-route data relay devices, and the data processing performance is described by each of the intra-route data relay devices. Has received the performance survey information,
The processing performance notification information is generated based on the received processing performance investigation information.

Each of the intra-route data relay apparatuses has a processing performance threshold value which is a threshold value relating to data processing performance, and determines whether or not the data processing performance of the set relay destination data relay apparatus conforms to the processing performance threshold value. It is characterized by doing.

One of the data transmitting device and the data receiving device generates processing performance notification information including a processing performance threshold value that is a threshold value for data processing performance, and transmits the generated processing performance notification information to the data processing device. A data transmission device and a transmission to the other of the data reception device, causing the processing performance notification information to be sequentially transferred between the plurality of data relay devices in the route according to the data relay route,
The in-path data relay device receives the transferred processing performance notification information, and the data processing performance of the set relay destination data relay device matches the processing performance threshold included in the received processing performance notification information. It is characterized by determining whether or not to perform.

[0035] Any of the intra-route data relay devices may include:
The processing performance investigation information for investigating the data processing performance of each of the out-of-path data relay apparatuses is transmitted to each of the out-of-path data relay apparatuses. Receiving the processing performance notification information indicating the data processing performance of the off-route data relay device, and determining whether the data processing performance of each of the off-route data relay devices conforms to a predetermined level based on the received processing performance notification information It is characterized by judging whether or not it is.

One of the data transmitting device and the data receiving device generates processing performance notification information indicating a plurality of types of data processing performance, and transmits the generated processing performance notification information to the data transmitting device and the data transmitting device. Transmitting to the other of the data receiving devices, causing the processing performance notification information to be sequentially transferred among the plurality of in-route data relay devices according to the data relay route, wherein each of the in-path data relay devices is transferred. Receiving the processing performance notification information, and determining whether or not each of the plurality of types of data processing performance of the set relay destination data relay device conforms to a predetermined level based on the received processing performance notification information. It is characterized by.

The other of the data transmitting device and the data receiving device generates processing performance research information for designating a specific data processing performance among a plurality of types of data processing performance as a research target.

The other of the data transmitting device and the data receiving device receives the processing performance notification information sequentially transferred between the plurality of data relay devices in the path, and based on the received processing performance notification information. A predetermined data relay path is set between the data transmitting device and the data receiving device.

The other of the data transmitting device and the data receiving device transmits the processing performance investigation information at a predetermined interval to one of the data transmitting device and the data receiving device. And

Each of the intra-route data relay devices receives the processing performance notification information indicating the CPU usage rate of the respective intra-route data relay device, and, based on the received processing performance notification information, sets the set relay destination data. It is characterized in that it is determined whether or not the CPU usage rate of the relay device conforms to a predetermined level.

Each of the intra-route data relay devices receives the processing performance notification information indicating the usage rate of the transmission / reception buffer of each of the intra-route data relay devices, and, based on the received processing performance notification information, performs the setting relay. It is characterized in that it is determined whether or not the usage rate of the transmission / reception buffer of the destination data relay device conforms to a predetermined level.

Each of the intra-route data relay devices receives the processing performance notification information indicating the average data transfer amount of each of the intra-route data relay devices, and, based on the received processing performance notification information, sets the set relay destination. It is characterized in that it is determined whether or not the average data transfer amount of the data relay device conforms to a predetermined level.

Each of the intra-route data relay devices receives the processing performance notification information indicating the maximum data transfer amount of the respective intra-route data relay device, and, based on the received processing performance notification information, sets the relay destination. It is characterized in that it is determined whether or not the maximum data transfer amount of the data relay device conforms to a predetermined level.

[0044] A data relay device according to the present invention is a data relay device connected to a plurality of data communication devices and selecting a data relay destination from the plurality of data communication devices. A processing performance notification information receiving unit that receives, from the data communication device, processing performance notification information indicating the data processing performance of the specific data communication device; and the processing performance notification received by the processing performance notification information receiving unit. A processing performance determining unit that determines whether data processing performance of the specific data communication device conforms to a predetermined level based on the information; and the plurality of data communication devices based on a determination result by the processing performance determining unit. And a relay destination selection unit that selects a data relay destination from

The relay destination selecting unit selects any one of the plurality of data communication devices other than the specific data communication device as a data relay destination based on the result of the determination by the processing performance determining unit. It is characterized by the following.

The data relay device further includes a processing performance threshold value recording unit that records a processing performance threshold value that is a threshold value relating to data processing performance, and the processing performance judgment unit determines a data processing performance of the specific data communication device. Is determined whether or not the threshold value matches the processing performance threshold value.

The processing performance notification information receiving unit receives, from each of the plurality of data communication devices, processing performance notification information for notifying the data processing performance of each data communication device. Based on the processing performance notification information received by the performance notification information receiving unit, determine whether the data processing performance of each of the data communication devices conforms to a predetermined level, the relay destination selection unit, the processing performance A data relay destination is selected from the plurality of data communication devices based on a determination result by the determination unit.

[0048] A data relay method according to the present invention is a data relay method that is connected to a plurality of data communication devices and selects a data relay destination from the plurality of data communication devices. Receiving, from the data communication device, processing performance notification information indicating the data processing performance of the specific data communication device; and the processing performance notification received by the processing performance notification information receiving step. A processing performance determining step of determining whether data processing performance of the specific data communication device conforms to a predetermined level based on the information; and the plurality of data communication devices based on a determination result of the processing performance determining step And a relay destination selecting step of selecting a data relay destination from

In the relay destination selecting step, any one of the plurality of data communication apparatuses other than the specific data communication apparatus is selected as a data relay destination based on a result of the determination in the processing performance determining step. It is characterized by the following.

[0050] The data relay method further includes a processing performance threshold value recording step of recording a processing performance threshold value which is a threshold value relating to data processing performance, wherein the processing performance determining step includes a data processing performance of the specific data communication apparatus. Is determined whether or not the threshold value conforms to the processing performance threshold value.

The processing performance notification information receiving step receives, from each of the plurality of data communication devices, processing performance notification information for notifying the data processing performance of each data communication device. Based on the processing performance notification information received in the performance notification information receiving step, it is determined whether or not the data processing performance of each of the data communication devices conforms to a predetermined level. A data relay destination is selected from the plurality of data communication devices based on a result of the determination in the determining step.

[0052]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 Hereinafter, a first embodiment of the present invention will be described with reference to FIG. FIG. 1 shows an example of a data communication system according to the present embodiment.

In FIG. 1, reference numeral 1 denotes a host A serving as a transmission host (data transmission device), and reference numeral 2 denotes a host B serving as a reception host (data reception device). A computer having the same function as the host A1 or a computer having the same function as the host B2 is the host A
1 and the host B2 are connected at positions equivalent to the network to which they are connected, but are omitted in this example for simplicity. Reference numerals 3 to 10 denote routers (data relay devices) arranged on the network. A route (data relay route) 11 has been created. Therefore, the routers 3, 4, 5, and 6 are intra-route data relay devices included in the data relay route. The routers 7 and 8 and the routers 9 and 10 are routers (out-of-path data relay devices) that are not involved in the data delivery path 11. Reference numeral 12 denotes an alternative data transfer path modified according to the present invention. The data delivery route 11 is set by, for example, any of the methods described in the related art 1.

FIG. 2 shows that the host A1 periodically updates the host B2.
Packets sent to the network (processing performance survey information)
Is shown. First, when the host A1 transmits a packet to the host B2, the diagnostic packet includes the IP address 30 of the host A1 as the transmission source, the IP address 31 of the host B2 as the reception destination, and the transmitted packet. Packet type 32 indicating that
Only, and the payload 41 is empty. When the diagnostic packet passes through the first router 3 located on the data delivery path 11 in FIG. State information and attribute information (data processing performance information) 34 are added.

Next, the configuration of each router will be described.
FIG. 40 is a diagram illustrating the configuration of each router. In the figure, 3000 is another router, host A1 or host B
2 is a packet transmission / reception unit that receives various packets from the router 2 and transmits various packets to another router, the host A1, or the host B2. Reference numeral 3100 denotes a packet type determining unit that determines the type of the packet received by the packet transmitting / receiving unit 3000. Packet type determination unit 310
0 determines the packet type of any one of the normal packet, the diagnostic packet, and the diagnostic result packet, and determines the normal packet processing unit 3200 and the diagnostic packet processing unit 330 based on the determination result.
0, the packet is passed to one of the diagnosis result packet processing units 3400.

Reference numeral 3200 denotes a normal packet processing unit for processing a normal packet. Note that the processing for a normal packet is out of the scope of the present invention, and thus the details are omitted.

Reference numeral 3300 denotes a diagnostic packet processing unit for performing processing on the diagnostic packet. Diagnostic packet processing unit 3
Reference numeral 300 includes a dynamic state information / attribute information detecting unit 3310 and a diagnostic packet processing unit 3320. Further, the diagnostic packet processing unit 3320 further includes a dynamic state information / attribute information giving unit 3330 and an own IP address giving unit 3340. The dynamic state information / attribute information detection unit 3310 detects dynamic state information and attribute information of the own router. The dynamic state information / attribute information adding unit 3330 adds the dynamic state information and attribute information of the own router detected by the dynamic state information / attribute information detection unit 3310 to the payload of the diagnostic packet. The own IP address assigning unit 3340 assigns the IP address of the own router to the payload of the diagnostic packet.

Reference numeral 3400 denotes a diagnosis result packet processing unit that performs processing on the diagnosis result packet. The diagnosis result packet processing unit 3400 includes a downstream router IP address search unit 3410, a downstream router dynamic state information / attribute information acquisition unit 3
420, downstream router determination unit 3430, threshold recording unit 344
0, a downstream router selection unit 3450. Downstream router I
The P address search unit 3410 searches the payload of the diagnosis result packet for the IP address of the router one stage downstream of the own router. The downstream router dynamic state information / attribute information acquisition unit 3420 retrieves the dynamic state information and attribute information of the downstream router from the payload of the diagnostic packet based on the IP address of the downstream router searched by the downstream router IP address search unit 3410. get. Downstream router judgment unit 3430
Is the downstream router dynamic state information / attribute information acquisition unit 3420
It is determined whether or not the acquired dynamic status information and attribute information of the downstream router exceed a predetermined threshold. Threshold recording unit 3
Reference numeral 440 records a threshold value used when the downstream router determination unit 3430 determines dynamic state information and attribute information of the downstream router. The downstream router selection unit 3450 selects a downstream router based on the determination result by the downstream router determination unit 3430. In the above-described example, if the downstream router determination unit 3430 of the router 5 determines that the dynamic state information or attribute information of the router 6 is equal to or less than the threshold, the downstream router selection unit 3450 of the router 5 directly passes the router 6 downstream. If the router is selected as a router (relay destination) and it is determined that the dynamic state information and attribute information of the router 6 are equal to or larger than the threshold, the downstream router selection unit 3450 of the router 5 substitutes another router, for example, the router 10 for an alternative. Select as a downstream router (relay destination).

Reference numeral 3500 denotes a routing table, which is referred to when the packet transmitting / receiving unit 3000 transmits various packets.

Next, the internal operation sequence when a router on the data delivery path 11 receives a diagnostic packet will be described with reference to FIG. When the packet transmitting / receiving unit 3000 of the router receives the packet, in step 60, the packet type determining unit 3100 checks whether the received packet is a diagnostic packet by examining the packet type 32 in the diagnostic packet format shown in FIG. If the received packet is a diagnostic packet, the process proceeds to step 61; otherwise, the process proceeds to step 63. In step 61, the own IP address assigning unit 3340 of the diagnostic packet processing unit 3320 adds the IP address of the own router to the payload of the diagnostic packet received. Next, in step 62, the diagnostic packet processing unit 3320
The dynamic state information / attribute information adding unit 3330 adds the dynamic state information and attribute information of the own router to the payload of the diagnostic packet received. The dynamic status information / attribute information added to the payload of the diagnostic packet by the dynamic status information / attribute information adding unit 3330 is detected by the dynamic status information / attribute information detection unit 3310. afterwards,
In step 63, the packet transmitting / receiving unit 3000 performs a normal packet transfer process of the router using the routing table 3500, and ends the process. Step 6
The normal packet transfer process of No. 3 is a process generally provided in an existing router, and thus the description thereof is omitted here.

The above processing is performed inside the router 3, and thereafter, the diagnostic packet passes through the router 3 and then passes through the second router 4 located on the data delivery path. When the diagnostic packet passes through the router 4, the router 4 adds the IP address 35 of the second router 4 and the dynamic status information and attribute information 36 of the second router 4 to the payload 41 of the diagnostic packet. In this way, when a diagnostic packet passes through each router, the IP address of the router and the dynamic status information and attribute information existing on the path between the sending host and the receiving host are converted into a diagnostic packet inside the router. To the payload. In the example shown in FIG. 1, the payload of the diagnostic packet upon arrival at the host B2 is provided with the IP address of the fourth router 6 and the dynamic state information and attribute information of the fourth router. Will be.
The processing of transmitting the diagnostic packet from the host A1 to the host B2 may be performed by an explicit transmission request from an application running on the host A1, or the internal processing of the operating system running on the host A1 may be performed. May include a function for transmitting a diagnostic packet.

When the diagnostic packet transmitted from the host A1 to the host B2 arrives at the host B2,
2 duplicates the payload of the received diagnostic packet into a diagnostic result packet (processing performance notification information) shown in FIG. 4 and transmits it to the host A1. In the format of the diagnostic result packet, the order of the IP address 42 of the transmission source host B2 and the IP address 43 of the reception destination host A1 is switched, and the packet type 32 is merely a value representing the diagnosis result packet. Each component of the host B
No difference from the payload 41 of the diagnostic packet arriving at 2.

When the diagnostic result packet is transmitted from the host B2 to the host A1, the diagnostic result packet is delivered to the host A1 while being transferred by the router on the way using the data delivery path 11. The router 6 on the transfer path 11 for first transferring the diagnostic result packet transmitted from the host B2 has no router downstream of the own router on the transfer path 11 formed between the host A1 and the host B2. This is known from the payload 41 of the diagnostic result packet, and the diagnostic result packet is transferred to the router 5 which is a router upstream of the own router. The router 5 that has received the diagnosis result packet from the downstream router 6 checks the payload 41 of the diagnosis result packet, and thereby knows that the router 6 exists downstream from the own router. Here, the IP address of the router 6 downstream of the router 5 and dynamic state information and attribute information are searched from the payload 41 of the diagnosis result packet. The router 5 that has obtained the IP address of the router 6 and the dynamic status information and attribute information has a threshold (processing performance threshold) for the dynamic status information and attribute information set in advance by the router administrator in its own router. Then, by comparing with the dynamic state information and attribute information of the router 5 obtained from the payload 41 of the diagnosis result packet, it is determined whether or not to change the data transfer path from the router 5 to the host B2.

Next, the internal operation sequence when the router on the data delivery path 11 receives the diagnosis result packet will be described with reference to FIG. When the router receives the packet, in step 80, the packet type determination unit 3100 determines whether the received packet is a diagnosis result packet by checking the packet type 3 in the diagnosis result packet format shown in FIG.
If the received packet is a diagnosis result packet, the process proceeds to step 81, and if not, the process proceeds to step 63. In step 81, the downstream router IP address search unit 341
0 searches for the IP address of its own router from the payload of the diagnostic result packet received by
The position where the P address and the dynamic state information and the attribute information are recorded is specified. In step 82, it is checked whether or not there is a record of another router behind the position where the IP address of the own router on the payload and the dynamic state information or attribute information are recorded, and the record of the other router is located behind. If there is, the process proceeds to step 83, and if there is no record of another router behind, the own router is determined to be the router located at the most downstream in the data transfer path, and the process proceeds to step 63. In step 83, the downstream router dynamic state information / attribute information acquisition unit 3420 searches for the payload, and acquires the dynamic state information and attribute information of the router one stage downstream of the own router. In step 84, the dynamic state information and attribute information obtained by the downstream router determination unit 3430 exceed the threshold (processing performance threshold) for the dynamic state information and attribute information set in the router in advance by the router administrator. It is checked whether or not the acquired dynamic state information or attribute information exceeds the threshold. If the acquired dynamic state information or attribute information does not exceed the threshold, the process proceeds to step S85. The process moves to step 63. In step 85, the downstream router selection unit 3450 checks whether there is a router that is an alternative downstream router in addition to the router whose own router has forwarded data as a downstream router, and determines that there is a router that is an alternative router. If so, the process proceeds to a step 86, and if there is no router serving as the substitute router, the process proceeds to a step 63. In step 86, the downstream router selection unit 3450
Sets the router to be used by the own router as a downstream router (relay destination) as an alternative router, and moves the processing to step 63. In step 63, the normal packet transfer processing of the router is performed, and the processing ends. Step 63
The normal packet transfer process is a process generally provided in existing routers, and thus the description thereof is omitted here.

As described above, the processing of the diagnosis result packet is performed inside each router located on the data delivery path 11. The operation of the router 5 in the present embodiment shown in FIG. The change of the data transfer path will be described. Now, it is assumed that the router 5 has received the diagnosis result packet transferred from the router 6 existing downstream of the router 5. The router 5 acquires the values of the dynamic state information and the attribute information of the router 6 downstream of the router 5 according to the processing flow shown in FIG. If the threshold value is not exceeded, the router 6 is selected as a relay destination, and the diagnostic result packet is transferred to the router 4 that is upstream of the router 5. Also, the router 5 receives the diagnostic result packet transferred from the router 6 downstream of the router 5, and manages the value of the dynamic state information and attribute information of the downstream router 5 obtained from the diagnostic result packet and the router 5. As a result of comparison with a threshold value set in advance by a user, if the threshold value is exceeded, a router 10 operable as another downstream router is detected inside the router 5, and the router 5 uses the next time for data transfer. The setting is automatically changed to use the router 10 as a downstream router. After that, the diagnostic packet is transferred to the router 4 located upstream of the router 5. At this stage, the data delivery route 11 that has been used up to now has the router 5 to the router 10
The alternative data delivery path 12 formed by the host B2 will be used.

In the above description, the diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and the diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to carry out in the flow of.

By using the present invention, dynamic status information and attributes of each router which is located on the data transfer path defined between the host A1 and the host B2 and executes the data transfer processing, which cannot be achieved by the prior art. It is possible to change the data transfer route according to the change in information, and as a result, it is possible to create a route in which the dynamic state information and attribute information of the used route are constant, and the dynamic route information and attribute information during data transfer. It is possible to construct a network system that suppresses a data transfer delay caused by the change of the data transfer and determines an optimal data delivery route.

Embodiment 2 Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. FIG. 6 shows a format of a diagnostic packet transmitted from the host A1 to the host B2 in the present embodiment, and FIG. 7 shows a format of a diagnostic result packet transmitted from the host B2 to the host A1 in the present embodiment. Is shown.

In the diagnostic packet of FIG. 6, the data delivery path 11 for the diagnostic packet described in the first embodiment is used.
When the diagnostic packet passes through each of the above routers, the CPU utilization of each router is given to the dynamic status information and attribute information given by each router. Therefore, in the second embodiment, the data delivery route 11
When the diagnostic packet passes through the first router, the IP address 33 of the first router and the CPU of the first router
When the diagnostic packet passes through the second router with the usage rate of 100, the IP addresses 35 and 2 of the second router
The CPU usage rate 101 is given. As described above, when passing through the (n-1) -th router on the data delivery route 11, the IP address 37 of the (n-1) -th router and the CPU usage rate 102 of the (n-1) -th router are assigned. When passing through the n-th router on 11, the IP address 39 of the n-th router and the CPU usage rate 103 of the n-th router are given.

After that, the diagnostic packet arrives at the host B2, and the payload 41 of the diagnostic packet is duplicated in the host B2 as the payload of the diagnostic result packet as described in the first embodiment. And transmits a diagnosis result packet using the data delivery path 11. When the diagnostic result packet passes through each router existing on the data transfer path, a certain router on the data delivery path indicating the flow of data from the host A1 to the host B2 is similar to the method described in the first embodiment. Using the method of the above, the IP address of the downstream router of the own router and the CPU usage rate are calculated as follows:
A threshold value of the CPU usage rate set in advance by the own router administrator in the own router is compared, and whether or not to change the downstream data transfer path is selected.

FIG. 8 shows the internal processing when each router on the data transfer path transfers the diagnostic packet transmitted from the host A1. When each router transfers a diagnostic packet, the processing flow is almost the same as that of the first embodiment. At step 150, a process of adding the CPU usage rate of the own router to the payload of the diagnostic packet is performed.

FIG. 9 shows the internal processing when each router on the data transfer path transfers the diagnostic result packet transmitted from the host B2. When each router transfers the diagnosis result packet, the processing flow is almost the same as that of the first embodiment.
A process of acquiring the U usage rate from the payload is performed, and the CPU usage rate of the router located one stage downstream obtained in step 152 is set to C
A process of comparing a threshold value of the PU usage rate is performed.

In the above description, the diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and the diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to perform in the flow of.

By using the present invention, among the routers present on the data delivery route 11, data transfer with a reduced transfer delay that occurs when the router CPU usage rate is high can be performed, and an optimum data delivery route can be determined. It is possible to construct a network system that performs

Embodiment 3 Hereinafter, a third embodiment of the present invention will be described with reference to FIGS. FIG. 10 is a diagram illustrating a state in which the host A1 is connected to the host B2 in the present embodiment.
Indicates the format of the diagnostic packet sent to
FIG. 11 shows a format of a diagnosis result packet transmitted from the host B2 to the host A1 in the present embodiment.

In the diagnostic packet of FIG. 10, when the diagnostic packet passes through the first router, the transmission / reception buffer usage rate 200 of the first router is added to the diagnostic packet described in the second embodiment. When the diagnostic packet passes through the second router, the transmission / reception buffer usage rate 201 of the second router is added. In this way, when the diagnostic packet passes through the (n-1) -th router on the data delivery path 11, the transmission / reception buffer usage rate 202 of the (n-1) -th router is added, and the diagnostic packet passes through the n-th router. When using, the transmission / reception buffer usage rate 2 of the n-th router
03 is added. In the case of the present invention, by setting the router in advance by each router administrator, the CPU usage rate of each router may be omitted and may not be added to the diagnosis result packet.

Then, the diagnostic packet arrives at the host B2, and the payload 41 of the diagnostic packet is copied to the payload of the diagnostic result packet in the host B2, as described in the first embodiment, and the host B2 transmits to the host A1. And transmits a diagnosis result packet using the data delivery path 11. When the diagnostic packet passes through each router existing on the data transfer path, the router on the data delivery path indicating the flow of data from the host A1 to the host B2 becomes the same as the method described in the second embodiment. By using the method, the downstream router IP address and the CPU usage rate of the own router, and the transmission / reception buffer usage rate added in the present embodiment are set in advance in the own router by the own router administrator.
By comparing with the threshold value of the PU usage rate or the threshold value of the transmission / reception buffer usage rate, it is determined whether or not to change the downstream data transfer path. In the case of the present invention, by setting each router in advance by each router administrator, the C of each router can be set.
In some cases, the PU usage rate is omitted and is not added to the diagnosis result packet. The other processes are the same as those in the first and second embodiments, and the description is omitted here.

FIG. 12 shows the internal processing when each router on the data transfer path transfers the diagnostic packet transmitted from the host A1. When each router transfers a diagnostic packet, the processing flow is almost the same as that of the second embodiment. In step 250, a process of adding the transmission / reception buffer usage rate of the own router to the payload of the diagnostic packet is performed. In the case of the present invention, the process of adding the CPU usage rate to the payload in step 150 may be omitted by setting the router in advance by each router administrator.

FIG. 13 shows the internal processing when each router on the data transfer path transfers the diagnostic result packet transmitted from the host B2. When each router transfers the diagnosis result packet, the processing flow is almost the same as that of the second embodiment.
A process of acquiring the PU usage rate and the transmission / reception buffer usage rate from the payload is performed.
A process of comparing a threshold value of the CPU usage rate and a threshold value of the transmission / reception buffer usage rate preset in the own router by the administrator is performed. In the case of the present invention, each router administrator sets the router in advance, thereby obtaining the CPU usage rate of the router one stage downstream of the own router in step 251 and the self-processing in step 252. In some cases, the process of comparing the CPU usage rate of the router one stage downstream of the router with the threshold value of the CPU usage rate set in advance by the administrator for the own router may be omitted. The other processes are the same as those in the first and second embodiments, and the description is omitted here.

In the above description, the diagnosis packet is transmitted from the transmission host A1 to the reception host B2, and the diagnosis result packet is transmitted from the reception host B2 to the transmission host A1. It is also possible to carry out in the flow of.

By using the present invention, it is possible to prevent data loss due to overflow of the transmission / reception buffer, which occurs when the usage rate of the transmission / reception buffer of the router among the routers existing on the data delivery path 11 is high. It is possible to construct a network system for determining a data delivery route.

Embodiment 4 Hereinafter, a fourth embodiment of the present invention will be described with reference to FIGS. Figure 1
4 shows a format of a diagnostic packet transmitted from the host A1 to the host B2 in the present embodiment, and FIG. 15 shows a format of a diagnostic result packet transmitted from the host B2 to the host A1 in the present embodiment. ing.

In the diagnostic packet of FIG. 14, when the diagnostic packet passes through the first router, an average data transfer amount 300 of the first router is added to the diagnostic packet described in the third embodiment. When the diagnostic packet passes through the second router, the average data transfer amount 301 of the second router is added. In this way, when the diagnostic packet passes through the (n-1) -th router on the data delivery route 11, the average data transfer amount 302 of the (n-1) -th router is added, and the diagnostic packet passes through the n-th router. When doing
The average data transfer amount 303 of the n-th router is added. In the case of the present invention, by setting each router in advance by each router administrator, the C of each router can be set.
In some cases, the PU usage rate and the transmission / reception buffer usage rate of each router are omitted and are not added to the diagnosis result packet.

Then, the diagnostic packet arrives at the host B2, and the payload 41 of the diagnostic packet is duplicated in the host B2 as the payload of the diagnostic result packet as described in the first embodiment, and the host B2 transmits the diagnostic packet to the host A1. And transmits a diagnosis result packet using the data delivery path 11. When the diagnostic packet passes through each router existing on the data transfer path, the router on the data delivery path indicating the flow of data from the host A1 to the host B2 becomes the same as the method described in the third embodiment. Using the method, the downstream router IP address of the own router, the CPU usage rate, the transmission / reception buffer usage rate, and the average data transfer amount added in the present embodiment are set in advance in the own router by the own router administrator. The threshold value of the CPU usage rate, the threshold value of the transmission / reception buffer usage rate, and the threshold value of the average data transfer amount are compared with each other to determine whether to change the downstream data transfer path. In addition,
In the case of the present invention, by setting the router in advance by each router administrator, the CPU usage rate and the transmission / reception buffer usage rate of each router may be omitted and may not be added to the diagnosis result packet.

FIG. 16 shows the internal processing when each router on the data transfer path transfers the diagnostic packet transmitted from the host A1. When each router transfers a diagnostic packet, the processing flow is almost the same as in the third embodiment. In step 350, a process of adding the average data transfer amount of the own router to the payload of the diagnostic packet is performed.
In the case of the present invention, the processing of adding the CPU usage rate to the payload of step 150 and the processing of step 2 are performed by setting the router in advance by each router administrator.
The process of adding the transmission / reception buffer usage rate to the 50 payloads may be omitted. Other processes are the same as those in the first, second, and third embodiments, and a description thereof will not be repeated.

FIG. 17 shows the internal processing when each router on the data transfer path transfers the diagnostic result packet transmitted from the host B2. When each router transfers the diagnosis result packet, the processing flow is almost the same as that of the third embodiment.
A process of acquiring the PU usage rate, the transmission / reception buffer usage rate, and the average data transfer amount from the payload is performed.
The CPU usage rate, transmission / reception buffer usage rate, and average data transfer rate of the one-stage downstream router obtained in step 2 are compared with the CPU usage rate threshold value, transmission / reception buffer usage rate threshold value, and threshold value preset in the router by the administrator. A process of comparing the threshold value of the data transfer amount is performed. In the case of the present invention, the processing of acquiring the CPU usage rate of the router one stage downstream of the own router and the transmission / reception buffer usage rate in step 351 are performed by setting the router in advance by each router administrator. Processing, and one of the own routers in step 352.
The CPU usage rate and the transmission / reception buffer usage rate of the downstream router are set to the own router in advance by the CPU set by the administrator.
In some cases, the process of comparing the usage rate threshold value with the transmission / reception buffer usage rate may be omitted. Other processes are the same as those in the first, second, and third embodiments, and a description thereof will not be repeated.

In the above description, the diagnosis packet is transmitted from the transmission host A1 to the reception host B2, and the diagnosis result packet is transmitted from the reception host B2 to the transmission host A1. It is also possible to carry out in the flow of.

By using the present invention, it is possible to prevent data congestion on a physical transmission path between a router and a router downstream of the router among the routers existing on the data delivery path 11. Thus, it is possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

Embodiment 5 Hereinafter, a fifth embodiment of the present invention will be described with reference to FIGS. Figure 1
8 shows a format of a diagnostic packet transmitted from the host A1 to the host B2 in the present embodiment, and FIG. 19 shows a format of a diagnostic result packet transmitted from the host B2 to the host A1 in the present embodiment. ing.

In the diagnostic packet of FIG. 18, when the diagnostic packet passes through the first router, the maximum data transfer amount 400 of the first router is added to the diagnostic packet described in the fourth embodiment. When the diagnostic packet passes through the second router, the maximum data transfer amount 401 of the second router is added. In this way, when the diagnostic packet passes through the (n-1) -th router on the data delivery route 11, the maximum data transfer amount 402 of the (n-1) -th router is added, and the diagnostic packet passes through the n-th router. When doing
The maximum data transfer amount 403 of the n-th router is added. In the case of the present invention, by setting each router in advance by each router administrator, the C of each router can be set.
In some cases, the PU usage rate, the transmission / reception buffer usage rate of each router, and the average data transfer amount of each router are omitted and are not added to the diagnosis result packet.

Then, the diagnostic packet arrives at the host B2, and the payload 41 of the diagnostic packet is duplicated in the host B2 as the payload of the diagnostic result packet as described in the first embodiment. And transmits a diagnosis result packet using the data delivery path 11. When the diagnostic packet passes through each router existing on the data transfer path, the router on the data delivery path indicating the flow of data from the host A1 to the host B2 uses the same method as that described in the fourth embodiment. Using the method, the downstream router IP address of the own router, the CPU usage rate, the transmission / reception buffer usage rate, the average data transfer amount, and the maximum data transfer amount added in the present embodiment are stored in the own router in advance by the own router. The threshold value of the CPU usage rate, the threshold value of the transmission / reception buffer usage rate, the threshold value of the average data transfer amount, or the threshold value of the maximum data transfer amount set by the user, and selects whether or not to change the downstream data transfer path. . In the case of the present invention, by setting the router in advance by each router administrator, the CPU usage rate, transmission / reception buffer usage rate, and average data transfer amount of each router are omitted and are not added to the diagnosis result packet. There is also.

FIG. 20 shows internal processing when each router on the data transfer path transfers a diagnostic packet transmitted from the host A1. When each router transfers a diagnostic packet, the processing flow is almost the same as that of the fourth embodiment. In step 450, a process of adding the maximum data transfer amount of the own router to the payload of the diagnostic packet is performed.
In the case of the present invention, the processing of adding the CPU usage rate to the payload of step 150 and the processing of step 2 are performed by setting the router in advance by each router administrator.
The process of adding the transmission / reception buffer usage rate to the payload of 50 and the process of adding the average data transfer amount to the payload in step 350 may be omitted. Other processes are the same as those in the first and second embodiments, the third and fourth embodiments, and the description thereof is omitted.

FIG. 21 shows the internal processing when each router on the data transfer path transfers the diagnosis result packet transmitted from the host B2. When each router transfers the diagnosis result packet, the flow of processing is almost the same as that of the fourth embodiment.
The PU utilization rate, the transmission / reception buffer utilization rate, the average data transfer amount, and the maximum data transfer amount are obtained from the payload.
The PU usage rate, the transmission / reception buffer usage rate, the average data transfer rate, and the maximum data transfer rate are set in the own router by the CPU usage threshold value, the transmission / reception buffer usage rate threshold value, and the average data transfer rate threshold value which are set in advance by the administrator. And a process of comparing the threshold value of the maximum data transfer amount with the threshold value. In the case of the present invention, the process of acquiring the CPU usage rate of the router one stage downstream of the own router and the transmission / reception buffer usage rate in step 451 by setting the router in advance by each router administrator. And the process of acquiring the average data transfer amount, and the CPU usage rate, the transmission / reception buffer usage rate, and the average data transfer amount of the router one step downstream of the own router in step 452 are set in the own router in advance by the administrator. In some cases, the process of comparing the threshold value of the CPU usage rate, the transmission / reception buffer usage rate, and the threshold value of the average data transfer amount may be omitted. Other processes are the same as those in the first and second embodiments, the third and fourth embodiments, and the description thereof is omitted.

In the above description, the diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and the diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to carry out in the flow of.

By using the present invention, among the routers on the data distribution path 11, data congestion at the time of a peak distribution load on a physical transmission path between a certain router and a router downstream of the router is prevented. This makes it possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

Embodiment 6 FIG. Hereinafter, a sixth embodiment of the present invention will be described with reference to FIG. FIG. 22 shows a part of the router on the data transfer path defined between the host A1 and the host B2 and the routers located around the router in the present embodiment.

In FIG. 22, reference numerals 500, 501, 502, and 503 denote routers having the functions described in the first to fifth embodiments, and 504 denotes a router 500 and a router 50.
2 shows a data transfer path defined between the host A1 and the host B2 where the host 2 exists. Therefore, the router 500 and the router 501 are data relay devices in the route, and the router 501
The router 502 is an off-route data relay device. 50
Reference numeral 5 denotes a diagnostic packet transmitted from the host A1 to the host B2. Reference numerals 506, 507, and 508 duplicate the diagnostic packet 505 by the router 500 located on the data transfer path 504 and transfer the duplicate from the communication interface of the router 500. The diagnostic packet is shown. In addition,
Regarding the data transfer from the host A1 to the host B2, it is assumed that the data transfer from the router 501 to the host B2 is impossible because the physical transmission path is not connected, and the data transfer from the router 502 and the router 503 to the host B2. It is assumed that the transfer is possible because the physical transmission path is connected. Reference numeral 509 denotes a packet returned as an error packet in the internal processing of the router 501 because the router 501 does not reach the host B2 when the diagnostic packet 506 is received by the router 501. Reference numeral 510 denotes a diagnostic packet 507 inside the router 502. 511 represents a diagnostic packet obtained by processing the diagnostic packet 508 inside the router 503. The packets 509 to 511 correspond to the diagnosis result packets described in the first to fifth embodiments, and are processing performance notification information.

A diagnostic packet is transmitted from the host A1 to the host B2, and the above-described first, second, third, fourth, or fourth embodiment is executed by a router upstream of the router 500. Form 5
The diagnostic packet 505 is transmitted to the router 5 using the method shown in FIG.
00 is transferred. The router 500 that has received the diagnostic packet 505 uses the method described in the first embodiment, the second embodiment, the third embodiment, the fourth embodiment, or the fifth embodiment to apply a method to the payload of the diagnostic packet 505. Then, the CPU usage rate, the transmission / reception buffer usage rate, the average data transfer amount, and the maximum data transfer amount set in advance by the administrator of the router 500 are given as the IP address of the own router and the dynamic state information or attribute information.
Next, a copy of the diagnostic packet 505 for the communication interface of the own router is made inside the router 500, and the diagnostic packet 506 is sent to the router 501 located downstream of the router 500, and the diagnostic packet 507 is sent to the router 502. 503 is transferred as a diagnostic packet 508.

Of the diagnostic packets 506, 507, and 508 created by duplicating the diagnostic packet 505 by the router 500, the diagnostic packet 506 transferred to the router 501 arrives at the router 501 and then arrives at the router 501.
It is determined by internal processing that it does not arrive at the host B2 located downstream, and the IP address of the router 501, the CPU usage rate as dynamic state information and attribute information, the transmission / reception buffer usage amount, the average data transfer amount, and the maximum data The transfer amount is not added to the payload, the payload of the diagnostic packet 506 is copied as it is to the error packet, and the packet type is returned to the router 500 as an error packet. Among the diagnostic packets created by duplicating the diagnostic packet 505 by the router 500, the diagnostic packet 507 transferred to the router 502 is transmitted to the router 502, and is processed in the router 502 by the IP address of the router 502. Then, the CPU usage rate, the transmission / reception buffer usage rate, the average data transfer amount, and the maximum data transfer amount as dynamic state information and attribute information are added to the payload, and the payload is transferred to the host B2 as the diagnostic packet 510. The diagnostic packet 508 created by duplicating the diagnostic packet 505 by the router 500 is also transferred to the host B as a diagnostic packet 511 through the internal processing of the router 503.

The router 500, which has received the error packet 509 returned from the router 501,
09 by referring to the packet type of the host B2.
When the router 501 cannot be used as a data transfer path to the router 501, the error packet 509 is discarded by internal processing of the router 501.

FIG. 23 shows the internal processing of the router when a diagnostic packet is received using the method of this embodiment. This processing shows the contents of the processing when the present invention is applied to the fifth embodiment, but also applies to the first, second, third, and fourth embodiments. Similar processing can be added. In the present embodiment, in step 550, the router 501 checks whether it is possible to reach the host B2 located downstream of the router itself. Is generated, and the payload of the diagnostic packet is copied into an error packet in step 552.
A process of returning an error packet to 0 is performed. Other processes are the same as those in the first, second, third, fourth, and fifth embodiments, and therefore, description thereof will be omitted.

FIG. 24 shows the internal processing of the router 500 when an error packet is received in the case of using this embodiment.
This processing shows the contents of the processing when the present invention is applied to the fifth embodiment, but also applies to the first, second, third, and fourth embodiments. Similar processing can be added. In the present embodiment, the router 500 determines whether or not the packet received in step 554 is an error packet, based on the packet type. Other processes are the same as those in the first, second, third, fourth, and fifth embodiments, and therefore, description thereof will be omitted.

In the above description, a diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and a diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to carry out in the flow of.

By using the present invention, a plurality of diagnostic packets are duplicated in advance on the data transfer path 11 and the data transfer path 504 representing a part of the data transfer path 11 and transmitted to the downstream router connected to the communication interface of a certain router. By transferring data in advance, it is possible to determine the optimal data delivery route at that stage, and if the physical transmission path is not connected to the final data transfer destination, data cannot be reached As a result, it is possible to construct a network system that can determine an optimal data delivery route for preventing transfer delay and data loss at an early stage.

Embodiment 7 FIG. Hereinafter, a seventh embodiment of the present invention will be described with reference to FIGS. FIG.
5 shows a format of a diagnostic packet transmitted from the host A1 to the host B2 in the present embodiment, and FIG. 26 shows a format of a diagnostic result packet transmitted from the host B2 to the host A1 in the present embodiment. ing.

The diagnostic packet shown in FIG. 25 is different from the diagnostic packet described in the fifth embodiment in that the dynamic state information and attribute information acquisition designation 60 at the beginning of the payload inside the host A1 are changed.
0 is added. Dynamic status information and attribute information acquisition specification 60
In 0, the type of dynamic status information and attribute information assigned by each router arranged on the data delivery path 11 between the host A1 and the host B2 is specified. Each router arranged on the data delivery route 11 refers to the setting value of the dynamic state information and the attribute information acquisition designation 600, and the designated dynamic state information and the attribute information are referred to as the IP of the own router. Attached to the payload 41 together with the address. In addition,
In the present invention, the dynamic state information and the attribute information acquisition designation 600 are added to the diagnostic packet of the fifth embodiment. However, in the first and second embodiments, the third and third embodiments, and In the fourth embodiment and the sixth embodiment, it can be similarly added.

FIG. 26 shows the format of a diagnosis result packet in this embodiment. The diagnostic packet shown in FIG. 25 arrives at the host B2, and the payload 41 of the diagnostic packet is copied into the payload of the diagnostic result packet in the host B2 as described in the first embodiment. , The diagnostic result packet is transferred using the data delivery path 11. When the diagnostic packet passes through each router existing on the data transfer path, the router on the data delivery path indicating the flow of data from the host A1 to the host B2 receives dynamic state information and attribute information acquisition designation 600. And the dynamic state information and attribute information of the router one stage downstream attached to the payload 41 and the dynamic state information and attribute set in advance in the own router by the router administrator. The information is compared with a threshold value to determine whether to change the downstream data transfer path. In the present invention, the dynamic state information and the attribute information acquisition designation 600 are added to the diagnostic packet of the fifth embodiment. However, in the first and second embodiments, the third and the third embodiments, The same can be added in the fourth and sixth embodiments.

FIG. 27 shows the internal processing when each router on the data transfer path transfers the diagnostic packet transmitted from the host A1. This processing shows the content of the processing when the present invention is applied to the fifth embodiment. 6, the same processing can be added. When each router transfers a diagnostic packet, the processing flow is almost the same as in the fifth embodiment, and the values set in the dynamic state information and attribute information acquisition designation 600 from the payload of the diagnostic packet in step 650 are as follows. In step 651, the dynamic state information and the attribute information acquisition designated value are inspected to determine whether the recording of the CPU usage rate is designated. Step 652
In step 653, the dynamic state information and attribute information acquisition designated value are checked to determine whether recording of the transmission / reception buffer usage rate is designated. In step 653, the dynamic state information and attribute information acquisition designated value are examined. It is determined whether or not the record of the average data transfer amount is designated. In step 654, the dynamic state information and the attribute information acquisition designated value are examined to determine whether or not the record of the maximum data transfer amount is designated. The other processes are the same as those in the first, second, third, fourth, fifth, and sixth embodiments, and thus the description thereof is omitted.

FIG. 28 shows the internal processing when each router on the data transfer path transfers the diagnostic result packet transmitted from the host B2. This processing shows the content of the processing when the present invention is applied to the fifth embodiment. 6, the same processing can be added. When each router transfers the diagnosis result packet, the processing flow is almost the same as that in the fifth embodiment. In step 655, the dynamic state information and the value of the attribute information acquisition designation 600 are acquired from the payload, and step 656 is performed. In step 657, the items specified by the dynamic status information and attribute information acquisition specification 600 are obtained from the items of the router one stage downstream, and the items specified by the dynamic status information and attribute information acquisition specification 600 in step 657. And a threshold value preset in the own router by the administrator. The other processes are the same as those in the first, second, third, fourth, fifth, and sixth embodiments, and thus the description thereof is omitted.

In the above description, the diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and the diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to carry out in the flow of.

By using the present invention, dynamic status information and attribute information acquired by each router and added to the payload of the diagnostic packet can be dynamically specified for routers existing on the data delivery path 11. In addition, the dynamic state information and attribute information of the one-stage downstream router recorded in the payload of the diagnosis result packet and the dynamic state information and attribute information preset in the own router by the router administrator are stored. It is possible to dynamically determine the comparison items between the threshold value and the optimum data delivery route that prevents transfer delay and data loss by changing the comparison target of the dynamic state information and attribute information. A network system can be constructed.

Embodiment 8 FIG. Hereinafter, an eighth embodiment of the present invention will be described with reference to FIGS. 29 and 30. FIG.
9 shows a format of a diagnostic packet transmitted from the host A1 to the host B2 in the present embodiment, and FIG. 30 shows a format of a diagnostic result packet transmitted from the host B2 to the host A1 in the present embodiment. ing.

The diagnostic packet shown in FIG. 29 is different from the diagnostic packet described in the fifth embodiment in that the dynamic state information and attribute information threshold 700 at the beginning of the payload inside the host A1 are changed.
Is added. The threshold value 700 of the dynamic status information and the attribute information is set in advance by each router on the data delivery route 11 by a router administrator, and the type of the dynamic status information and the attribute information that must be acquired by the router. The minute threshold is stored. The threshold 700 of the dynamic state information and the attribute information is given when the host A1 transmits a diagnostic packet to the host B2. Then, the data delivery route 11
Each of the above routers does not refer to the dynamic state information and the threshold value 700 of the attribute information when transferring the diagnostic packet,
The data transfer processing described in the fifth embodiment is performed. In the present invention, the dynamic state information and the threshold value 700 of the attribute information are added to the diagnostic packet of the fifth embodiment. The same applies to the fourth, sixth, and seventh embodiments.

FIG. 30 shows the format of a diagnosis result packet in this embodiment. The diagnostic packet shown in FIG. 29 arrives at the host B2, and the payload 41 of the diagnostic packet is copied into the payload of the diagnostic result packet in the host B2 as described in the first embodiment. , The diagnostic result packet is transferred using the data delivery path 11. When the diagnostic packet passes through each router existing on the data transfer path, the router on the data delivery path indicating the flow of data from the host A1 to the host B2 transmits the dynamic state information and the threshold value 700 of the attribute information. , And compares the dynamic state information and attribute information of the router one stage downstream attached to the payload 41 with the set value of the threshold 700 of the dynamic state information and attribute information, Select whether to change the downstream data transfer path. In the present invention, the threshold value 70 of the dynamic state information and the attribute information is added to the diagnostic packet of the fifth embodiment.
0 is added, however, the first and second embodiments
In the third, fourth, fifth, sixth, and seventh embodiments, the present invention can be similarly added.

The internal processing when each router on the data transfer path transfers the diagnostic packet transmitted from the host A1 is as described in the first, second, third, fourth, and fourth embodiments. Since this embodiment is equivalent to the fifth, sixth, and seventh embodiments, the description thereof is omitted.

FIG. 31 shows the internal processing when each router on the data transfer path transfers the diagnosis result packet transmitted from the host B2. This processing shows the content of the processing when the present invention is applied to the fifth embodiment. Equivalent processing can be added to the sixth and seventh embodiments. When each router transfers the diagnosis result packet, the processing flow is almost the same as in the fifth embodiment.
At step 50, dynamic state information and attribute information thresholds are obtained from the payload, and at step 751, the C of the router one stage downstream is obtained.
A process of comparing the PU usage rate, the transmission / reception buffer usage rate, the average data transfer amount, and the maximum data transfer amount with threshold values of dynamic state information and attribute information acquired from the payload is performed. Other processes are the same as those in the first, second, third, fourth, fifth, sixth, and seventh embodiments, and thus the description thereof is omitted. .

In the above description, the diagnostic packet is transmitted from the transmitting host A1 to the receiving host B2, and the diagnostic result packet is transmitted from the receiving host B2 to the transmitting host A1. It is also possible to carry out in the flow of.

By using the present invention, the CPU usage rate, transmission / reception buffer usage rate, average data transfer amount, It is possible to compare the maximum data transfer amount with the threshold value of the dynamic status information and attribute information assigned to the diagnosis result packet,
By changing the thresholds of the dynamic state information and the attribute information at the time of transmitting each diagnostic packet, it becomes possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

Embodiment 9 FIG. Hereinafter, a ninth embodiment of the present invention will be described with reference to FIG. FIG. 32 shows the host A1 that generates / transmits the diagnostic packet, receives the diagnostic result packet, obtains the IP address of each router on the data delivery path 11 added to the payload of the diagnostic result packet, and executes the accumulation processing. It is an example of the internal structure.

Inside the host A1, the kernel area 801 and the user area 800 are processed separately by the OS running on the host. In the present embodiment, processing of the data transfer path diagnosis application 802 and other communication applications 803 operating in the user area 800 will be exemplified. A plurality of applications that perform processing other than these applications simultaneously operate on the host A1, but their description is omitted in the present embodiment. The data transfer path diagnosis application 802 includes therein a diagnosis packet generation unit 804, a diagnosis packet transmission unit 805, a diagnosis result packet reception unit 807, a diagnosis result accumulation unit 806, and a path search unit 809. The storage table 808 is managed.

Data transfer path diagnosis application 80
2 is an arbitrary timing, by the diagnostic packet generator,
Create a diagnostic packet to be transmitted to the data delivery route 11,
Thereafter, the diagnostic packet transmitting unit 805 transmits the generated diagnostic packet to the network. The format of the diagnostic packet generated by the diagnostic packet generator 804 is the same as that of the first, second, third, fourth, fifth, sixth, and seventh embodiments, and The format of the diagnostic packet shown in the eighth embodiment is adopted.

From the network to which the host A1 is connected, the first embodiment, the second embodiment, the third embodiment, the fourth embodiment, the fifth embodiment, the sixth embodiment, the seventh embodiment, and the seventh embodiment will be described. When the diagnosis result packet shown in the mode 8 arrives, the diagnosis result packet receiving unit 80 of the data transfer path diagnosis application 802
7 and then passed to the diagnosis result storage unit 806.
The diagnosis result storage unit 806 includes a diagnosis result packet receiving unit 807.
From the obtained payload of the diagnostic result packet, the IP address and dynamic state information and attribute information assigned by each router on the data delivery path 11 are extracted and recorded in the diagnostic result accumulation table as data transfer path information. By executing this process every time a diagnosis result packet is received, the diagnosis result accumulation table stores the data delivery path 11
The route information and the dynamic state information and the attribute information based on the IP address of each of the above routers are stored. Embodiment 1
The dynamic change processing of the data transfer path, which is performed in the second, third, fourth, fifth, sixth, seventh, and eighth and eighth embodiments, Regarding the updated data transfer path, dynamic state information and attribute information will be accumulated in the diagnosis result accumulation table 808 when the next diagnosis result packet is received.

Here, another communication application 803
Shows a case where data transmission is performed using a source routing method of an IP packet, which does not follow the data transfer path between the host A1 and the host B2 determined on the network. Other applications 803 are IP
Data transfer path diagnosis application 8 at the stage of creating a packet using the packet source routing method 8
02, a request is made to the route search unit 809 in the data transfer route to extract a route candidate whose dynamic state information and attribute information of each router is below a certain threshold. Upon receiving the path candidate extraction request from another communication application 803, the path search unit 809 sets the diagnosis result accumulation table 80
8, data transfer path candidates in which the dynamic state information and the attribute information of each router in the middle of the path are equal to or less than the threshold value are presented to another application 803. At the stage when the data transfer route candidate is presented by the route search unit 809,
The other communication application 803 determines the actual condition based on a determination condition other than the dynamic state information and attribute information of each router on the data transfer path (for example, a condition that a certain router must pass through the router). First, a data transfer path used in the source routing method is determined.

FIG. 33 shows the processing of the diagnostic packet generator 804 in the data transfer path diagnostic application 802. At an arbitrary timing during the operation of the data transfer path diagnosis application 802, a diagnostic packet is generated in step 850, and the diagnostic packet generated in step 851 is passed to the diagnostic packet transmitting unit 805, and the process ends. Next, FIG. 34 shows the processing of the diagnostic packet transmitting unit 805 inside the data transfer path diagnostic application 802.
In step 852, the diagnostic packet is received from the diagnostic packet generation unit 804, and in step 853, the diagnostic packet is transmitted to the network to which the host A1 is connected, and the process ends. Next, FIG. 35 shows the processing of the diagnosis result packet receiving unit 807 inside the data transfer path diagnosis application 802. In step 854, the diagnostic result packet is received from the network to which the host A1 is connected, and in step 855, the received diagnostic result packet is transferred to the diagnostic result storage unit 806, and the process is terminated. Next, FIG. 36 shows the processing of the diagnosis result storage unit 806 inside the data transfer path diagnosis application 802. In step 856, the diagnosis result packet receiving unit 807
Step 8:
At 57, the IP address of each router and the dynamic state information and attribute information recorded in the payload of the diagnosis result packet are extracted and added to the diagnosis result accumulation table 808, and the process is terminated. Next, FIG. 37 shows the processing of the route search unit 809 inside the data transfer route diagnosis application 802. In step 860, a data path candidate extraction request from another communication application 803 is received, and the diagnosis result is determined based on the threshold for the dynamic state information and attribute information of the router specified in the data path candidate extraction request received in step 861. Diagnosis result accumulation table 80 managed by the accumulation unit 806
8 to extract data path candidates by searching inside,
The information is presented to another communication application 803, and the process ends.

By using the present invention, the IP address of each router added to the payload of the diagnosis result packet created by each router on the data delivery path 11
The address and dynamic state information and attribute information are stored in the host A1.
And can be dynamically stored in another communication application that wants to perform packet transmission that does not follow a determined data transfer path by using a source routing method of an IP packet operating on the host A1. It becomes possible to provide a method for selecting an optimal data delivery route for preventing the transfer delay and the data loss when the state information and the attribute information become constant.

Although the data transfer path diagnosis application 802 operates as an application of the host A1 on the diagnostic packet transmission side in the present embodiment, the transmission host has the same function as the data transfer path diagnosis application 802. It is also possible to obtain the same effect by giving it to the kernel area 801 of A1. Further, in the present embodiment, the description has been given on the assumption that the function of the data transfer path diagnosis application 802 is provided on the host A1 on the diagnostic packet transmitting side, but the present embodiment is also provided on the host B2 on the diagnostic packet receiving side. May be provided, in which case the diagnostic packet can be transmitted from the host B2 as well.

Embodiment 10 FIG. Hereinafter, a tenth embodiment of the present invention will be described with reference to FIG. FIG. 38 shows a case where the periodic wake-up unit 900 is connected to the data transfer path diagnosis application 8.
2 shows an example of the internal structure of the host A1 provided in the host computer 02.

The periodic wake-up unit 900 operates when the data path diagnosis application 802 starts up or after the host A starts up.
The diagnostic packet generator 804 operates at regular intervals in response to an instruction from the first administrator, and performs a process of prompting the diagnostic packet generator 804 to generate a diagnostic packet. Thereby, the periodic wake-up unit 900 operates the diagnostic packet generation unit 804 at regular intervals,
The diagnostic packet generator 804 generates diagnostic packets at regular intervals, and the process of transmitting diagnostic packets to the network described in the ninth embodiment is performed at regular intervals. Based on this, the diagnostic result packet received from the network also arrives at the host A1 at regular intervals, and as a result, the diagnostic result accumulation table 8
At 08, the path information and the dynamic state information by the IP address of each router on the data distribution path 11 and the attribute information are stored by the diagnosis result packet receiving unit 807 and the diagnosis result storing unit 806. As a result, the contents of the diagnosis result accumulation table 808 are continuously updated at regular intervals, and a new data transfer path candidate is updated by the path search unit 809 in response to a data transfer path extraction request from another communication application 803. Can be presented. The processing of the other components is the same as that of the above-described ninth embodiment, and a description thereof will not be repeated.

FIG. 39 shows the processing of the periodic wake-up section 900 inside the data transfer path diagnosis application 802. At the start of the data transfer path diagnosis application 802 or at intervals set by the administrator of the host A1 after the start, the computer wakes up periodically at step 950 and prompts the diagnostic packet generator to generate a diagnostic packet at step 951. The process ends. Note that the processing contents of the other components are the same as in the above-described ninth embodiment, and a description thereof will be omitted.

By using the present invention, the IP address of each router added to the payload of the diagnosis result packet created by each router on the data delivery path 11
It is possible to periodically update the address and dynamic state information and attribute information to new information, and to use a source routing method of an IP packet operating on the host A1 to perform a packet that does not follow a predetermined data transfer path. It is possible to provide a method for selecting the most suitable data delivery route for keeping the latest dynamic state information and attribute information constant and preventing transfer delay and data loss for other communication applications that want to execute transmission.

In the present embodiment, the data transfer path diagnosis application 802 has been described as operating as an application of the host A1 on the diagnostic packet transmitting side. However, the transmission host has the same function as the data transfer path diagnosis application 802. It is also possible to obtain the same effect by giving it to the kernel area 801 of A1. Further, in the present embodiment, the description has been given on the assumption that the function of the data transfer path diagnosis application 802 is provided on the host A1 on the diagnostic packet transmitting side. However, in this embodiment, the function is provided on the host B2 on the diagnostic packet receiving side. The illustrated function may be provided, and in this case, the diagnostic packet can be transmitted from the host B2.

Here, the features of the present invention described above are summarized as follows. According to the present invention, in an IP network in which a plurality of routers are arranged between a sending host and a receiving host, and a data delivery route is determined by the router, the following means are provided to provide dynamic state information of a used route. The present invention relates to a method for constructing a network system that can create a route with constant information and attribute information, suppresses a data transfer delay caused by a dynamic change in route information and attribute information during data transfer, and determines an optimal data delivery route. (A) The transmitting host transmits a diagnostic packet for diagnosing a route to the receiving host at regular intervals. (B) When the router P1 existing on the data transfer path between the sending host and the receiving host transfers the diagnostic packet from the sending host to the router P2 downstream from the router P1,
The IP address of the router P1 and dynamic status information and attribute information are added to the payload (data portion) of the diagnostic packet. (C) Similarly, when transferring the diagnostic packet to the router P3 located downstream of the delivery route, the router P2 also attaches the IP address of the router P2 and dynamic state information and attribute information to the payload of the diagnostic packet. (D) As described above in (b) and (c), when the diagnostic packet to which the IP address of each of the routers on the route and the dynamic state information and the attribute information are added arrives at the receiving host, the packet is received. The host copies the IP address and dynamic state information and attribute information of the router in the middle of the route given to the payload of the diagnostic packet into the payload of the diagnostic result packet and returns it to the sending host. (E) Each router existing in the middle of the route looks at the payload of the diagnostic result packet, and can determine the dynamic state information and attribute information of the router downstream from the own router on the route from the sending host to the receiving host. If the dynamic status information and the attribute information can be detected and are equal to or more than a threshold value set in advance by the own router administrator, the route is changed so as to pass through another router.

According to the present invention, in the above-mentioned network system, the CPU utilization rate of the own router is used as the dynamic state information and the attribute information of the own router which are added to the payload of the diagnostic packet by the router existing on the route. By using the CPU usage rate of the downstream router as a factor that causes each router to change the route according to the diagnosis result packet, the router C
It is characterized in that data transfer can be performed with a transfer delay suppressed when the PU usage rate is high.

According to the present invention, in the above-described network system, the use rate of the transmission / reception buffer inside the router is used as the dynamic state information and attribute information of the own router which is attached to the payload of the diagnostic packet by the router existing on the route. Or add the usage rate of the transmission / reception buffer of the own router. The diagnostic result packet is characterized in that it is possible to prevent the data loss due to the overflow of the transmission / reception buffer by using the usage rate of the transmission / reception buffer of the downstream router as a factor that causes each router to change the route.

According to the present invention, in the above-mentioned network system, whether the average data transfer amount of the own router is used as dynamic state information or attribute information of the own router attached to the payload of the diagnostic packet by a router present on the way is determined. Or
Add the average data transfer amount of the own router. Diagnosis result packets prevent each router from changing the route by using the average data transfer volume of downstream routers to prevent transfer delays and data loss due to data congestion on the physical transmission path with downstream routers It is characterized by being able to do.

According to the present invention, in the above-mentioned network system, whether the maximum data transfer amount of the own router is used as the dynamic state information or attribute information of the own router attached to the payload of the diagnostic packet by the router existing on the way is determined. Or
Add the maximum data transfer amount of the own router. By using the maximum data transfer amount of the downstream router as a factor that causes each router to change the route by the diagnosis result packet, transfer delay and data due to peak data congestion on the physical transmission path with the downstream router It is characterized in that loss can be prevented.

According to the present invention, in the above-described network system, by providing the following means, a plurality of transfer routes existing between the sending host and the receiving host can be detected by each router on the route, and the plurality of routes can be detected. The present invention relates to a method for constructing a network system capable of selecting an optimum transfer path from among the above. (A) The diagnostic packet transmitted by the transmitting host to the receiving host is internally copied by the router on the route for the transmitting / receiving interface of the own router,
The diagnostic packet is sequentially transferred to the downstream router for all the transmitting and receiving interfaces. (B) Among the diagnostic packets distributed to a plurality of paths as described in (a) above, the diagnostic packet arriving at the receiving host is returned as a diagnostic result packet by the receiving host along the path that has been passed so far. Returned to the sending host. (C) Among the diagnostic packets distributed over a plurality of paths as in (a) above, the diagnostic packets that have not arrived at the receiving host due to the absence of a physical path, etc.
An error packet is returned from the router on the route to the sending host. (D) Based on the diagnostic result packet and the error packet returned to the transmitting host, each router existing on the intermediate route sees the payload of the packet, and thus a plurality of delivery routes exist between the transmitting host and the receiving host. Understand that. (E) Further, an optimum route that is equal to or less than a predetermined threshold is determined for the plurality of routes that have been grasped.

The present invention relates to a method for constructing a network system, which comprises the following means in the above-mentioned network system, and enables each router to be notified of a factor for changing a data delivery route. (A) The diagnostic packet transmitted by the transmitting host has a field for designating an element to be recorded in the diagnostic packet, out of dynamic state information and attribute information of the router. (B) Before transmitting the diagnostic packet to the receiving host, the transmitting host specifies dynamic status information and attribute information to be recorded in each router on the data delivery route. (C) The router that has received the diagnostic packet issued by the transmitting host in (b) transmits the dynamic state information and attribute information designated by the transmitting host, which are indicated in the diagnostic packet, to the IP address of the own router. It is recorded in the payload of the diagnostic packet together with the address.

[0139] The present invention relates to a network construction method which comprises the following means in the above-mentioned network system and enables each router to be notified of a threshold value for changing a data delivery route. (A) The diagnostic packet transmitted by the transmitting host has a field for designating a threshold for changing a route for dynamic state information and attribute information of the router. (B) Before transmitting the diagnostic packet to the receiving host, the transmitting host specifies a threshold value for a route change based on dynamic state information and attribute information for each router on the data delivery path. (C) When the diagnostic packet transmitted from the transmitting host to the receiving host is relayed by a router existing in the middle of the data delivery path, dynamic status information and attribute information are added to the payload in the same manner as described above. It is recorded along with the IP address of its own router. (D) The diagnostic packet arriving at the receiving host duplicates its payload into a diagnostic result packet and is returned to the transmitting host. (E) When a diagnostic result packet returned from the receiving host to the transmitting host is transferred by a router in the middle of the data delivery path, a threshold for dynamic state information and attribute information of the received result packet is recorded. By referring to the field, the own router knows the threshold for changing the route, uses that threshold, compares it with the dynamic state information and attribute information of the router downstream from the own router, and If the value is a value, the route is changed.

According to the present invention, in the above-described network system, when the transmitting host receives the diagnostic result packet returned from the receiving host to the transmitting host, the router on the route, which is recorded in the payload of the diagnostic packet, adds the diagnostic result packet. The sending host collects and manages the IP address and dynamic state information and attribute information. When an IP packet source routing request is issued from a user application running on the sending host, it is possible to select and specify the optimum route from the collected and managed IP addresses and dynamic status information and attribute information. It is characterized by the following.

According to the present invention, in the above-described network system, an IP address and dynamic status information and attribute information collected and managed by a transmitting host are received as a reply to a diagnostic packet issued at regular intervals by the transmitting host. By updating based on the diagnostic result packet returned from the host,
It is characterized in that it is possible to always select and designate an optimal route for a source routing request of an IP packet issued from a user application.

[0142]

By using the present invention, dynamic state information of each router which is located on a data transfer path defined between a sending host and a receiving host and performs data transfer processing, which cannot be achieved by the prior art. Changes in the data transfer route due to changes in data and attribute information. As a result, it is possible to create a route where the dynamic status information and attribute information of the router are constant, and the dynamic route information and attribute A data transfer delay caused by a change in information can be suppressed, and a network system that determines an optimal data delivery route can be constructed.

By using the present invention, among the routers present on the data transfer path, data transfer can be performed with a transfer delay suppressed when the CPU utilization of the router is high, and an optimum data delivery path is determined. It becomes possible to construct a network system.

By using the present invention, it is possible to prevent data loss due to overflow of the transmission / reception buffer, which occurs when the transmission / reception buffer usage rate of the router is high, among the routers existing on the data transfer path. It is possible to construct a network system that determines a delivery route.

By using the present invention, it is possible to prevent data congestion on a physical transmission path between a router and a router downstream of the router among routers existing on the data transfer path. Thus, it is possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

By using the present invention, it is possible to prevent data congestion during a peak delivery load on a physical transmission path between a router and a router downstream of the router among routers present on the data transfer path. This makes it possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

By using the present invention, a plurality of diagnostic packets are duplicated in advance on a data transfer path, and are forwarded to a downstream router connected to a communication interface of a certain router in advance, so that an optimal packet at that stage can be obtained. It is possible to determine the data delivery route, and if the physical transmission path is not connected to the final data transfer destination, it becomes possible to know that the data is unreachable, and as a result, the transfer delay It is possible to construct a network system that can determine an optimal data delivery route for preventing data loss or data loss at an early stage.

By using the present invention, dynamic status information and attribute information acquired by each router and added to the payload of the diagnostic packet can be dynamically specified for routers existing on the data transfer path. In addition, the dynamic state information and attribute information of the one-stage downstream router recorded in the payload of the diagnosis result packet and the threshold value of the dynamic state information and attribute information preset in the own router by the router administrator A network that determines the optimum data delivery route that prevents transfer delays and data loss by changing the comparison target of dynamic state information and attribute information A system can be constructed.

By using the present invention, the CPU usage rate, transmission / reception buffer usage rate, average data transfer amount, maximum It is possible to compare the data transfer amount and the threshold value of the dynamic status information and attribute information assigned to the diagnosis result packet, and change the threshold value of the dynamic status information and attribute information at the time of transmitting each diagnostic packet. Thus, it is possible to construct a network system that determines an optimal data delivery route for preventing transfer delay and data loss.

By using the present invention, the IP address and dynamic status information and attribute information of each router attached to the payload of the diagnosis result packet created by each router on the data transfer path can be transmitted to the transmission host. Can be stored in the IP, and operates on the sending host.
Even for other communication applications that want to perform packet transmission that does not follow the determined data transfer path using the packet source routing method, the dynamic state information and attribute information are constant, preventing transfer delay and data loss It is possible to provide a method for selecting an optimal data delivery route to be performed.

By using the present invention, the IP address of each router and the dynamic status information and attribute information added to the payload of the diagnosis result packet created by each router on the data transfer path are periodically updated. Can be updated to new information, and using the source routing method of IP packets operating on the sending host,
For other communication applications that want to carry out packet transmission that does not follow the determined data transfer path, the latest dynamic state information and attribute information become constant, and an optimal data delivery path that prevents transfer delays and data loss It is possible to provide a method for making the selection.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of a data communication system according to the present invention.

FIG. 2 is a diagram showing an example of a format of a diagnostic packet according to the first embodiment.

FIG. 3 is a flowchart showing a diagnostic packet transfer process according to the first embodiment;

FIG. 4 is a diagram showing an example of a format of a diagnosis result packet according to the first embodiment.

FIG. 5 is a flowchart showing a diagnosis result packet transfer process according to the first embodiment;

FIG. 6 is a diagram showing an example of a format of a diagnostic packet according to the second embodiment.

FIG. 7 is a diagram showing an example of a format of a diagnosis result packet according to the second embodiment.

FIG. 8 is a flowchart showing a diagnostic packet transfer process according to the second embodiment.

FIG. 9 is a flowchart illustrating a process of transferring a diagnosis result packet according to the second embodiment;

FIG. 10 is a diagram showing an example of a format of a diagnostic packet according to the third embodiment.

FIG. 11 is a diagram showing an example of a format of a diagnosis result packet according to the third embodiment.

FIG. 12 is a flowchart showing a diagnostic packet transfer process according to the third embodiment;

FIG. 13 is a flowchart showing a diagnosis result packet transfer process according to the third embodiment;

FIG. 14 is a diagram showing an example of a format of a diagnostic packet according to the fourth embodiment.

FIG. 15 is a diagram showing an example of a format of a diagnosis result packet according to the fourth embodiment.

FIG. 16 is a flowchart showing a diagnostic packet transfer process according to the fourth embodiment.

FIG. 17 is a flowchart showing a process of transferring a diagnosis result packet according to the fourth embodiment.

FIG. 18 is a diagram showing an example of a format of a diagnostic packet according to the fifth embodiment.

FIG. 19 is a diagram showing an example of a format of a diagnosis result packet according to the fifth embodiment.

FIG. 20 is a flowchart showing a diagnostic packet transfer process according to the fifth embodiment.

FIG. 21 is a flowchart showing a diagnosis result packet transfer process according to the fifth embodiment.

FIG. 22 illustrates a part of the data communication system according to the sixth embodiment.

FIG. 23 is a flowchart showing a diagnostic packet transfer process according to the sixth embodiment.

FIG. 24 is a flowchart showing processing for an error packet according to the sixth embodiment.

FIG. 25 is a diagram showing an example of a format of a diagnostic packet according to the seventh embodiment.

FIG. 26 is a diagram showing an example of a format of a diagnosis result packet according to the seventh embodiment.

FIG. 27 is a flowchart showing a diagnostic packet transfer process according to the seventh embodiment.

FIG. 28 is a flowchart showing a diagnosis result packet transfer process according to the seventh embodiment.

FIG. 29 is a diagram showing an example of a format of a diagnostic packet according to the eighth embodiment.

FIG. 30 is a diagram showing an example of a format of a diagnosis result packet according to the eighth embodiment.

FIG. 31 is a flowchart showing a diagnosis result packet transfer process according to the eighth embodiment.

FIG. 32 is a diagram showing a configuration of a transmission host according to the ninth embodiment.

FIG. 33 is a flowchart showing the processing of the diagnostic packet generator according to the ninth embodiment.

FIG. 34 is a flowchart showing processing of a diagnostic packet transmitting unit according to the ninth embodiment.

FIG. 35 is a flowchart showing processing of a diagnosis result packet receiving unit according to the ninth embodiment;

FIG. 36 is a flowchart showing a process of a diagnosis result accumulating unit according to the ninth embodiment;

FIG. 37 is a flowchart showing processing of a route search unit according to the ninth embodiment.

FIG. 38 is a diagram showing a configuration of a transmission host according to the tenth embodiment.

FIG. 39 is a flowchart showing processing in a periodic wake-up section according to the tenth embodiment.

FIG. 40 is a diagram showing a configuration of a router.

FIG. 41 is a view showing a conventional technique 1 (RIP).

FIG. 42 is a view showing a conventional technique 1 (HELLO protocol).

FIG. 43 is a view showing Conventional Technique 1 (OSPF protocol).

FIG. 44 is a view showing a conventional technique 2;

FIG. 45 is a view showing a conventional technique 3;

[Explanation of symbols]

1 sending host, 2 receiving hosts, 3 routers, 4 routers, 5 routers, 6 routers, 7 routers, 8 routers, 9 routers, 10 routers, 11 data delivery routes, 12 changed data delivery routes, 800 user areas, 801 Kernel area, 802 Data transfer path diagnosis application, 803 Other communication applications, 804 Diagnostic packet generator, 805 Diagnostic packet transmitter, 806 Diagnostic result storage, 807 Diagnostic result packet receiver, 808 Diagnostic result accumulation table, 80
9 route search section, 900 cycle wake-up section, 3000 packet transmission / reception section, 3100 packet type determination section, 320
0 normal packet processor, 3300 diagnostic packet processor, 3310 dynamic state information / attribute information detector, 332
0 diagnostic packet processing section, 3330 dynamic state information / attribute information provision section, 3340 own IP address provision section, 34
00 diagnosis result packet processing unit, 3410 downstream router IP address search unit, 3420 downstream router dynamic state information / attribute information acquisition unit, 3430 downstream router judgment unit, 3
440 threshold recording unit, 3450 downstream router selection unit, 3
500 routing table.

Claims (23)

[Claims] 1. A data transmission apparatus comprising: a data transmission apparatus; a data reception apparatus; and a plurality of data relay apparatuses disposed between the data transmission apparatus and the data reception apparatus. A data communication system including a plurality of intra-route data relay devices included in a data relay route set between a data transmission device and the data reception device, wherein any of the data transmission device and the data reception device One of them generates processing performance notification information indicating the data processing performance of each intra-path data relay device, and directs the generated processing performance notification information to the other of the data transmitting device and the data receiving device. Transmitting the processing performance notification information sequentially among the plurality of intra-route data relay devices according to the data relay route; Receiving the transferred processing performance notification information, and determining, based on the received processing performance notification information, a data processing performance of a set relay destination data relay device that is set as its own data relay destination in the data relay path by a predetermined amount. A data communication system, which determines whether or not the level is suitable, and selects its own data relay destination based on the determination result. 2. An at least one off-route data relay device not included in the data relay route is connected to one of the plurality of on-route data relay devices. Any of the off-route data relay devices, based on the determination result of the determination on the data processing performance of the set relay destination data relay device, the data relay destination of its own, the off-route data relay device of any of the off-route data relay device The data communication system according to claim 1, wherein the data communication system is changed to a relay device. 3. The data processing performance of the in-path data relay device, wherein one of the data transmitting device and the data receiving device is transmitted from the other of the data transmitting device and the data receiving device. 2. The data communication system according to claim 1, further comprising: receiving processing performance investigation information for examining the processing performance; and generating the processing performance notification information based on the received processing performance investigation information. 3. 4. The other of the data transmitting device and the data receiving device generates the processing performance research information and transmits the generated processing performance research information to one of the data transmitting device and the data receiving device. To cause the plurality of intra-path data relay apparatuses to sequentially transfer the processing performance inspection information according to the data relay path, and each of the intra-path data relay apparatuses receives the transferred processing performance inspection information. Described its own data processing performance in the received processing performance investigation information,
Transferring the processing performance investigation information describing its own data processing performance to another in-route data relay device, wherein one of the data transmitting device and the data receiving device is between the plurality of in-route data relay devices; Receiving the processing performance investigation information sequentially described in the above, and describing the data processing performance by each of the intra-path data relay devices, and generating the processing performance notification information based on the received processing performance investigation information. The data communication system according to claim 3. 5. The in-path data relay device has a processing performance threshold value that is a threshold value for data processing performance, and determines whether the data processing performance of the set relay destination data relay device matches the processing performance threshold value. The data communication system according to claim 1, wherein the determination is made as follows. 6. One of the data transmission device and the data reception device generates processing performance notification information including a processing performance threshold value that is a threshold value related to data processing performance, and generates the generated processing performance notification information. Is transmitted to the other of the data transmitting device and the data receiving device, and the processing performance notification information is sequentially transferred between the plurality of data relay devices in the route according to the data relay route. The relay device receives the transferred processing performance notification information, and determines whether or not the data processing performance of the set relay destination data relay device matches the processing performance threshold included in the received processing performance notification information. The data communication system according to claim 1, wherein the determination is performed. 7. The in-path data relay device transmits processing performance investigation information for investigating data processing performance of each out-of-path data relay device to each of the out-of-path data relay devices, and Receives processing performance notification information indicating the data processing performance of each out-of-path data relay device from each of the out-of-path data relay devices as a response to the performance investigation information, and, based on the received processing performance notification information, The data communication system according to claim 2, wherein it is determined whether or not the data processing performance of the external data relay device conforms to a predetermined level. 8. One of the data transmission device and the data reception device generates processing performance notification information indicating a plurality of types of data processing performance, and transmits the generated processing performance notification information to the data transmission device. Transmitting to the other of the device and the data receiving device, causing the processing performance notification information to be sequentially transferred among the plurality of in-route data relay devices according to the data relay route, Receiving the transferred processing performance notification information and determining whether each of the plurality of types of data processing performance of the set relay destination data relay device conforms to a predetermined level based on the received processing performance notification information The data communication system according to claim 1, wherein the data communication is performed. 9. The other of the data transmitting device and the data receiving device generates processing performance research information for designating a specific data processing performance among a plurality of types of data processing performance as a research target. The data communication system according to claim 4, wherein 10. The other of the data transmitting device and the data receiving device receives the processing performance notification information sequentially transferred among the plurality of intra-route data relay devices, and receives the received processing performance notification information. The data communication system according to claim 1, wherein a predetermined data relay path is set between the data transmitting device and the data receiving device based on the data transmission route. 11. The other of the data transmitting device and the data receiving device transmits the processing performance investigation information at a predetermined interval to one of the data transmitting device and the data receiving device. The data communication system according to claim 4, wherein: 12. Each of the intra-route data relay devices receives processing performance notification information indicating a CPU usage rate of each of the intra-route data relay devices, and performs the setting relay based on the received processing performance notification information. CPU of destination data relay device
The data communication system according to claim 1, wherein it is determined whether the usage rate conforms to a predetermined level. 13. Each of the intra-route data relay devices receives processing performance notification information indicating a usage rate of a transmission / reception buffer of each of the intra-route data relay devices, and based on the received processing performance notification information, The data communication system according to claim 1, wherein it is determined whether or not the usage rate of the transmission / reception buffer of the set relay destination data relay device conforms to a predetermined level. 14. Each of the intra-route data relay devices receives processing performance notification information indicating an average data transfer amount of each of the intra-route data relay devices, and performs the setting based on the received processing performance notification information. The data communication system according to claim 1, wherein it is determined whether an average data transfer amount of the relay destination data relay device conforms to a predetermined level. 15. Each of the intra-route data relay devices receives processing performance notification information indicating a maximum data transfer amount of each of the intra-route data relay devices, and performs the setting based on the received processing performance notification information. The data communication system according to claim 1, wherein it is determined whether or not the maximum data transfer amount of the relay destination data relay device conforms to a predetermined level. 16. A data relay device connected to a plurality of data communication devices and selecting a data relay destination from the plurality of data communication devices, wherein the data relay device is selected from a specific data communication device among the plurality of data communication devices. A processing performance notification information receiving unit that receives processing performance notification information indicating the data processing performance of the specific data communication device, based on the processing performance notification information received by the processing performance notification information receiving unit, A processing performance determining unit that determines whether the data processing performance of the data communication device conforms to a predetermined level; and selecting a data relay destination from the plurality of data communication devices based on a determination result by the processing performance determining unit. A data relay device comprising: a relay destination selecting unit. 17. The relay destination selecting unit, as a data relay destination, any one of the plurality of data communication devices other than the specific data communication device based on a result of the determination by the processing performance determining unit. 17. The data relay device according to claim 16, wherein the data relay device is selected. 18. The data relay device further includes a processing performance threshold recording unit that records a processing performance threshold value that is a threshold value related to data processing performance, wherein the processing performance determination unit is configured to store data of the specific data communication device. 17. The data relay device according to claim 16, wherein it is determined whether a processing performance matches the processing performance threshold. 19. The processing performance notification information receiving unit receives, from each of the plurality of data communication devices, processing performance notification information for notifying the data processing performance of each data communication device; Based on the processing performance notification information received by the processing performance notification information receiving unit, determine whether the data processing performance of each data communication device conforms to a predetermined level, the relay destination selection unit, 17. The data relay device according to claim 16, wherein a data relay destination is selected from the plurality of data communication devices based on a determination result by the processing performance determining unit. 20. A data relay method connected to a plurality of data communication devices and selecting a data relay destination from the plurality of data communication devices, wherein the data relay device is configured to: A processing performance notification information receiving step of receiving processing performance notification information indicating data processing performance of a specific data communication device; and, based on the processing performance notification information received by the processing performance notification information receiving step, A processing performance determining step of determining whether the data processing performance of the data communication device conforms to a predetermined level; and selecting a data relay destination from the plurality of data communication devices based on a determination result of the processing performance determining step. A relay destination selecting step. 21. The relay destination selecting step, wherein a data communication device other than the specific data communication device among the plurality of data communication devices is set as a data relay destination based on a result of the determination by the processing performance determining step. The data relay method according to claim 20, wherein the data relay method is selected. 22. The data relay method further includes a processing performance threshold value recording step of recording a processing performance threshold value which is a threshold value relating to data processing performance, wherein the processing performance determining step includes a step of recording data of the specific data communication device. 21. The data relay method according to claim 20, wherein it is determined whether a processing performance matches the processing performance threshold. 23. The processing performance notification information receiving step: receiving, from each of the plurality of data communication devices, processing performance notification information for notifying the data processing performance of each data communication device; Based on the processing performance notification information received in the processing performance notification information receiving step, determine whether the data processing performance of each data communication device conforms to a predetermined level, the relay destination selection step, 21. A data relay destination is selected from the plurality of data communication devices based on a result of the determination in the processing performance determining step.
The data relay method described in 1.