WO2015182629A1 - 監視システム、監視装置及び監視プログラム - Google Patents
監視システム、監視装置及び監視プログラム Download PDFInfo
- Publication number
- WO2015182629A1 WO2015182629A1 PCT/JP2015/065156 JP2015065156W WO2015182629A1 WO 2015182629 A1 WO2015182629 A1 WO 2015182629A1 JP 2015065156 W JP2015065156 W JP 2015065156W WO 2015182629 A1 WO2015182629 A1 WO 2015182629A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- message
- monitoring system
- target device
- node
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0751—Error or fault detection not based on redundancy
- G06F11/0754—Error or fault detection not based on redundancy by exceeding limits
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
- G06F11/0709—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3495—Performance evaluation by tracing or monitoring for systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0805—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
- H04L43/0817—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/81—Threshold
Definitions
- the disclosed subject matter relates to a monitoring device and a monitoring program therefor.
- nodes In recent years, in a network in which a plurality of communication nodes (hereinafter referred to as “nodes”) are connected, a system in which nodes are black boxed and internal information such as CPU utilization cannot be used due to device specifications, operation standards, and the like has been known. Yes.
- Patent Document 1 discloses a technique related to a network troubleshooting framework for detecting and diagnosing a failure occurring in a network. According to the disclosed technique, a failure occurring in the network is detected roughly as follows. First, nodes that communicate with each other transmit data describing the behavior and configuration of a network configured by the node group to the manager node. The manager node has a network simulation function and estimates network performance based on the received data. Then, it is determined whether the estimated network performance is different from the network performance measured at each node. If they are different, determine one or more faults that may be the cause.
- Patent Document 2 describes “Data Processing System Modeling Unit” for modeling a target system using a mathematical model based on the birth and death process, and the performance value for the load amount on the target system. And a “Performance Measurement Calculation Unit” device that calculates and notifies based on the measured value of the service response time of the target system (for example, see claim 32).
- the manager node performs network simulation using network setting information transmitted from the node (see paragraphs [0007], [0008], [0009], and [0010], for example).
- the network setting information is information inside the node measured by the agent module operating at each node, and includes, for example, signal strength, traffic statistics, and routing table information (for example, paragraphs [0011], [0012], [0013], [0014]).
- Patent Document 1 does not disclose a method for detecting a network failure when network setting information cannot be measured or transmitted by each node.
- a node may be black-boxed according to the device specifications of the node, the network operation standard, or the like.
- the agent module cannot be installed on the node, and the manager node cannot acquire the network setting information of the node. Therefore, it is difficult for the manager node to perform network simulation using the network setting information.
- a monitoring system for detecting a node failure or a change in the state of a node from information input to an apparatus constituting a network system and information output from the apparatus.
- the performance of each node is estimated by measuring and analyzing transmission / reception traffic of one or more nodes.
- the performance of each node is further estimated several times and their changes are examined. When a change exceeding a predetermined range is detected for a certain node, it is detected as a failure of the node.
- a network TAP device (hereinafter referred to as a TAP device) is used for traffic measurement.
- a TAP device is a device that replicates a network signal and transmits it to a measuring device.
- the TAP device is installed at one or more locations in the network.
- the buffer amount of the node is estimated.
- the state outside the node for example, the traffic volume is measured.
- the information may be combined to predict the occurrence of congestion in the node. This makes it possible to predict the occurrence of congestion due to call loss or retransmission when burst traffic arrives.
- a node in which a failure has occurred may be specified by gradually narrowing down measurement points.
- the monitoring system includes a measurement unit and an analysis unit,
- the measurement unit measures traffic information related to the message using a device that monitors a message input to the target device and a message output from the target device,
- the analysis unit calculates one or more indicators based on the predetermined relational expression and the measured traffic information, and based on a comparison between one indicator or a plurality of indicators and a threshold value, It is characterized by detecting that the target device has changed to a specific state.
- the monitoring device includes a measurement unit and an analysis unit,
- the measurement unit measures traffic information related to the message using a device that monitors a message input to the target device and a message output from the target device,
- the analysis unit calculates one or more indexes based on the predetermined relational expression and the measured traffic information, and based on a comparison between one index or a plurality of index changes and a threshold value, It is characterized by detecting that the target device has changed to a specific state.
- Another aspect is a monitoring program that causes a computer to function as the monitoring device when executed by the computer.
- a monitoring system a monitoring apparatus, and a monitoring program that detect the state of a node from information input to a device configuring a network and information output from the device, and further use the detected state. Can do.
- FIG. 6 is a diagram illustrating a configuration example of association setting information according to Embodiment 1.
- FIG. 6 is a diagram illustrating a configuration example of a session table according to the first embodiment.
- FIG. 6 is a diagram illustrating a configuration example of state history information according to Embodiment 1.
- FIG. It is a figure which shows the hardware structural example of each apparatus of a monitoring system.
- 3 is a flowchart illustrating traffic analysis processing according to the first embodiment. 4 is a flowchart illustrating logical node sorting processing according to the first embodiment. 3 is a flowchart illustrating call loss extraction processing according to the first embodiment.
- FIG. 3 is a flowchart illustrating system state calculation processing according to the first and second embodiments. 3 is a flowchart illustrating system state determination processing according to the first and second embodiments.
- FIG. 10 is a diagram illustrating a configuration example of system configuration information according to the third embodiment. 10 is a flowchart illustrating a measurement priority control process according to the third embodiment. 10 is a flowchart illustrating selective signal processing according to the third embodiment. The schematic flowchart in a monitoring system is shown.
- the network monitoring system disclosed in this specification is a network monitoring system that monitors a network system, and the network system includes a plurality of nodes, and the nodes communicate with each other via the network. .
- the network monitoring system has various types of traffic from a low load to a high load based on limited measurement information when several types of communication traffic having different internal processing loads of the monitoring target system are input to the target system.
- a state calculation process is performed to calculate the response characteristics of the target system with respect to the load with a small amount of calculation.
- the network monitoring system is a precondition for classifying several types of communication traffic with different processing loads inside the monitored system into individual communication traffic so that modeling processing is not required in the state calculation processing. Process.
- the network monitoring system performs the above-described state calculation process for calculating a value indicating the internal state of the target system, for example, the maximum processing performance, in order to detect the occurrence of a failure in the monitored system.
- the network monitoring system detects a change in the value to determine that the internal state or configuration of the target system has changed, and performs state determination processing that outputs an alert.
- a bursty mass message is transmitted to the monitoring target system, and the message received by the target system cannot be stored in the buffer, and the transmitted message is discarded. Predict that early. Therefore, when the network monitoring system measures that a message in the target system has been sent, it stores the number of messages that are waiting to be processed in the target system, and the target system processes the message. When a message that will be transmitted later is not measured, it is determined that message discard has occurred in the target system, and the number of stored messages is also reported to the state calculation process. Process. In addition, the network monitoring system performs the state calculation process, which estimates the physical state of the target system, for example, the buffer size, using the number of staying messages at the time of message discard reported from the preprocessing. Do. The network monitoring system predicts that message discard due to buffer overflow will occur when an amount of communication traffic exceeding the buffer size estimated by the state calculation process is transmitted to the target system, and outputs an alert. Judgment processing is performed.
- the configuration information of the target system stored in advance is used. Sends instructions to the measurement device to increase the measurement frequency of communication traffic near the node that is logically close to the node that detected the state change, and to decrease the measurement frequency of other communication traffic
- the measurement priority control process is performed.
- the network monitoring system receives an instruction from the measurement priority control process, the network monitoring system performs a selective signal reception process that changes the measurement frequency according to the instruction.
- Embodiment 1 Next, Embodiment 1 will be described with reference to the drawings. Here, the embodiment is disclosed using an example of detecting the occurrence of a failure in the network system.
- FIG. 1 is a block diagram illustrating a configuration example of the network system 10 and the monitoring system 20.
- the network system 10 includes, for example, a plurality of nodes 11 (indicated as 11a to 11e as an example in FIG. 1) and a system manager 12 forming a network.
- the node 11 communicates with other nodes 11 via the network.
- the system manager 12 manages the node 11 group.
- the network system 10 further includes a plurality of TAP devices (network taps) 13 (shown as examples 13a to 13d in FIG. 1).
- the TAP device 13 duplicates a packet transmitted via the network at a predetermined measurement location of the network system 10, and is duplicated using, for example, the network cable 14 (shown as 14a to 14d as an example in FIG. 1) as a medium. This is a device for transmitting the received packet to the measurement unit 21 of the monitoring system 20.
- the monitoring system 20 includes, for example, one or more measurement units 21, pre-processing units (traffic report creation units) 22, and analysis units 23, respectively.
- the measurement unit 21, the preprocessing unit 22, and the analysis unit 23 are described as separate devices. However, each unit is physically or logically included in one physical device (monitoring device). It may be provided.
- the measurement unit 21, the preprocessing unit 22, and the analysis unit 23 may be referred to as a monitoring side, a preprocessing unit, and an analysis unit of the monitoring device, respectively.
- Each of the measurement unit and the analysis unit may be implemented as one device in the apparatus, for example, hardware. For example, it can be implemented as a DPI device with an analysis function.
- the measurement unit 21 monitors the network, intercepts communication data (message) transmitted / received between the nodes 11 of the network system 10 using the TAP device 13 or the like, and performs signal inspection processing 212 to detect the communication data. The contents are inspected, and inspection report data is transmitted to the preprocessing unit 22.
- the inspection report data includes, for example, protocol information (including a message destination IP address, transmission source IP address, interface information, and procedure information), measurement time (for example, date and time information when the message was intercepted), and association attributes. Information (such as IMSI (International Mobile Subscriber Identity)).
- protocol information including a message destination IP address, transmission source IP address, interface information, and procedure information
- measurement time for example, date and time information when the message was intercepted
- association attributes for example, date and time information when the message was intercepted
- Information such as IMSI (International Mobile Subscriber Identity)
- the interface information and procedure information will be described later in the description of the association setting information 221.
- the preprocessing unit 22 receives the inspection report data from the measurement unit 21, analyzes the inspection report data, calculates the communication traffic status of the network system 10 including one or more nodes 11, and calculates The state of communication traffic is transmitted to the analysis unit 23 as traffic report data.
- the communication traffic refers to communication data (message) transmitted / received by the node 11.
- it is a request signal and a response message of a control signal that communicates between a plurality of nodes 11 and an application protocol such as HTTP (Hypertext Transfer Protocol).
- HTTP Hypertext Transfer Protocol
- the unit of communication traffic data transmitted and received by the node 11 will be referred to as a message and described.
- a message received by the node 11 is called an arrival message, and a message to be transmitted is called a departure message.
- the message may be an IP packet.
- the traffic report data is summary information regarding messages transmitted / received by the node 11 and includes supplementary information regarding a residence time from when a node 11 receives a message to transmission to another node 11, retransmission, and call loss. Details of the contents of the traffic report data will be described later.
- the preprocessing unit 22 includes a storage unit that stores association setting information 221 and a storage unit that includes a session table 222. Either or both of the association setting information 221 and the session table 222 may be outside the preprocessing unit 22, and FIG. 1 shows an example in which the session table 222 is outside the preprocessing unit 22.
- Each storage unit of the association setting information 221 and the session table 222 may be a separate storage area of one storage device.
- FIG. 2 is a diagram illustrating a configuration example of the association setting information 221 according to the first embodiment.
- the association setting information 221 is setting information used for the logical node sorting process 224.
- the logical node sorting process 224 associates the arrival message with the departure message in each node 11 of the network system 10 and the processing load and processing flow from when the node 11 receives the arrival message to when the departure message is transmitted. This is a process of distinguishing the difference and sorting the associated arrival message and departure message sessions into different logical nodes according to the processing load and processing flow.
- the logical node and logical node sorting process 224 will be described later.
- the association setting information 221 is set in advance by an administrator or an operator.
- the association setting information 221 includes, for example, arrival message interface information 2211 and procedure information 2212 (collectively referred to as arrival message information), departure message interface information 2213 and procedure information 2214 (collectively referred to as departure message information),
- arrival message information 2211 and procedure information 2212 collectively referred to as arrival message information
- departure message interface information 2213 and procedure information 2214 collectively referred to as departure message information
- the attribute information 2215 is included as association information
- the processing type 2216 is included as a node model.
- Interface information (2211, 2213) is information indicating the type of communication standard between nodes 11.
- the procedure information (2212, 2214) is information indicating the processing contents included in the arrival message and the departure message.
- the association information attribute information 2215 is information used to associate an arrival message with a departure message.
- the interface information (2211, 2213) is “S1AP”. And information such as “S6a”. Further, the procedure information (2212, 2214) includes information such as “Attach Request” and “Create Session Request”.
- the attribute information 2215 includes information indicating the identification number of the mobile phone user, for example, called IMSI.
- the process type 2216 is identification information for distinguishing the difference in processing load and processing flow from when the arrival message is received by the node 11 to when the departure message is transmitted.
- “YYY_Q1” first processing type
- the processing type for the process of sending a departure message after inquiring to another node 11 is “YYY_Q2” (second processing type). If the inquired nodes are different, “YYY_Q2” may be further divided into a plurality of “YYY_Q2-1” and “YYY_Q2-2”.
- YYY is a character string indicating the type of the node 11, such as “MME”.
- MME the type of the node 11
- it may be classified according to the size of the delay time and may be assigned with different processing types, or may be classified with an appropriate granularity according to the processing contents at the node and attached with processing types. Good.
- FIG. 3 is a diagram illustrating a configuration example of the session table 222.
- the session table 222 is a table for managing the status of the preprocessing unit 22 associating the arrival message with the departure message as a session.
- the session table 222 includes one or more entries (session entries). Each entry in the session table 222 includes, as arrival message information, a measurement time 2220, interface information 2221, procedure information 2222, a retransmission flag 2223, and a staying residence time 2224. Each entry of the session table 222 includes measurement time 2225, interface information 2226, procedure information 2227, attribute information 2228, and a call loss flag 2229 as departure message information. Furthermore, each entry of the session table 222 includes physical node information 2230 and a processing type 2231 as logical node information.
- the measurement times (2220 and 2225) are areas for storing measurement time information included in the inspection report data.
- the interface information (2221 and 2226) is an area for storing the interface information (2211 or 2213) of the association setting information 221.
- the procedure information (2222 and 2227) is an area for storing the procedure information (2212 or 2214) of the association setting information 221.
- the resend flag 2223 is 2 when the measurement unit 21 measures the arrival message having the same content a plurality of times (that is, when the preprocessing unit 22 receives the inspection report data of the arrival message having the same content a plurality of times).
- the arrival message after the first time is determined to be a retransmitted message, and is an area to be stored as flag information.
- the arrival count 2224 is the number of messages remaining in the same logical node at the time when the arrival message is measured. That is, the number of message pairs in which the arrival message is measured but the departure message is not measured. In one example, the arrival count 2224 is a value obtained by counting the number of entries having the same logical node information in the session table 222.
- Attribute information 2228 is an area for storing attribute information 2215 of association setting information 221.
- the call loss flag 2229 does not receive the inspection report data of the corresponding departure message within a predetermined time (timeout time) even though the preprocessing unit 22 has received the inspection report data of the arrival message. In this case, it is determined that a call loss has occurred in the destination message destination node 11 (arrival message receiving node), and is stored as flag information.
- the flag information of the retransmission flag 2223 and the call loss flag 2229 is, for example, either a value indicating true (TRUE) or a value indicating false (FALSE).
- the processing at the physical node 11 is classified and managed as one or a plurality of logical nodes according to the processing type.
- the logical node information is information for identifying a node that processes an arrival message and outputs a departure message.
- the logical node information includes physical node information 2230 and a processing type 2231.
- the physical node information 2230 is information for physically identifying the device (hardware) of the node 11.
- the IP address of the node 11 is used.
- the destination IP address of the arrival message is used as the IP address of the node 11.
- the source IP address of the departure message may be used.
- the process type 2231 is the same information as the process type 2216 of the association setting information 221. Although details will be described later, the preprocessing unit 22 stores the value of the processing type 2216 of the entry retrieved from the association setting information 221 as the processing type 2231.
- the preprocessing unit 22 identifies a logical node by using a set of physical node information 2230 and a processing type 2231. For example, if the same node 11 receives two types of arrival messages and the processing types 2231 are different from each other, the preprocessing unit 22 has received the two types of arrival messages by logically separate logical nodes. Consider it a thing.
- the analysis unit 23 makes the same determination using the logical node information.
- the analysis unit 23 receives the traffic report data from the preprocessing unit 22, and uses the received traffic report data and a predetermined algorithm, one or more values indicating the performance and / or internal state of the network system 10. Is calculated as state information.
- the analysis unit 23 stores the history of the state information, calculates a change amount of one or more values of the state information from the history of the state information, and compares the change amount with a predetermined threshold value. As a result of the comparison, if the amount of change is equal to or greater than the threshold value, the analysis unit 23 determines that the network system 10 has changed to a specific state. A more detailed process of the analysis unit 23 will be described later.
- the analysis unit 23 includes a traffic report buffer 231 and a storage unit for state history information 233.
- the traffic report buffer 231 stores traffic report data.
- the state history information 233 will be described with reference to FIG.
- the state history information 233 includes, for example, management information 2331, physical node information 2332 and processing type 2333 as logical node information, message arrival number information 2334 as traffic information, maximum processing performance information 2335 as estimated state information, and buffer size 2336. And information including the predicted call loss number information 2337 is stored.
- the analysis unit 23 includes a storage area for the state history 233 separately for each logical node information (a set of physical node information and processing type) in order to make it easy to refer to the estimated state information for each logical node.
- the measurement time 2331 of the management information stores the measurement time extracted from the traffic report data.
- the physical node information 2332 and the processing type 2333 of the logical node information store the physical node information and the processing type of the logical node information extracted from the traffic report data.
- the message arrival number 2334 of the traffic information is the number of message arrivals counted based on the traffic report data.
- the maximum processing performance 2335, the buffer size 2336, and the predicted call loss number 2337 of the estimated state information estimated values obtained by the analysis unit 23 are stored. Note that the message arrival rate may be stored in addition to or instead of the number of message arrivals.
- FIG. 5 shows an example of the hardware configuration of each device such as the measurement unit 21, the preprocessing unit 22, and the analysis unit 23.
- These devices include a CPU (processing unit) 1001, a main storage device 1002, an external storage device 1005 such as an HDD, a reading device 1003 that reads information from a portable storage medium 1008 such as a CD-ROM or DVD-ROM, and a display.
- a computer including an input / output device 1006 such as a keyboard and a mouse, a communication device 1004 such as a NIC (Network Interface Card) for connecting to the network 19, and an internal communication line 1007 such as a bus connecting these devices. 1000. Note that some of the components may be omitted.
- the session table 222, the storage unit of the association setting information 221 and the storage unit of the state history information 233 can be realized by using a partial area of the main storage device 1002.
- Each device loads various programs stored in the external storage device 1005 to the main storage device 1002 and is executed by the CPU 1001, and is connected to the network 19 using the communication device 1004 as necessary.
- the communication device 1004 By performing network communication with other devices or receiving packets from the network TAP device 13, various processes and various types of storage in each embodiment can be realized.
- the program may be stored in advance in the external storage device 1005, or may be introduced from another device via the network 19 or the storage medium 1008 as necessary.
- the CPU of the preprocessing unit 20 executes each process of the traffic analysis process 223, the logical node sorting process 224, the call loss extraction process 225, and the report process 226 shown in FIG. Further, for example, the CPU of the analysis unit 23 executes each process of the system state calculation process 232, the system state determination process 234, and the measurement priority control process 236 shown in FIG. Note that the measurement priority control processing 236 is omitted in the first embodiment, and will be described in the third embodiment.
- Traffic analysis processing 2223 When the traffic analysis processing 223 receives the inspection report data from the measurement unit 21 in the preprocessing unit 22, the traffic analysis processing 223 extracts information necessary for session management in the session table 222, stores the information in the session table 222, and This is a process of creating traffic report data from information for analysis processing in the analysis unit 23 and transmitting the traffic report data to the analysis unit 23.
- FIG. 6 is a flowchart illustrating the process performed by the preprocessing unit 22 in the traffic analysis process 223.
- the preprocessing unit 22 obtains protocol information (message destination IP address, transmission source IP address, interface type, and procedure information), measurement time, and association attribute from the inspection report data received from the measurement unit 21.
- Information (such as IMSI) is extracted (step S11).
- the preprocessing unit 22 refers to the existing session table 222 using the extracted protocol information as a search condition, and searches for a session entry in which the protocol information matches the departure message information (step S12). For example, an entry whose interface type and procedure information match is specified. The new registration of the session table 222 will be described later.
- the preprocessing unit 22 calculates the difference between the measurement times of the arrival message and the departure message as the residence time (step S14).
- the case where there is a corresponding session entry in step S13 corresponds to, for example, the case where an arrival message received by a certain node 11 is processed and a corresponding departure message is output.
- the measurement time 2220 of the arrival message is stored in the corresponding session entry, and the measurement time in the inspection report data can be used as the measurement time of the departure message.
- the preprocessing unit 22 may store the measurement time in the inspection report data in the area of the measurement time 2225 of the departure message information in the session table 222.
- the calculated residence time is stored as appropriate in association with the logical node information, for example, and is read out at the time of traffic reporting.
- the preprocessing unit 22 transmits traffic report data related to the entry for which the session has ended to the analysis unit 23, deletes the corresponding session entry, and ends the processing (step S15).
- the traffic report data is summary information regarding messages transmitted and received by the node 11.
- the content of the traffic report data includes, for example, a measurement time, logical node information, a staying time, a staying number at arrival, a retransmission flag, and a call loss flag.
- the traffic report data measurement time includes the same information as the departure message information measurement time 2225 managed by the session table 222.
- the call loss time includes the time when the traffic report data is generated because there is no departure message.
- the logical node information of the traffic report data includes the same information as the physical node information 2230 and the processing type 2231 managed by the session table 222.
- the stay time of the traffic report data is the time that the message stays in the node 11 from when the node 11 receives the message until it is transmitted to another node 11, and is the calculation result of step S14.
- the number of stays at the arrival of traffic report data is the same information as the number of stays at arrival 2224 managed by the session table 222.
- the traffic report data retransmission flag is the same information as the retransmission flag 2223 managed by the session table 222.
- the call loss flag of the traffic report data is the same information as the call loss flag 2229 managed by the session table 222.
- step S13 the preprocessing unit 22 refers to the existing session table 222 using the protocol information extracted from the inspection report data as a search condition, and from the inspection report data. A session entry in which the extracted protocol information matches the arrival message information is searched (step S16).
- step S13 for example, when the node 11 receives an arrival message and then receives an arrival message with the same content in a state where the corresponding departure message is not transmitted, in other words, This corresponds to the case where a retransmission message is received.
- step S17 If there is a matching session entry in step S17 (step S17), the preprocessing unit 22 stores TRUE in the retransmission flag 2223 of the corresponding session entry (step S18), and ends the process.
- the preprocessing unit 22 creates a new session entry in the session table 222 (step S19).
- the preprocessing unit 22 stores the measurement time, interface type, and procedure information extracted from the inspection report data in the corresponding areas (2220 to 2222) of the arrival message information of the new session entry.
- the preprocessing unit 22 proceeds to the processing flow in the logical node sorting process 224 (step S20).
- the logical node sorting process 224 distinguishes the difference in processing load and processing flow from when the node 11 receives the arrival message to when the departure message is transmitted. This is a process for classifying sessions into different logical nodes according to the processing load and processing flow.
- FIG. 7 is a flowchart illustrating the processing performed by the preprocessing unit 22 in the logical node sorting processing 224.
- the preprocessing unit 22 confirms the completion of the new session entry creation step S19 (step S31).
- the preprocessing unit 22 matches the interface information 2211 of the arrival message information and the procedure information 2212 from the association setting information 221 using the combination of the interface information and procedure information of the protocol information extracted from the inspection report data as a search condition.
- the entry to be searched is searched (step S32).
- the preprocessing unit 22 sets the protocol information (including interface information 2213 and procedure information 2214) of the departure message of the entry of the matched association setting information 221 in the interface information 2226 and procedure information 2227 of the departure message information of the new session entry. (Step S33). Thereby, when inspection report data based on a departure message is subsequently received, it can be determined that there is a session entry that matches the departure message information in steps S12 and S13.
- the preprocessing unit 22 uses the inspection report to report information (specific identification number) corresponding to the attribute information 2215 (type information indicating IMSI in one example) specified in the association information of the entry of the matched association setting information 221. It is extracted from the attribute information for associating the data message, and is additionally stored in the attribute information 2228 of the departure message information of the new session entry (step S34).
- the preprocessing unit 22 stores the processing type 2216 of the entry of the matched association setting information 221 in the processing type 2231 of the logical node information of the new session entry (step S35).
- the preprocessing unit 22 stores the destination IP address included in the protocol information of the inspection report data in the physical node information 2230 of the logical node information of the new session entry (Step S36).
- the preprocessing unit 22 counts the number of session entries having the same logical node information (including a combination of the physical node information 2230 and the processing type 2231) from the session table 222, and uses the value as the number of stays at the arrival of a new session entry. It memorize
- the call loss extraction processing 225 did not receive the inspection report data of the corresponding departure message within the predetermined time (timeout time) even though it received the inspection report data of the arrival message in the preprocessing unit 22. In this case, it is determined that the call loss has occurred at the destination node 11 of the arrival message, and the determination criterion is stored in the corresponding session entry of the session table 222.
- FIG. 8 is a flowchart illustrating the process performed by the pre-processing unit 22 in the call loss extraction process 225.
- the preprocessing unit 22 repeats the next processing from the first session entry to the last session entry in the session table 222 (steps S41 and S44).
- the preprocessing unit 22 determines whether the current time exceeds the time obtained by adding a predetermined timeout time to the arrival message information measurement time 2220 (step S42).
- a predetermined timeout time is used as the predetermined timeout time. If exceeded, the preprocessing unit 22 stores TRUE in the call loss flag 2229 of the corresponding session entry, and transmits traffic report data to the analysis unit 23 (step S43). If not, skip the process and go to the next session entry.
- the analysis unit 23 stores the traffic report data in the traffic report buffer 231.
- the system state calculation processing 232 receives traffic report data from the preprocessing unit 22 in order to detect the occurrence of a failure for each logical node in the analysis unit 23, and from the information included in the traffic report data, the internal state of the logical node In one example, the maximum processing performance is calculated.
- FIG. 9 is a flowchart illustrating a process performed by the analysis unit 23 in the system state calculation process 232.
- the analysis unit 23 stores the state information in a temporary storage area.
- Step S54 and Step S55 in FIG. 9 are omitted. Steps S54 and S55 will be described in the second embodiment.
- the analysis unit 23 reads a plurality of buffered traffic report data from the traffic report buffer 231 every predetermined unit time (step S51).
- the unit time is, for example, a value on the order of seconds to several tens of seconds, and a value described in advance in the setting file is used.
- the analysis unit 23 sorts the traffic report data for each logical node information (a set of physical node information and processing type) included in the traffic report data, and for each logical node information, the following is performed based on the corresponding traffic report data. (A) and (b) are calculated (step S52).
- (A) Count the number of message arrivals of the corresponding traffic report data, divide by unit time, calculate the average value, and store the obtained average value as the message arrival rate Lambda of the status information.
- the counted number of message arrivals may be stored in the status information.
- the number of message arrivals corresponds to, for example, the number of traffic reports, but can be appropriately counted according to the transmission method of traffic report data.
- the corresponding traffic report data refers to the traffic report data within the unit time for the predetermined logical node information.
- the average value is calculated by dividing the total residence time included in the corresponding traffic report data by the number of message arrivals, and the obtained average value is stored as the average residence time W.
- the analysis unit 23 calculates the maximum processing performance Mu for each logical node information of the traffic report data based on the following relational expression, and stores it as the maximum processing performance Mu of the state information (step S53).
- the analysis unit 23 determines the measurement time extracted from the traffic report data, the number of message arrivals (and / or average message arrival rate Lambda) included in the state information, and the physical node of the logical node information extracted from the traffic report data.
- the maximum processing performance Mu of the information, the processing type, and the state information respectively, the measurement time 2331 (time rounded in unit time) of the state history information 233, the number of message arrivals (rate) 2334, and the logical node information
- the physical node information 2332, the processing type 2333, and the maximum processing performance 2335 of the estimated state information are stored (step S56), and the processing ends.
- the system state determination processing 234 determines that the internal state or configuration of the logical node has changed by detecting a change in the value indicating the internal state of the logical node calculated by the system state calculation processing 232 in the analysis unit 23. For example, it is a process of outputting an alert considering that a failure has occurred.
- FIG. 10 is a flowchart illustrating a process performed by the analysis unit 23 in the system state determination 234.
- the analysis unit 23 calculates the amount of change in the value of the maximum processing performance 2335 of the estimated state information for each logical node information (a combination of the physical node information 2332 and the processing type 2333) from the state history information 233 (step S61). ). Since the status information for each unit time is stored in the status history information 233, the analysis unit 23 can calculate the amount of change in the value of the maximum processing performance 2335 from the two most recent entries for the target logical node, for example. it can. An appropriate entry may be used in addition to the two most recent entries.
- the analysis unit 23 compares the change amount with a predetermined threshold value (step S62).
- a predetermined threshold value e.g., a value previously described in the setting file is used as the threshold value.
- step S63 If the amount of change is equal to or greater than a predetermined threshold (step S63), the analysis unit 23 determines that the state of the logical node has changed, and outputs a system alert to the system manager 12 (step S64). In the first embodiment, steps S65 to S67 are omitted. Steps S65 to S67 will be described in the second embodiment. On the other hand, when the amount of change is not equal to or greater than a predetermined threshold (step S63) and after execution of step S64, the system state determination process is terminated. In the above description, the change amount is used, but the change rate may be used.
- the target system when several types of communication traffic having different processing loads inside the target system are input to the target system, it is possible to create response characteristics of the target system for the processing of each communication traffic. . Further, general-purpose response characteristics of the target system can be created using limited measurement information without performing time-consuming modeling work. Furthermore, it is possible to detect a node communication failure or the like from the measurement information.
- the packet discard is estimated by estimating the physical configuration such as the buffer size of the target system (target node).
- the traffic report data includes a retransmission flag and a call loss flag. Further, the processing of the analysis unit 23 is different from that of the first embodiment. Other configurations and processes are the same as those in the first embodiment, and a description thereof will be omitted.
- the system state calculation processing 232 uses the call loss flag and the staying number on arrival included in the traffic report data received from the preprocessing unit 22 in the analysis unit 23, and the node 11 (logical node) This is a process of estimating the physical state of, for example, the buffer size. In addition, it is a process of outputting an alert by predicting that a large number of burst messages are transmitted to a certain logical node, and the received message is discarded without being able to store the received message in the buffer, and that the transmitted message is discarded.
- Embodiment 2 which the analysis unit 23 performs by the system state calculation process 232 is demonstrated.
- the analysis unit 23 stores the state information in a temporary storage area.
- step S51 to step S53 Since the processing from step S51 to step S53 is the same as that in the first embodiment, description thereof is omitted.
- the analysis unit 23 extracts logical node information (a combination of physical node information and processing type), a call loss flag, and a staying number on arrival from the traffic report data. And the analysis unit 23 calculates
- requires the minimum value of the staying number at the time of arrival for every logical node information from the traffic report data in which the call loss flag TRUE.
- a state in which the call loss flag is TRUE is a state in which a message has arrived but has not been output, and a part of the staying number on arrival may be discarded. This value is used as a predicted value of the buffer size on the assumption that packet discarding occurs even with the minimum number of staying arrivals obtained here.
- the analysis unit 23 stores the minimum value in the buffer size of the state information (Step S54).
- the buffer size is represented by the number of messages, but may be represented by other units.
- the analysis unit 23 determines whether the number of message arrivals exceeds the buffer size value stored in the status information for each logical node information (a set of physical node information and processing type) of the traffic report data. If exceeded, the excess number is stored in the predicted call loss number of the state information (step S55).
- the analysis unit 23 measures the measurement time extracted from the traffic report data (the time rounded in unit time), the number of message arrivals (and / or the average message arrival rate Lambda) included in the state information, and the logical node information.
- (Rate) 2334, physical node information 2332 of logical node information, processing type 2333, maximum processing performance 2335 of estimated state information, buffer size 2336, and predicted call loss number 2337 are stored (step S56), and processing is performed. finish.
- Steps S61 to S64 are the same as those in the first embodiment.
- the analysis unit 23 divides the message arrival number 2334 from the storage unit of the state history information 233 for each logical node information (a set of the physical node information 2332 and the processing type 2333) by a predetermined minute unit time.
- the number of message arrivals in minute time units is calculated, and the calculated value is compared with the buffer size 2336 (steps S65 and S66).
- the minute unit time is a time shorter than the unit time of step S51, and is, for example, about 100 microseconds to about 1 second, and uses a value described in advance in the setting file.
- the analysis unit 23 causes the message discard due to the microburst to occur in the logical node indicated by the set of the physical node information 2332 and the processing type 2333.
- a system alert indicating that there is a high possibility (or has occurred) is output to the system manager 12 (step S67).
- the system alert output to the system manager 12 may include a predicted call loss number 2337.
- the occurrence of congestion due to bursty traffic to the receiving side node can be detected as soon as possible.
- a large amount of bursty communication traffic is input to the target system instantaneously, it is possible to estimate the physical configuration of the target system necessary for estimating the packet discard status of the target system.
- the analysis unit 23 of the present embodiment further includes a system configuration storage unit 235 (see FIG. 1).
- the system configuration storage unit 235 is a storage area that manages the configuration of the network system 10. Further, the CPU of the analysis unit 23 further executes measurement priority control 236. Other configurations and processes are the same as those in the first embodiment, and a description thereof will be omitted.
- the system configuration storage unit 235 manages the system configuration of the network system 10 (node connection relationship) using a tree structure.
- the node (data node 2350) constituting the tree structure includes information regarding the node 11.
- Each data node 2350 includes physical node information 2351, TAP device information 2352, and network interface number 2353.
- the physical node information 2351 is information (similar to the physical node information 2230) for physically identifying the device of the node 11.
- the TAP device information 2352 is information for identifying the TAP device 13 corresponding to the node device 11.
- the network interface number 2353 is an area for storing the network interface number of the measurement unit 21 connected to the TAP device.
- the configuration information of the network system 10 is set (stored) in advance in the system configuration storage unit 235 by the administrator or operator of the network system 10.
- FIG. 12 is a flowchart illustrating the process of the third embodiment performed by the analysis unit 23 in the measurement priority control process 236.
- the analysis unit 23 confirms that a change in the state of a certain logical node (for example, the occurrence of a failure) has been detected in the system state determination processing 234 described in the above embodiment (step S71).
- a detection method the same method as in Embodiment 1 or 2 can be used.
- the analysis unit 23 uses the configuration of the network system 10 stored in the system configuration storage unit 235 to calculate the distance of each TAP device 13 to the node 11 to which the logical node that detected the state change belongs. Further, the network interface number of the measurement unit 21 to which each TAP device 13 is connected is extracted from the network interface number 2353 (step S72).
- the analysis unit 23 identifies one or a plurality of TAP devices 13 corresponding to data nodes closer than a predetermined distance, and measures the network interface number of the measurement unit 21 to which the TAP device 13 is connected.
- a control instruction including an instruction to increase the processing priority (measurement priority) and lower the measurement processing priority for the network interface number of the measurement unit 21 connected to the TAP device 13 at a distance farther than a predetermined distance.
- the data is transmitted to the measurement unit 21 (step S73), and the process ends.
- FIG. 13 is a flowchart illustrating the process of the third embodiment performed by the measurement unit 21 in the selective signal reception process 211.
- the measurement unit 21 receives a control instruction from the analysis unit 23 (step S81).
- the measurement unit 21 increases the measurement frequency for the network interface number having a high measurement priority in the selective signal reception 211. Further, the measurement frequency for the network interface number having a low measurement priority is reduced (step S82).
- the measurement unit 21 may appropriately select the data received from the TAP device 13 at a measurement frequency according to the control instruction described above (FIG. 311).
- the measurement unit 21 may output a measurement frequency change instruction to the corresponding TAP device 13 to change the transmission frequency from the TAP device 13.
- the measurement frequency of communication traffic near the measurement point where the failure is detected is increased, and the measurement frequency of other communication traffic is decreased.
- FIG. 14 shows a schematic flowchart in the monitoring system.
- the measurement unit 21 uses a device (a TAP device 13 in the example of FIG. 1) that monitors a message input to the target device (the node 11 in the example of FIG. 1) and a message output from the target device.
- the traffic information related to the message is measured.
- step S92 the analysis unit 23, based on the measured traffic information, the message arrival rate, which is the number of messages received per unit time, the message arrival time in the target device, the performance of the device, An index (maximum processing performance Mu in the above example) is obtained using a relational expression with the index representing the state.
- step S93 the analysis unit 23 detects that the target device has changed to a specific state based on the obtained change in the index.
- the monitoring system that monitors the network system
- the network system includes a plurality of nodes, The above node communicates with other nodes via the network
- the monitoring system includes a measurement unit, a preprocessing unit, and an analysis unit
- the measurement unit monitors the network, intercepts communication data transmitted and received by the network system, inspects the content of the communication data, transmits inspection report data to the preprocessing unit
- the pre-processing unit receives inspection report data from the measurement unit, analyzes the inspection report data, calculates a state of communication traffic of the network system including a node and / or a plurality of nodes, and calculates
- the communication traffic status is sent to the analysis unit as traffic report data
- the analysis unit is
- the traffic report data is received from the preprocessing unit, and the received traffic report data and a predetermined algorithm are used to obtain one or more values indicating the performance and / or internal state of the network system,
- As state information A history of the state information is stored, a change amount of one or a plurality
- Configuration example 3 When several types of communication traffic with different processing loads in the network system are input to the network system, the analysis unit can perform various loads from low load to high load based on limited measurement information. The response characteristics of the target system are calculated with a relatively small amount of calculation. The preprocessing unit sorts several types of communication traffic having different processing loads inside the network system into individual communication traffic.
- Configuration example 4 The analysis unit calculates one or a plurality of values indicating the internal state of the network system in order to detect the occurrence of a failure in the network system, and detects a change in the value, thereby detecting the internal state of the network system. It is determined that the configuration has changed, and an alert is output.
- Configuration example 5 When the preprocessing unit measures that a message in the network system has been transmitted, the preprocessing unit stores the number of staying messages waiting for processing in the network system, and the network system processes the message. If the message that would be transmitted after the measurement is not measured, it is determined that message discard has occurred in the network system, and the stored number of staying messages is also reported to the analysis unit.
- the analysis unit estimates the physical state (for example, buffer size) of the network system using the number of staying messages reported from the preprocessing unit at the time of message discard, and the estimated buffer size When an amount of communication traffic exceeding 1 is transmitted to the network system, it is predicted that message discard due to buffer overflow will occur, and an alert is output.
- the physical state for example, buffer size
- Configuration example 6 When the analysis unit detects that the state of the node of the network system has changed, communication traffic in the vicinity of the node that has detected the state change using the configuration information of the network system stored in advance. An instruction is transmitted to the measurement apparatus so as to increase the measurement frequency and decrease the measurement frequency of other communication traffic.
- the measurement unit When receiving the instruction from the analysis unit, the measurement unit changes the measurement frequency according to the instruction.
- Patent Document 2 In the technology disclosed in Patent Document 2 described above, “Data Processing System Modeling Unit” creates a performance model for the entire communication traffic to the target system.
- a performance model for the entire communication traffic to the target system.
- the traffic volume and ratio for each type may change.
- Patent Document 2 does not disclose a technique for individually creating a performance model.
- “Performance Measurement Calculation Unit” calculates the performance value for the load on the target system using the mathematical model of the target system modeled by “Data Processing System Modeling Unit”.
- the mathematical model of the target system is a model with different response characteristics depending on the load amount for the entire communication traffic. Therefore, the “Performance Calculation” device needs to measure the service response time with respect to the communication traffic amount of various loads from low load to high load on the target system.
- this disclosed technique is used for the purpose of detecting a system failure such as congestion in advance, there is a case where communication traffic that places a heavy load on the target system cannot always be measured in advance.
- the response characteristics of the target system can be estimated from the amount of communication traffic that does not cause the target system to be heavily loaded.
- Patent Document 2 creates a mathematical model of the target system for various loads, and thus it takes a very long time to complete the creation of a certain model.
- the viewpoint of the system administrator it is not desirable to take a long time before the target system can be monitored.
- the system monitoring is performed in the shortest possible preparation time, it is possible to grasp the response characteristics of the target system even from the amount of communication traffic that does not cause a high load on the target system. it can.
- general-purpose response characteristics of the target system can be estimated using limited measurement information without performing time-consuming modeling work.
- bursty traffic may be instantaneously transmitted to a certain node from another node or a group of nodes via the network.
- the receiving side node cannot receive a large amount of traffic and discards it. Thereafter, when a larger amount of traffic arrives at the receiving side node due to retransmission traffic from the transmitting side node, the receiving side node may fall into a congestion state due to high load. If congestion worsens, the receiving node may go down.
- Patent Document 2 “Data Processing System Modeling Unit” creates a performance model of the target system using a mathematical model. In order to incorporate the probability of packet discard in the target system into the model when a large amount of bursty communication traffic is input to the target system instantaneously, a model of the physical state such as the communication buffer size of the target system is required. Need to create. However, Patent Document 2 does not disclose a technique for creating a model of a physical state such as a communication buffer size of the target system.
- the occurrence of congestion due to bursty traffic to the receiving side node can be detected as soon as possible.
- a large amount of bursty communication traffic is input to the target system instantaneously, it is possible to estimate the physical configuration of the target system necessary for estimating the packet discard status of the target system.
- DPI Deep Packet Inspection
- the failure is detected at a measurement point where a monitoring target system is connected to a network so that a single DPI device can measure a plurality of points.
- each of the above-described configurations, functions, processing units, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit.
- Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor.
- Information such as programs, tables, and files that realize each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.
- control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
当該監視システムは、計測ユニットと、分析ユニットと、を備え、
計測ユニットは、対象装置に入力されるメッセージ及び該対象装置から出力されるメッセージを監視する装置を用いて該メッセージに関するトラフィック情報を計測し、
分析ユニットは、所定の関係式と、計測したトラフィック情報と、に基づき、1つ以上の指標を計算し、1つの指標、もしくは、複数の指標の変化と、閾値と、の比較に基づいて、該対象装置が特定の状態に変化したことを検知する、という特徴を備える。
当該監視装置は、計測部と、分析部と、を備え、
計測部は、対象装置に入力されるメッセージ及び対象装置から出力されるメッセージを監視する装置を用いて該メッセージに関するトラフィック情報を計測し、
分析部は、所定の関係式と、計測したトラフィック情報と、に基づき、1つ以上の指標を計算し、1つの指標、もしくは、複数の指標の変化と、閾値と、の比較に基づいて、該対象装置が特定の状態に変化したことを検知する、という特徴を備える。
まず、各実施の形態の概要を説明する。本明細書で開示するネットワーク監視システムは、ネットワークシステムを監視するネットワーク監視システムであって、ネットワークシステムは複数のノードを備え、ノードは、ネットワークを経由して、他のノードと相互に通信を行う。
次に、実施の形態1を、図面を参照して説明する。ここでは、ネットワークシステムの障害発生を検知する例を用いて実施の形態を開示する。
トラフィック解析処理223は、前処理ユニット22において、計測ユニット21から検査報告データを受信すると、セッションテーブル222でセッション管理を行うのに必要な情報を抽出し、当該情報をセッションテーブル222に記憶し、分析ユニット23での分析処理のための情報からトラフィック報告データを作成し、分析ユニット23に当該トラフィック報告データを送信する処理である。
論理ノード仕分け処理224は、前処理ユニット22において、ノード11が到着メッセージを受信してから出発メッセージを送信するまでの、処理負荷や処理フローの違いを区別し、関連付けした到着メッセージと出発メッセージのセッションを、処理負荷や処理フローに応じて異なる論理ノードに仕分ける処理である。
呼損抽出処理225は、前処理ユニット22において、到着メッセージの検査報告データを受信したにもかかわらず、対応する出発メッセージの検査報告データを、所定の時間(タイムアウト時間)内に受信しなかった場合に、到着メッセージの宛先のノード11で呼損が発生したと判断し、セッションテーブル222の該当するセッションエントリに判断基準を記憶する処理である。
システム状態計算処理232は、分析ユニット23において、論理ノード毎の障害発生を検知するため、前処理ユニット22からトラフィック報告データを受信し、当該トラフィック報告データに含まれる情報から、論理ノードの内部状態、一例では最大処理性能を計算する処理である。
システム状態判定処理234は、分析ユニット23において、システム状態計算処理232で算出した、論理ノードの内部状態を示す値の変化を検出することで、論理ノードの内部状態や構成が変化したことを判定し、例えば障害発生とみなしてアラートを出力する処理である。
次に、瞬間的に大量のバースト的通信トラフィックが対象システムに入力された場合に、対象システムのパケット廃棄の状況を推測する実施の形態について、図9及び図10を用いて説明する。例えば、対象システム(対象ノード)のバッファサイズなどの物理的な構成を推測してパケット廃棄を推測する。
本実施の形態のシステム状態計算処理232は、分析ユニット23において、前処理ユニット22から受信したトラフィック報告データに含まれる、呼損フラグ及び到着時滞留数を用いて、ノード11(の論理ノード)の物理的な状態、例えばバッファサイズなど、を推測する処理である。また、ある論理ノードにバースト的な大量メッセージが送信され、論理ノードが受信したメッセージをバッファに記憶しきれずに、送信されたメッセージが廃棄されたことを予測し、アラートを出力する処理である。
実施の形態3では、実施の形態1又は2の構成及び処理に加えて、ネットワークシステムのある計測地点で障害を検出した際に、障害を検出した計測地点の近辺の通信トラフィックの計測頻度を増加し、それ以外の通信トラフィックの計測頻度を減少させることで、障害の発生箇所を、効率的に絞り込む。本実施の形態について、図12、図13及び図11を用いて説明する。
以下、上述の監視システムの構成例を例示する。
図14は、監視システムにおける概略フローチャートを示す。
ネットワークシステムを監視する監視システムは、
上記ネットワークシステムは複数のノードを備え、
上記ノードは、ネットワークを経由して、他のノードと相互に通信を行うものであり、
上記監視システムは、計測ユニットと、前処理ユニットと、分析ユニットと、を備え、
上記計測ユニットは、上記ネットワークを監視して、上記ネットワークシステムが送受信する通信データを傍受し、当該通信データの内容を検査し、上記前処理ユニットに、検査報告データを送信し、
上記前処理ユニットは、上記計測ユニットから検査報告データを受信し、当該検査報告データを解析して、ノード、及び/又は、複数ノードを備える上記ネットワークシステムの、通信トラフィックの状況を計算し、計算した通信トラフィックの状況を、トラフィック報告データとして上記分析ユニットに送信し、
上記分析ユニットは、
上記前処理ユニットからトラフィック報告データを受信し、受信した当該トラフィック報告データと、所定のアルゴリズムと、を用いて、上記ネットワークシステムの性能及び/又は内部状態を示す、1つ又は複数の値を、状態情報として計算し、
当該状態情報の履歴を記憶し、状態情報の当該履歴から、当該状態情報の1つ又は複数の値の変化量を計算し、当該変化量と所定の閾値とを比較し、比較した結果、変化量が閾値以上であれば、上記ネットワークシステムが特定の状態に変化したことを検知する。
上記ネットワークシステム内での処理負荷が異なる数種類の通信トラフィックが、上記ネットワークシステムに入力されている場合に、分析ユニットは、限られた計測情報から、低負荷から高負荷となる様々な負荷に対する、対象システムの応答特性を比較的少ない計算量で計算する。前処理ユニットは、上記ネットワークシステムの内部の処理負荷が異なる数種類の通信トラフィックを、それぞれ個別の通信トラフィックに仕分ける。
上記分析ユニットは、上記ネットワークシステムの障害発生を検知するため、上記ネットワークシステムの内部状態を示す1つ又は複数の値を計算し、当該値の変化を検出することで、上記ネットワークシステムの内部状態や構成が変化したことを判定し、アラートを出力する。
上記前処理ユニットは、上記ネットワークシステムにあるメッセージが送信されたことを計測した際に、上記ネットワークシステムで処理待ちになっている滞留メッセージ数を記憶しておき、上記ネットワークシステムが当該メッセージを処理した後に本来送信するであろうメッセージが計測されなかった場合に、上記ネットワークシステムでメッセージ廃棄が発生したことを判定して、記憶した上記滞留メッセージ数も合わせて上記分析ユニットに報告する。
上記分析ユニットは、上記ネットワークシステムの上記ノードの状態が変化したことを検出した際に、予め記憶している上記ネットワークシステムの構成情報を用いて、状態変化を検出した上記ノードの近辺の通信トラフィックの計測頻度を増加し、それ以外の通信トラフィックの計測頻度を減少させるように、上記計測装置に指示を送信する。
以下、従来技術と比較した本実施の形態の効果について説明する。
Claims (15)
- 監視システムであって、
計測ユニットと、分析ユニットと、を備え、
前記計測ユニットは、対象装置に入力されるメッセージ及び該対象装置から出力されるメッセージに関するトラフィック情報を計測し、
前記分析ユニットは、
所定の関係式と、計測したトラフィック情報と、に基づき、1つ以上の指標を計算し、
前記指標、もしくは、前記指標の変化と、閾値と、の比較に基づいて、該対象装置が特定の状態に変化したことを検知する
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
計測した該対象装置毎のトラフィック情報を、該対象装置での処理種別に応じて1つもしくは複数の論理ノードに仕分ける処理ユニットをさらに備え、
前記分析ユニットは、該論理ノード毎に、1つもしくは複数の前記指標が変化したと判断した場合に、該論理ノードが特定の状態に変化したことを検知する
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
前記分析ユニットは、
該対象装置のバッファサイズの予測値を求め、
計測するトラフィック情報に基づくメッセージ数が、求められたバッファサイズの予測値を超えると、メッセージ廃棄のアラートを出力する
ことを特徴とする監視システム。 - 請求項3に記載の監視システムであって、
前記分析ユニットは、
計測したトラフィック情報に基づきメッセージの廃棄を判断し、
メッセージが廃棄されたときの前記対象装置におけるメッセージ滞留数をバッファサイズの予測値とする
ことを特徴とする監視システム。 - 請求項2に記載の監視システムであって、
前記分析ユニットは、
該論理ノードのバッファサイズの予測値を求め、
計測するトラフィック情報に基づくメッセージ数が、求められたバッファサイズの予測値を超えると、メッセージ廃棄のアラートを出力する
ことを特徴とする監視システム。 - 請求項5に記載の監視システムであって、
前記分析ユニットは、
計測したトラフィック情報に基づきメッセージの廃棄を判断し、
メッセージが廃棄されたときの該対象装置の論理ノードにおけるメッセージ滞留数をバッファサイズの予測値とする
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
前記分析ユニットは、
前記対象装置又は該対象装置の前記論理ノードが特定の状態に変化したことを検知すると、該対象装置から予め定められたネットワーク上の距離内にある他の対象装置のトラフィック情報計測頻度を上げる
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
前記関係式は、単位時間あたりの到着メッセージ数である、該対象装置へのメッセージ到着率と、該対象装置でのメッセージ滞留時間と、該対象装置の性能又は状態を表す指標と、の関係式である
ことを特徴とする監視システム。 - 請求項8に記載の監視システムであって、
前記関係式は、待ち行列理論に基づき予め定められ、以下の関係を満たす
ことを特徴とする監視システム:
Mu=Lambda+1/W
ここで、Muは対象装置の性能又は状態を表す指標、Lambdaは単位時間内のメッセージ数に基づく対象装置への平均メッセージ到着率、Wは単位時間内のメッセージについての対象装置での平均滞留時間である。 - 請求項1に記載の監視システムであって、
前記分析ユニットは、
前記計測ユニットが計測した前記トラフィック情報から前記閾値を生成する
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
前記分析ユニットは、
前記指標それぞれの履歴を記憶し、
前記履歴を用いて、前記指標のそれぞれの変化量を計算し、
当該変化量と予め記憶している前記閾値とを比較する
ことを特徴とする監視システム。 - 請求項1に記載の監視システムであって、
前記特定の状態への変化は、対象装置の障害の発生である
ことを特徴とする監視システム。 - 請求項2に記載の監視システムであって、
前記特定の状態への変化は、該論理ノードの障害の発生である
ことを特徴とする監視システム。 - 監視装置であって、
計測部と、分析部と、を備え、
前記計測部は、対象装置に入力されるメッセージ及び該対象装置から出力されるメッセージに関するトラフィック情報を計測し、
前記分析部は、
所定の関係式と、計測したトラフィック情報と、に基づき、1つ以上の指標を計算し、
前記指標、もしくは、前記指標の変化と、閾値と、の比較に基づいて、該対象装置が特定の状態に変化したことを検知する
ことを特徴とする監視装置。 - 計算機に実行させることにより、前記計算機を監視装置として機能させる監視プログラムであって、
前記監視装置は、
対象装置に入力されるメッセージ及び該対象装置から出力されるメッセージに関するトラフィック情報を計測し、
所定の関係式と、計測したトラフィック情報と、に基づき、1つ以上の指標を計算する処理と、
前記指標、もしくは、前記指標の変化と、閾値と、の比較に基づいて、該対象装置が特定の状態に変化したことを検知する処理と、を実行する
ことを特徴とする監視プログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016523520A JPWO2015182629A1 (ja) | 2014-05-30 | 2015-05-27 | 監視システム、監視装置及び監視プログラム |
US15/314,516 US20170206125A1 (en) | 2014-05-30 | 2015-05-27 | Monitoring system, monitoring device, and monitoring program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014113225 | 2014-05-30 | ||
JP2014-113225 | 2014-05-30 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015182629A1 true WO2015182629A1 (ja) | 2015-12-03 |
Family
ID=54698953
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/065156 WO2015182629A1 (ja) | 2014-05-30 | 2015-05-27 | 監視システム、監視装置及び監視プログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US20170206125A1 (ja) |
JP (1) | JPWO2015182629A1 (ja) |
WO (1) | WO2015182629A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019142414A1 (ja) * | 2018-01-19 | 2019-07-25 | 日本電気株式会社 | ネットワーク監視システム、方法及びプログラムを格納した非一時的なコンピュータ可読媒体 |
US11281830B2 (en) * | 2019-03-11 | 2022-03-22 | Intel Corporation | Method and apparatus for performing profile guided optimization for first in first out sizing |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11777834B2 (en) * | 2016-11-01 | 2023-10-03 | T-Mobile Usa, Inc. | IP multimedia subsystem (IMS) communication testing |
WO2019112498A1 (en) * | 2017-12-06 | 2019-06-13 | Telefonaktiebolaget Lm Ericsson (Publ) | Automatic transmission point handling in a wireless communication network |
CN116386340A (zh) * | 2023-06-06 | 2023-07-04 | 北京交研智慧科技有限公司 | 交通监测数据处理方法、装置、电子设备及可读存储介质 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010072955A (ja) * | 2008-09-18 | 2010-04-02 | Fujitsu Ltd | 監視装置、監視方法及びコンピュータプログラム |
WO2011074659A1 (ja) * | 2009-12-18 | 2011-06-23 | 日本電気株式会社 | 移動体通信システム、その構成装置、トラヒック平準化方法およびプログラム |
-
2015
- 2015-05-27 US US15/314,516 patent/US20170206125A1/en not_active Abandoned
- 2015-05-27 WO PCT/JP2015/065156 patent/WO2015182629A1/ja active Application Filing
- 2015-05-27 JP JP2016523520A patent/JPWO2015182629A1/ja not_active Withdrawn
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010072955A (ja) * | 2008-09-18 | 2010-04-02 | Fujitsu Ltd | 監視装置、監視方法及びコンピュータプログラム |
WO2011074659A1 (ja) * | 2009-12-18 | 2011-06-23 | 日本電気株式会社 | 移動体通信システム、その構成装置、トラヒック平準化方法およびプログラム |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019142414A1 (ja) * | 2018-01-19 | 2019-07-25 | 日本電気株式会社 | ネットワーク監視システム、方法及びプログラムを格納した非一時的なコンピュータ可読媒体 |
JPWO2019142414A1 (ja) * | 2018-01-19 | 2021-01-07 | 日本電気株式会社 | ネットワーク監視システム、方法及びプログラム |
JP7234942B2 (ja) | 2018-01-19 | 2023-03-08 | 日本電気株式会社 | ネットワーク監視システム、方法及びプログラム |
US11281830B2 (en) * | 2019-03-11 | 2022-03-22 | Intel Corporation | Method and apparatus for performing profile guided optimization for first in first out sizing |
Also Published As
Publication number | Publication date |
---|---|
US20170206125A1 (en) | 2017-07-20 |
JPWO2015182629A1 (ja) | 2017-04-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6097889B2 (ja) | 監視システム、監視装置、および検査装置 | |
JP5666685B2 (ja) | 障害解析装置、そのシステム、およびその方法 | |
WO2015182629A1 (ja) | 監視システム、監視装置及び監視プログラム | |
CN105827472B (zh) | 网络数据流类型检测方法及装置 | |
JP5767617B2 (ja) | ネットワーク障害検出システムおよびネットワーク障害検出装置 | |
JP4058038B2 (ja) | 負荷監視装置および負荷監視方法 | |
CN108322320B (zh) | 业务生存性分析方法及装置 | |
US10592327B2 (en) | Apparatus, system, and method for analyzing logs | |
CN104584483A (zh) | 用于自动确定服务质量降级的原因的方法和设备 | |
JP3957712B2 (ja) | 通信監視システム | |
JP2018148350A (ja) | 閾値決定装置、閾値決定方法及びプログラム | |
JP5963974B2 (ja) | 情報処理装置及び情報処理方法及びプログラム | |
JP5883770B2 (ja) | ネットワーク異常検知システム、および、分析装置 | |
CN114430383A (zh) | 探测节点筛选方法、装置、电子设备和存储介质 | |
WO2021103800A1 (zh) | 故障修复操作推荐方法、装置及存储介质 | |
CN104518896A (zh) | 基于内部网关协议路由介数的网络脆弱性分析方法和装置 | |
US11265237B2 (en) | System and method for detecting dropped aggregated traffic metadata packets | |
JP6432377B2 (ja) | メッセージログ除去装置、メッセージログ除去方法、及びメッセージログ除去プログラム | |
JP2006033715A (ja) | ネットワークe2e性能評価システムと方法およびプログラム | |
US10511502B2 (en) | Information processing method, device and recording medium for collecting logs at occurrence of an error | |
KR20110071425A (ko) | 동적 플로우 샘플링 장치 및 방법 | |
CN117093429B (zh) | 一种服务器稳定性评价方法及*** | |
WO2023093527A1 (zh) | 告警关联规则生成方法、装置、电子设备和存储介质 | |
US10031788B2 (en) | Request profile in multi-threaded service systems with kernel events | |
JP4112590B2 (ja) | 異なり数上位nキーの推定方法および推定システム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15799194 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016523520 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15314516 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15799194 Country of ref document: EP Kind code of ref document: A1 |