JP2015531552A - Flow information collection system, method and program - Google Patents

Abstract

Provided is a flow information collection system capable of collecting information on a flow at a low processing cost. The base node information storage unit 71 is a node that is a collection target of flow identification information or a part thereof and information of the flow, and is a node that can collect information on each packet transfer path corresponding to the flow. The correspondence relationship with the base node that is one or more determined nodes is stored. When the flow identification information or a part thereof is designated, the information acquisition means 72 identifies the base node corresponding to the flow identification information or a part thereof, and identifies the flow identification information or a part thereof from the base node Get flow information that matches.

Description

  The present invention relates to a flow information collection system, a flow information collection method, and a flow information collection program that collect information about a flow in order to monitor a flow in an open flow.

  As shown in Patent Document 1 and Non-Patent Documents 1 and 2, in recent years, a technique called OpenFlow has been proposed. OpenFlow captures communication as an end-to-end flow, and performs path control, failure recovery, load balancing, and optimization on a per-flow basis. The OpenFlow switch that functions as a forwarding node includes a secure channel for communication with the OpenFlow controller. The open flow switch operates according to the flow table. The OpenFlow controller instructs the OpenFlow switch to add or rewrite the flow table as appropriate. In the flow table, for each flow, a set of key information (FlowKey, matching key) to be compared with the packet header, an action (Action) defining the processing content, and flow statistical information (Stats) is defined. This set is called a flow entry. Also, a rule that defines an action for a received packet is determined by a combination of key information and action. Flow identification information is used as key information. The information contained in the packet header can identify the flow to which the packet belongs. Therefore, more specifically, information included in the header of the packet is used as key information.

  When a new packet is received, the OpenFlow switch searches the flow table for a flow entry that matches the identification information of the flow to which the packet belongs. As a result of the search, when a flow entry that matches the flow to which the received packet belongs is found, the OpenFlow switch executes the process described in the action field in the flow entry for the packet. On the other hand, if no entry matching the flow to which the received packet belongs is found as a result of the search, the OpenFlow switch forwards the received packet to the OpenFlow controller via the secure channel, and based on the header information of the received packet. Ask the OpenFlow controller to determine the route of the packet. Then, the OpenFlow switch receives a flow entry corresponding to the determined route from the OpenFlow controller, and updates the flow table.

  As described above, the OpenFlow switch determines the process for the packet based on the flow entry set in the OpenFlow controller. As processing for packets, OUTPUT that outputs packets to a specified interface is particularly frequently used. Note that the port specified at this time is not limited to a physical interface.

  In addition, the OpenFlow controller defines a flow entry for each node (OpenFlow switch) on the packet transfer path and sets a flow entry for each node in order to realize control of a flow. To do. The OpenFlow controller needs to be equipped with a flow control application for realizing such processing.

  Since the OpenFlow controller controls a plurality of flow entries (flow entries for each node) for one flow, each flow entry set along the transfer path of packets classified into the flow is held as one set. Need to manage.

  Therefore, when the number of flows in the network increases, the number of flow entries that the OpenFlow controller should hold increases.

  Also, the OpenFlow controller uses the OpenFlow message to actively or passively obtain the flow entry information set in the OpenFlow switch, and compares it with the flow entry information held by the OpenFlow controller itself. After that, the acquired information is notified to the route calculation application or discarded. Therefore, when the number of flow entries to be held by the OpenFlow controller increases as the number of flows increases, the load of the OpenFlow controller increases from the viewpoint of the amount of data and the load of message processing.

  If the OpenFlow controller does not hold the flow entry and it newly calculates the flow entry each time it changes the network status (for example, topology change) and sets it as a node, this load can be reduced. . However, in this case, an overhead for resetting all flow entries occurs, or strict end-to-end flow control is not performed. In addition, problems such as difficulty in monitoring the flow occur.

International Publication No. 2008/095010

Nick McKeown, 7 others, “OpenFlow: Enabling Innovation in Campus Networks”, [online], March 14, 2008, [Search February 26, 2010], Internet <http://www.openflowswitch.org //documents/openflow-wp-latest.pdf> “OpenFlow Switch Specification Version 0.9.0 (Wire Protocol 0x98)”, [online], July 20, 2009, [February 26, 2010 search], Internet <http://www.openflowswitch.org/documents /openflow-spec-v0.9.0.pdf>

  When a large number of flows are controlled by the open flows described in Patent Document 1 and Non-Patent Documents 1 and 2, the cost for managing the flow information increases. For example, a case where the OpenFlow controller attempts to acquire traffic statistical information as flow information is taken as an example. In this case, in the techniques described in Patent Document 1 and Non-Patent Documents 1 and 2, the OpenFlow controller needs to acquire statistical information from all OpenFlow switches on the packet communication path corresponding to the flow. The processing cost of the controller becomes high. The same applies when information other than traffic statistical information is collected as flow information.

  Therefore, an object of the present invention is to provide a flow information collection system, a flow information collection method, and a flow information collection program that can collect information about flows at a low processing cost.

  A flow information collection system according to the present invention includes a base node information storage unit that stores a correspondence relationship between flow identification information or a part thereof and a base node that is a node from which flow information is collected, and flow identification information or It is characterized by comprising information acquisition means for acquiring flow information from a base node corresponding to a part thereof.

  Also, the flow information collection method according to the present invention stores the correspondence between the flow identification information or part thereof and the base node that is the node from which the flow information is collected, and the flow identification information or part thereof is stored in the flow identification information or part thereof. Flow information is acquired from the corresponding base node.

  A flow information collection program according to the present invention is provided in a computer including a base node information storage unit that stores a correspondence relationship between flow identification information or a part thereof and a base node that is a node from which flow information is collected. An installed flow information collection program, characterized by causing a computer to execute an information acquisition process for acquiring flow information from a base node corresponding to flow identification information or a part thereof.

  According to the present invention, information about a flow can be collected at a low processing cost.

It is explanatory drawing which shows the example of the communication system containing a control apparatus (flow information collection system). It is a block diagram which shows the structural example of the flow information collection system of 1st Embodiment. It is a flowchart which shows the example of process progress at the time of the flow entry setting of 1st Embodiment. It is a schematic diagram which shows the example of a flow. It is a schematic diagram which shows the other example of a flow. It is a flowchart which shows the example of the process progress at the time of the flow entry setting of 2nd Embodiment. It is a schematic diagram which shows the example of a some flow. It is explanatory drawing which shows the information memorize | stored by step S11 regarding the flows 1 and 2 shown in FIG. It is a schematic diagram which shows a response | compatibility with the identification information of a group of a flow, and a base point node. It is a flowchart which shows the example of the process progress at the time of entry deletion notification reception. It is a flowchart which shows the example of progress of a process at the time of the entry deletion notification reception in 4th Embodiment. It is explanatory drawing which shows the example of the communication system with which the 5th Embodiment of this invention is applied. It is a block diagram which shows the structural example of the control apparatus of 5th Embodiment. It is a block diagram which shows the example of the minimum structure of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The flow information collection system of the present invention is realized by an open flow controller in open flow. Hereinafter, the flow information collection system of the present invention is referred to as a control device. An OpenFlow switch in OpenFlow is referred to as a node. The implementation of the present invention is not limited to the implementation in the open flow, and any technique may be used as long as it is a centralized control type network architecture.

  FIG. 1 shows an example of a communication system including a control device (flow information collection system). The communication system illustrated in FIG. 1 includes a control device 10 and nodes 1 to 3. The control device 10 is connected to each of the nodes 1 to 3 via the control secure channel 5. Further, the nodes 1 to 3 have a link with the outside of the communication system. Note that the number of nodes and the connection relationship between the nodes are arbitrary, and are not limited to the example shown in FIG.

  The control device 10 performs flow control spontaneously in response to inquiries from the nodes 1 to 3. Further, the control device 10 may perform flow control in response to a change in topology, an instruction from a user, registration of a new host, or the like.

  The control device 10 specifies flow identification information and calculates a transfer path of packets belonging to the flow. Then, the control device 10 identifies the base node from the nodes included in the route, and stores the base node (specifically, the base node identification information) and the flow identification information or a part thereof in association with each other. To do.

  Here, the base node will be described. The base node is a node that is a collection target of information about a flow, and is one or a plurality of nodes determined as nodes that can collect information on each packet transfer path corresponding to the flow.

  A packet belonging to a certain flow may be copied at a certain node and output from a plurality of ports. Therefore, the packet transfer path is not limited to one for one flow. For example, for one flow, the base node can be determined by designating one or more nodes constituting the packet transfer path for each packet transfer path corresponding to the flow. And when the node designated for every packet transfer path | route corresponding to one flow is one common node, there is one base node. Since it is preferable that the number of base nodes is small, it is preferable to use a common node for each packet transfer path corresponding to one flow as described above. In this case, for example, a node through which all packet transfer paths corresponding to the flow pass may be determined as a base node. Note that a plurality of base point nodes may be defined.

  Further, the control device 10 creates a flow entry including the flow identification information and an action defining which node the packet is transferred to for each node on the calculated packet transfer path. Then, a flow entry corresponding to the node is set. As a result, the flow is determined. The control device 10 itself may hold the flow entry set in each node, but may not hold it. When the control device 10 itself does not hold the flow entry set in each node, the control device 10 can reduce the management cost of the flow entry.

  Further, as described above, the control device 10 stores the base node and the flow identification information or a part thereof in association with each other. When the control device 10 collects information about a certain flow or a group of flows (for example, traffic statistical information), the control device 10 associates the information with the flow identification information (or a part thereof) corresponding to the flow or the group of flows. Collect information from the base node. Information collected in this way can be used as information over the entire packet forwarding path of a flow or group of flows.

Embodiment 1. FIG.
FIG. 2 is a block diagram illustrating a configuration example of the flow information collection system (control apparatus) according to the first embodiment of this invention. The control device 10 includes a topology management unit 11, a route calculation unit 12, a flow information acquisition unit 13, a flow entry calculation unit 14, a flow entry setting unit 15, a control message processing unit 16, and a node communication unit 17. Is provided.

  The topology management unit 11 collects and manages network topology information.

  The route calculation unit 12 calculates a packet transfer route corresponding to the flow.

  The flow information acquisition unit 13 collects information regarding the flow from the base node. Hereinafter, in the present embodiment, a case where traffic statistical information is collected as information about a flow will be described as an example. However, information other than traffic statistical information may be collected as information about a flow.

  The flow entry calculation unit 14 calculates a flow entry for each node on the packet transfer route based on the flow identification information and the packet transfer route calculated by the route calculation unit 12. For example, the flow entry calculation unit 14 may determine a flow entry by combining an action of outputting a packet from an output port for forwarding a packet to the next node and the flow identification information.

  The flow entry setting unit 15 requests the node to set the flow entry calculated by the flow entry calculation unit 14.

  The control message processing unit 16 converts the control content when each element in the control device 10 controls the node as a control message. Further, the control message processing unit 16 analyzes the control message received from the node.

  The node communication unit 17 is a communication interface that performs communication with a node.

  Accordingly, when each of the units 11 to 15 in the control device communicates with the node, the communication is performed via the control message processing unit 16 and the node communication unit 17. Hereinafter, in order to simplify the description, reference to the control message processing unit 16 and the node communication unit 17 is omitted.

  The flow information acquisition unit 13 includes a base node determination unit 131, an information acquisition unit 132, and a base node information storage unit 133.

  The base node determination unit 131 determines a base node from the packet transfer path corresponding to the flow.

  The base node information storage unit 133 stores, for each flow, base node identification information corresponding to the flow and flow identification information.

  The information acquisition unit 132 requests the base node for information (for example, traffic statistical information) regarding the flow corresponding to the base node, and acquires the information from the base node.

  The topology management unit 11, the route calculation unit 12, the flow information acquisition unit 13 (particularly, the base node determination unit 131 and the information acquisition unit 132), the flow entry calculation unit 14, the flow entry setting unit 15, and the control message processing unit 16 include, for example, It is realized by a CPU of a computer that operates according to a flow information collection program. In this case, the CPU reads the flow information collection program, and according to the program, the topology management unit 11, the route calculation unit 12, the flow information acquisition unit 13 (particularly, the base node determination unit 131 and the information acquisition unit 132), the flow entry calculation. It suffices to operate as the unit 14, the flow entry setting unit 15, and the control message processing unit 16. Each of these elements may be realized by separate hardware.

Next, the process progress of this embodiment is demonstrated.
First, processing progress at the time of flow entry setting will be described. FIG. 3 is a flowchart illustrating an example of processing progress at the time of setting the flow entry according to the present embodiment. For example, it is assumed that node 1 (see FIG. 1) receives a packet, and the flow entry corresponding to the identification information of the flow to which the packet belongs is not set in node 1. In this case, the node 1 sends the received packet to the control device 10.

  When the control device 10 receives the packet, the route calculation unit 12 specifies the identification information of the flow to which the packet belongs (step S1). As the flow identification information, for example, information included in a packet header can be used. For example, a combination of a packet source MAC (Media Access Control) address and a destination MAC address can be used as flow identification information.

  Further, the route calculation unit 12 refers to the transmission source and destination of the packet and the topology information managed by the topology management unit 11, and calculates the packet transfer route in the flow to which the packet belongs (step S2).

  Next, the flow entry calculation unit 14 calculates a flow entry for each node on the packet transfer path calculated in step S2 (step S3). The flow entry calculation unit 14 determines an action of outputting a packet from an output port for transferring a packet to the next node for each node on the packet transfer path, and combines the flow identification information and the action to set the flow for each node. An entry may be defined.

  Next, the flow entry setting unit 15 requests each node on the packet transfer path to set a flow entry determined for each node (step S4). Each node sets a flow entry in response to a request from the flow entry setting unit 15. As a result, the flow to which the packet received by the node 1 belongs is determined. An example of the flow is shown in FIG. FIG. 4 illustrates a case where a flow that follows the nodes 1, 2, and 3 is defined.

  There is also a flow in which a node on the transfer path copies a packet and the node outputs the packet from a plurality of ports. An example of this flow is shown in FIG. In the example shown in FIG. 5, there are a transfer route that follows the nodes 1, 2, and 3 and goes from the node 3 to the outside of the communication system, and a transfer route that follows the nodes 1 and 2 and goes from the node 2 to the outside of the communication system. That is, the transfer path corresponding to the flow shown in FIG. 5 is a path tree in which a plurality of paths are combined.

  After step S4, the base node determination unit 131 determines a base node corresponding to the flow (step S5). Hereinafter, an example of a method for determining the base node will be described.

  For example, the base node determination unit 131 determines the base node by designating one or more nodes constituting the packet transfer path for each packet transfer path corresponding to the flow. By designating the base node for each packet transfer path, all packet transfer paths corresponding to the flow can be monitored.

  In the flow illustrated in FIG. 4, since there is one packet transfer path corresponding to the flow, one or more nodes may be determined as the base node from the packet transfer path. There may be a plurality of base point nodes, but since it is better that the number of base point nodes is small, it is preferable to determine one base point node. Further, for example, in the flow illustrated in FIG. 5, there are two packet transfer paths corresponding to the flow. Therefore, one or more (preferably one) node is designated from each path, and each designated node is designated as a base node. It may be determined as When the node specified for each packet transfer path is different, there are a plurality of base nodes. However, in order to reduce the number of base nodes, it is preferable to specify a common node for each packet transfer path. For example, in the example shown in FIG. 5, it is preferable that the node 1 or the node 2 is a base node.

  In addition, the base node determination unit 131 may determine one or more (preferably one) nodes as base nodes from among the nodes through which all packet transfer paths corresponding to the flow pass, for example. For example, in the example shown in FIG. 5, the nodes through which two packet transfer paths pass are nodes 1 and 2. Accordingly, the base node determination unit 131 may determine a base node from among the nodes 1 and 2. In the example shown in FIG. 4, since there is one packet transfer path, a base node may be determined from among nodes 1 to 3 on the packet transfer path.

  As another method for determining the base node, there is a method in which a node that first receives a packet from outside the communication system (hereinafter referred to as an ingress edge) is determined as the base node. Depending on the flow, there may be a plurality of ingress edges. In this case, the base node determination unit 131 may set all the ingress edges as base nodes. In the example illustrated in FIG. 4, the node 1 corresponding to the ingress edge may be the base node.

  In addition to the ingress edge, the base node determination unit 131 may determine a node that outputs a packet last in the communication system (hereinafter referred to as an egress edge) as a base node. In the example illustrated in FIG. 4, the node 1 (ingress edge) and the node 3 (egress edge) may be set as base nodes. By using the ingress edge and egress edge as the base node, information about the flow is collected from the start point and end point of the flow, and the accuracy of flow monitoring can be improved.

  In addition, as described above, when determining the base node by designating one or more nodes constituting the packet transfer path for each packet transfer path corresponding to the flow, the base node determination unit 131 performs a certain period of time. A node having a small amount of messages transmitted / received to / from the control device may be designated. At this time, when the amount of messages transmitted / received with the control device per certain time is equal to or less than the threshold value, it may be determined that the amount of messages transmitted / received with the control device is small. Or you may determine the node with few amounts of transmission / reception messages with a control apparatus per fixed time by comparison with another node. By using a node with a small amount of messages sent to and received from the control device per fixed time as a base node, the processing load of control messages at the node can be made uniform in the network even in situations where many flows occur. It is possible to reduce the pressure on the secure channel communication band between the control device 10 and the node.

  After determining the base node, the base node determination unit 131 associates the identification information of the base node with the identification information of the flow identified in step S1, and stores it in the base node information storage unit 133 (step S6). . If there are a plurality of base point nodes, all the base node identification information is associated with the flow identification information.

  Note that, when a node having a flow entry set in step S4 transmits / receives a packet belonging to the flow, the node updates the traffic statistical information in the flow entry.

  Next, a process in which the control device 10 acquires information about a flow (in this example, traffic statistical information) will be described. The information acquisition unit 132 refers to the correspondence between the flow identification information stored in the base node information storage unit 133 and the identification information of the base node, and each information corresponding to the identification information of the flow for which statistical information is to be acquired. Specify the base node.

  Then, the information acquisition unit 132 transmits the flow identification information to the identified base node, and requests traffic statistical information that matches the flow identification information. Upon receiving this request, the base node searches the flow entry determined by the flow identification information from its own flow table, extracts traffic statistical information from the flow entry, and sends it to the control device 10. The information acquisition unit 132 receives this traffic statistical information.

  As a result, the information acquisition unit 132 can collect statistical information of the flow traffic from the base node, and can monitor the flow based on the information.

  By collecting the flow information from the base node determined by various methods described in step S5, it can be said that the collected information is information over the entire flow. That is, as illustrated in FIG. 5, even when there are a plurality of packet transfer paths corresponding to a flow, the information acquisition unit 132 can collect information about each path.

  For example, for each packet transfer path corresponding to a flow, by specifying one or more nodes constituting the packet transfer path, the base node is determined, so that an information collection omission that statistical information on any of the paths is not collected is eliminated. Can be prevented. The same applies to the case where the base node is a node through which all packet transfer paths corresponding to the flow pass, and the case where all the ingress edges are base nodes.

  Since information on all the packet transfer paths corresponding to the flow can be collected from the base node, the information acquisition unit 132 does not need to collect information from all the nodes on the flow. That is, the information acquisition unit 132 may collect information only from the base node. Therefore, the control device 10 can collect information regarding the flow at a low processing cost.

  In the present invention, the base node determination unit 131 stores the correspondence between the flow and the base node in the base node information storage unit 133. Therefore, even if the control device 10 does not hold the flow entry set for each node, the information acquisition unit 132 can identify the base node corresponding to the flow for which statistical information is to be acquired. Flow information can be collected.

Embodiment 2. FIG.
In the second embodiment of the present invention, a group of flows is defined. And the control apparatus of 2nd Embodiment collects the information regarding each flow which belongs to a group from a base node.

  The control device of the second embodiment can be realized with the same configuration as that in FIG. 2, and the second embodiment will be described below with reference to FIG. 2. Also in the second embodiment, a case where the control apparatus 10 collects traffic statistical information as information relating to a flow will be described as an example.

  However, in the second embodiment, the operations of the base node determination unit 131 and the information acquisition unit 132 are different from those in the first embodiment. Hereinafter, the second embodiment will be described with a focus on differences from the first embodiment.

  In the second embodiment, an overlapping part of identification information of each flow belonging to a group is used as group identification information.

  FIG. 6 is a flowchart illustrating an example of processing progress at the time of setting a flow entry in the second embodiment. The operations in steps S1 to S4 are the same as those in steps S1 to S4 in the first embodiment.

  After step S4, the base node determination unit 131 associates the identification information of the flow identified in step S1 with the identification information of each node on the packet transfer path corresponding to the flow, and generates a base node information storage unit 133. (Step S11). At this time, as illustrated in FIG. 5, when there are a plurality of packet transfer paths corresponding to the flow, the base node determination unit 131 determines the flow identification information and the packet for each packet transfer path corresponding to the flow. The base node information storage unit 133 stores the identification information of each node on the transfer path in association with each other.

  Assume that the processes of steps S1 to S11 are executed for a plurality of flows, and a plurality of flows are determined. FIG. 7 shows an example of a plurality of flows. FIG. 7 shows a flow 1 that passes through nodes 1, 2, 3 and exits from node 3, and a flow 2 that passes through nodes 1, 2 and exits from node 2.

  FIG. 8 is an explanatory diagram showing the information stored in step S11 for the flows 1 and 2 shown in FIG. In FIG. 8, a combination of a source MAC address and a destination MAC address of a packet is used as flow identification information. As shown in FIG. 8, regarding the flow 1, nodes 1 to 3 are stored as identification information of each node on the route. For the flow 2, the nodes 1 and 2 are stored as identification information of each node on the route.

  In FIG. 8, in order to simplify the description, a case where there is one packet transfer path corresponding to one flow is taken as an example. As described above, when there are a plurality of packet transfer paths corresponding to a flow, the base node determination unit 131 determines the flow identification information and each node on the packet transfer path for each packet transfer path corresponding to the flow. Are associated with each other.

  In the second embodiment, a group of flows is defined. The group of flows may be determined by the control device 10 according to a user request, or may be set from the outside. A group may be defined based on a network monitoring policy. The determination of the flow group may be performed by an arbitrary method as long as the condition that the group identification information is independent between the groups is satisfied. Hereinafter, a case where one group including flows 1 and 2 shown in FIG. 7 is defined will be described as an example.

  When the group is determined, the base node determination unit 131 determines an overlapping part of the identification information of each flow belonging to the group as the group identification information. In this example, the overlapping part of the identification information of flow 1 and the identification information of flow 2 (see FIG. 8) may be determined as group identification information. As shown in FIG. 8, in the identification information of flow 1 and the identification information of flow 2, the source MAC address is duplicated and the destination MAC address is different. Therefore, in this example, the source MAC address “0x0” that is an overlapping part in the identification information of the flows 1 and 2 may be used as the group identification information.

  In addition, the base node determination unit 131 determines a base node that is a target for collecting information on flows belonging to the group. For example, the base node determination unit 131 selects one or more base nodes from among the nodes on the packet transfer path of each flow belonging to the group, in the part where the packet transfer paths overlap, Is the base node. Even when there are a plurality of packet transfer paths corresponding to one flow, the base node may be determined in the same manner.

  In the example shown in FIGS. 7 and 8, the nodes in the portion where the packet transfer path of flow 1 and the packet transfer path of flow 2 overlap are nodes 1 and 2. Therefore, the base node determination unit 131 may determine either of the nodes 1 and 2 as the base node.

  The base node determination unit 131 associates the identification information of the flow group with the identification information of the base node, and causes the base node information storage unit 133 to store the information. FIG. 9 is a schematic diagram showing the correspondence between the flow group identification information and the base node in this example. In this example, the source MAC address “0x0”, which is group identification information, is associated with the node 1 or the node 2. In the following description, a case where the source MAC address “0x0” is associated with the node 1 is taken as an example.

  Next, a process in which the information acquisition unit 132 acquires information about a flow (in this example, traffic statistical information) will be described. The information acquisition unit 132 identifies a base node corresponding to the identification information of the group for which statistical information is to be acquired, and collects information on the flows belonging to the group (traffic statistical information) from the base node. Here, a case will be described as an example in which flow information is collected for the groups of flows 1 and 2 shown in FIG.

  For example, the information acquisition unit 132 derives group identification information by specifying an overlapping portion of identification information of each flow belonging to a group for which statistical information is to be acquired. This operation is the same as the operation in which the base node determination unit 131 determines the group identification information. In this example, the source MAC address “0x0” is derived as the group identification information. Next, the information acquisition unit 132 refers to the correspondence between the group identification information stored in the base node information storage unit 133 and the base node identification information, and identifies the group of the flow for which statistical information is to be acquired. Each base node corresponding to the information is specified. In this example, the node 1 associated with the source MAC address “0x0” is the base node.

  Then, the information acquisition unit 132 transmits the source MAC address “0x0”, which is group identification information, to the identified base node, and requests traffic statistical information that matches the group identification information.

  Upon receiving this request, the base node searches the flow entry that matches the group identification information from its own flow table. In this example, the base node searches for each flow entry whose transmission source MAC address is “0x0” in the flow identification information. Then, the base node extracts traffic statistical information from each flow entry obtained by the search, and sends the traffic statistical information to the control device 10. The information acquisition unit 132 receives this traffic statistical information.

  As a result, the information acquisition unit 132 can collect the statistical information of the traffic of each flow belonging to the group from the base node, and can monitor the flow based on the information. In this example, it is possible to obtain statistical information of all traffic transmitted from a terminal whose transmission source MAC address is “0x0”.

  As described above, in this embodiment, the information of each flow belonging to the group is acquired from only the base node, not from all the nodes on each flow. Therefore, information on flows belonging to the group can be collected at a low processing cost.

  Note that the base node may transmit statistical information to the control device 10 for each flow entry. In this case, the information acquisition unit 132 can collect statistical information for each flow belonging to the group.

Next, a modification of the second embodiment will be described.
The base node determination unit 131 may perform the operation of the first embodiment when the flow entry is set. That is, the base node determination unit 131 may execute steps S5 and S6 (see FIG. 3) after step S4. In this case, the base node information storage unit 133 stores the correspondence relationship between the flow and the base node, as in the first embodiment. Then, the information acquisition unit 132 may acquire flow information (for example, traffic statistical information) for each flow belonging to the group from which statistical information is to be acquired, as in the first embodiment. As a result, as in the second embodiment, information regarding each flow belonging to the group can be collected from the base node.

  In the above description of the first and second embodiments, the case where the control device 10 collects traffic statistical information has been described as an example. However, the information to be collected is not limited to traffic statistical information. Any information that is common in the route may be used.

  In addition, the control device 10 may store each packet transfer path corresponding to a flow, and the base node determination unit 131 may determine a base node when acquiring flow information.

Embodiment 3. FIG.
The control device according to the third embodiment of the present invention performs flow alive monitoring in addition to the same operation as that of the first embodiment.

  The control device of the third embodiment can be realized with the same configuration as that in FIG. 2, and the third embodiment will be described below with reference to FIG. 2.

  However, in addition to the operation described in the first embodiment, the flow entry setting unit 15 deletes the flow entry when a predetermined condition occurs and notifies the node that the flow entry has been deleted. (Hereinafter referred to as entry deletion notification) is instructed to be transmitted to the control device.

  In addition to the operation described in the first embodiment, the information acquisition unit 132 determines whether or not the flow has disappeared when the entry deletion notification is received.

  Hereinafter, the third embodiment will be described with a focus on differences from the first embodiment.

  In 3rd Embodiment, the operation | movement at the time of flow entry setting is the same as that of 1st Embodiment, and the control apparatus 10 performs step S1-S6 (refer FIG. 3).

  However, the flow entry setting unit 15 performs the following operation after step S5. In other words, the flow entry setting unit 15 determines that a condition that the flow entry is not used for a predetermined time for a node that sets the flow entry, or a predetermined time from the time when the flow entry is set even if the flow entry is used. Is instructed to delete the flow entry when the condition that the time elapses is satisfied. Further, the flow entry setting unit 15 instructs the node to transmit an entry deletion notification to the control device 10 when the flow entry is deleted.

  Note that the flow entry setting unit 15 uses a condition that the flow entry is not used for a predetermined time as a condition for deleting the flow entry, and a condition that a predetermined time has passed since the flow entry is set even if the flow entry is used. Both may be defined, or only one of them may be defined.

  However, it is preferable that the flow entry setting unit 15 sets the length of the predetermined time described in the flow entry deletion condition to be shorter at the base node than at the nodes other than the base node.

  In addition to the flow entry being set, each node that has received the above instruction from the flow entry setting unit 15 is set in step S4 according to the instruction from the flow entry setting unit 15 when a specified condition is satisfied. Delete the flow entry. Then, the node transmits an entry deletion notification to the control device 10. At this time, the node transmits the flow identification information included in the deleted flow entry and the node identification information included in the entry deletion notification.

  When the control device 10 receives the entry deletion notification, the information acquisition unit 132 determines whether the flow has disappeared based on the entry deletion notification. FIG. 10 is a flowchart showing an example of processing progress when an entry deletion notification is received.

  When the control device 10 receives an entry deletion notification from any node (step S31), the information acquisition unit 132 extracts the flow identification information and the transmission source node identification information included in the entry deletion notification. (Step S32). This flow identification information is the flow identification information included in the flow entry deleted by the node.

  Next, the information acquisition unit 132 determines whether or not the entry deletion notification transmission source is a base node corresponding to the identification information of the flow (step S33). The information acquisition unit 132 reads from the base node information storage unit 133 the base node identification information corresponding to the flow identification information obtained in step S32. If the identification information of the entry deletion notification source node matches the identification information of the base node, the information acquisition unit 132 determines that the entry deletion notification transmission source is the base node. On the other hand, if the two do not match, the information acquisition unit 132 determines that the transmission source of the entry deletion notification is not the base node.

  When the transmission source of the entry deletion notification is the base node (Yes in Step S33), the information acquisition unit 132 determines that the flow corresponding to the deleted flow entry has disappeared (Step S34).

  On the other hand, when the transmission source of the entry deletion notification is not the base node (No in step S33), the entry deletion notification is ignored (step S35). In step S35, the control device 10 may newly set a flow entry in order to maintain the flow. Alternatively, the flow entry may be set when the node newly receives a packet.

  In the present embodiment, in addition to the effect of the first embodiment, an effect that the life and death of the flow can be monitored is obtained.

  Further, as described above, it is preferable that the flow entry setting unit 15 sets the length of the predetermined time described in the flow entry deletion condition to be shorter at the base node than at the nodes other than the base node. Then, for example, when the condition that the flow entry has not been used for a predetermined time is determined, the nodes other than the base node have fewer opportunities to transmit the entry deletion notification than the base node. Therefore, the amount of messages received by the control device 10 from a node other than the base node can be reduced.

  Further, the flow entry setting unit 15 does not have to instruct transmission of an entry deletion notification to nodes other than the base node.

Embodiment 4 FIG.
The control device according to the fourth embodiment of the present invention performs life / death monitoring of a group of flows in addition to the same operation as that of the second embodiment.

  The control device of the fourth embodiment can be realized with the same configuration as that in FIG. 2, and the fourth embodiment will be described below with reference to FIG. 2.

  However, in addition to the operation in the second embodiment, the flow entry setting unit 15 deletes a flow entry when a predetermined condition occurs for a node, and transmits an entry deletion notification to the control device. Instruct.

  In addition to the operation in the second embodiment, the information acquisition unit 132 determines whether or not the flow group has disappeared when the entry deletion notification is received.

  Hereinafter, the fourth embodiment will be described with a focus on differences from the second embodiment.

  In the fourth embodiment, the operation at the time of setting a flow entry is the same as in the second embodiment, and the control device 10 executes steps S1 to S11 (see FIG. 6).

  However, the flow entry setting unit 15 performs the following operation after step S4, for example. In other words, the flow entry setting unit 15 determines that a condition that the flow entry is not used for a predetermined time for a node that sets the flow entry, or a predetermined time from the time when the flow entry is set even if the flow entry is used. Is instructed to delete the flow entry when the condition that the time elapses is satisfied. Further, the flow entry setting unit 15 instructs the node to transmit an entry deletion notification to the control device 10 when the flow entry is deleted.

  Note that the flow entry setting unit 15 uses a condition that the flow entry is not used for a predetermined time as a condition for deleting the flow entry, and a condition that a predetermined time has passed since the flow entry is set even if the flow entry is used. Both may be defined, or only one of them may be defined.

  In addition to the flow entry being set, each node that has received the above instruction from the flow entry setting unit 15 is set in step S4 according to the instruction from the flow entry setting unit 15 when a specified condition is satisfied. Delete the flow entry. Then, the node transmits an entry deletion notification to the control device 10. At this time, the node transmits the flow identification information included in the deleted flow entry and the node identification information included in the entry deletion notification.

  When the control device 10 receives the entry deletion notification, the information acquisition unit 132 determines whether the flow group has disappeared based on the entry deletion notification. FIG. 11 is a flowchart illustrating an example of processing progress when an entry deletion notification is received in the fourth embodiment.

  When the control device 10 receives an entry deletion notification from any node (step S41), the information acquisition unit 132 specifies the identification information of the group to which the flow indicated by the flow identification information included in the entry deletion notification belongs. (Step S42). The information acquisition unit 132 extracts the flow identification information included in the entry deletion notification, and from the group identification information stored in the base node information storage unit 133, the group that matches the flow identification information Identify the identification information. Since the group identification information is described as part of the flow identification information (see FIGS. 8 and 9), the information acquisition unit 132 sets the group identification information included in the extracted flow identification information. What is necessary is just to specify. In addition, the information acquisition unit 132 also extracts the identification information of the node that transmitted the entry deletion notification.

  Next, the information acquisition unit 132 determines whether or not the transmission source of the entry deletion notification is a base node corresponding to the group identification information specified in step S42 (step S43). The information acquisition unit 132 reads the base node identification information corresponding to the group identification information obtained in step S42 from the base node information storage unit 133. If the identification information of the entry deletion notification source node matches the identification information of the base node, the information acquisition unit 132 determines that the entry deletion notification transmission source is the base node. On the other hand, if the two do not match, the information acquisition unit 132 determines that the transmission source of the entry deletion notification is not the base node.

  When the transmission source of the entry deletion notification is not the base node (No in step S43), the entry deletion notification is ignored (step S44).

  In addition, when the transmission source of the entry deletion notification is a base node (Yes in step S43), the information acquisition unit 132 identifies the group identification information for the base node corresponding to the group identification information specified in step S42. Information on flow entries that conform to the above is requested (step S45). The information requested in step S45 may be information that can be specified for each flow entry. For example, the information acquisition unit 132 may request statistical information in the flow entry that matches the group identification information.

  In step S <b> 45, the base node that has received the request sends the flow entry information that matches the identification information of the designated group to the control device 10. However, if the corresponding information does not exist, the information is not sent to the control device 10.

  If the information acquisition unit 132 acquires information as a response to the request in step S45 (Yes in step S46), since a part of the flow belonging to the group specified in step S42 is alive, the group is also alive. (Step S47).

  On the other hand, if no information is acquired as a response to the request in step S45, it is determined that all of the flows belonging to the group specified in step S42 have disappeared, so that the group has also disappeared (step S48).

  In the present embodiment, in addition to the effect of the second embodiment, the effect that the alive monitoring of the group of flows can be performed is obtained.

  In addition, the flow entry setting unit 15 performs an operation for instructing the node that has set the flow entry to delete the flow entry when the above-described condition is satisfied and to transmit an entry deletion notification to the control device 10. It may be performed after the group base node is determined. In this case, as in the third embodiment, it is preferable that the length of the predetermined time described in the flow entry deletion condition is shorter at the base node than at the nodes other than the base node.

Embodiment 5. FIG.
FIG. 12 is an explanatory diagram showing an example of a communication system to which the fifth exemplary embodiment of the present invention is applied. A control device (flow information collection system) 30 according to the fifth embodiment is connected to an external device 40. The external device 40 specifies identification information for specifying a flow to the control device 30 and specifies an interface of a node that outputs a packet to the outside of the communication system in the flow. However, the identification information for designating the flow may be information that can generate a plurality of flows. Therefore, a set of setting contents designated by the external device 40 is referred to as a flow group setting. In addition, the flow identification information specified by the external device 40 is referred to as flow group identification information here.

  A node that outputs a packet to the outside of the communication system in the flow is referred to as a boundary node.

  FIG. 13 is a block diagram illustrating a configuration example of a control device according to the fifth embodiment. The control device 30 of the fifth embodiment includes a setting management unit 19 and an external device communication unit 20 in addition to the elements illustrated in FIG.

  The external device communication unit 20 is a communication interface with the external device 40 (see FIG. 12).

  The setting management unit 19 holds the flow group setting designated for the external device 40.

  In this embodiment, the topology management unit 11 collects and manages network topology information and notifies the external device 40 of the topology information. The external device 40 uses this topology information to determine the boundary node and its interface.

  In this embodiment, when the flow group identification information and the interface of the boundary node are designated by the external device 40, the route calculation unit 12 sends a packet from the boundary node to the outside of the communication system (see FIG. 12). Calculate the flow path to output.

  The flow entry calculation unit 14 creates a flow entry. The flow entry calculation unit 14 creates a flow entry corresponding to each node by associating an action for transferring a packet along the route calculated by the route calculation unit 12 with the flow group identification information.

  Here, the flow entry calculation unit 14 may use the flow group identification information designated by the external device 40 as it is and associate it with the action. Alternatively, the flow entry calculation unit 14 may detail the flow group identification information designated by the external device 40 and associate it with the action.

  A plurality of flows may be generated from the flow group identification information.

  The flow entry setting unit 15 requests the node to set the flow entry calculated by the flow entry calculation unit 14. This operation is the same as the operation in the other embodiments.

  The control message processing unit 16 and the node communication unit 17 are the same as the control message processing unit 16 and the node communication unit 17 in other embodiments.

  The base node determination unit 131 determines the base node, and stores the base node identification information and the flow group identification information specified by the external device 40 in the base node information storage unit 133.

  The base node determination unit 131 determines the base node by designating one or more nodes constituting the packet transfer path for each packet transfer path calculated by the path calculation unit 12 based on the flow group setting, for example. May be. Further, the base node may be determined by selecting a node from among the nodes through which all packet transfer paths calculated by the path calculation unit 12 pass based on the flow group setting.

  When the flow group identification information related to the flow for which information (for example, traffic identification information) is to be collected is designated from the external device 40, the information acquisition unit 132 refers to the base node information storage unit 133. The identification information of the base node corresponding to the flow group identification information is specified. And the information acquisition part 132 transmits flow group identification information with respect to the base node, and requests | requires the information (traffic identification information) regarding a flow.

  Upon receiving this request, the base node searches for a flow entry that matches the flow group identification information sent from the information acquisition unit 132, and returns the traffic identification information in the flow entry to the control device 30. And the information acquisition part 132 acquires the identification information of the traffic.

  The setting management unit 19 sends the information collected as described above from the base node by the information acquisition unit 132 according to the flow group identification information designated from the external device 40 to the external device 40.

  In this embodiment, the base node determination unit 131 specifies, for example, one or more nodes constituting the packet transfer path for each packet transfer path calculated by the path calculation unit 12 based on the flow group setting. Determine the base node. Alternatively, the base node determination unit 131 determines a base node by selecting a node from among the nodes through which all packet transfer routes calculated by the route calculation unit 12 pass based on the flow group setting. Furthermore, in the flow entry, the flow group identification information itself or information that details the flow group identification information is set. Therefore, by collecting the information in the flow entry that conforms to the flow group identification information from the base node, it is possible to collect the traffic statistical information in the paths of all flows specified by the specified flow group identification information. . This statistical information can be regarded as statistical information over all the flows.

  Then, the setting management unit 19 transmits the traffic statistical information acquired by the base node to the external device 40 by the information acquisition unit 132. As a result, the external device 40 can acquire and monitor the communication execution result for the flow determined as a result of the flow group setting set by the external device 40.

  Also in the fifth embodiment, the case of collecting traffic statistical information has been described as an example. However, the information to be collected is not limited to traffic statistical information, and may be information that is common in the route. Good.

  Further, the control device 30 may store each packet transfer path for each flow group identification information, and the base node determination unit 131 may determine the base node when acquiring flow information.

  Also in the fifth embodiment, information may be collected in the same manner as each of the first to fourth embodiments, or the life and death monitoring of the flow may be performed.

  Next, the minimum configuration of the present invention will be described. FIG. 14 is a block diagram showing an example of the minimum configuration of the present invention. The flow information collection system of the present invention includes a base node information storage unit 71 and an information acquisition unit 72.

  The base node information storage unit 71 (for example, the base node information storage unit 133) is configured to identify a plurality of flows that serve as flow identification information (flow identification information itself) or a part thereof (for example, flow group identification information). The correspondence relationship between the common part of the information) and the base node that is the node from which the flow information is collected is stored.

  The information acquisition unit 72 (for example, the information acquisition unit 132) acquires flow information from the base node corresponding to the flow identification information or a part thereof.

  With such a configuration, the flow information is acquired not from all nodes but from the base node. Therefore, information about the flow can be collected at a low processing cost.

  A part or all of the above embodiments can be described as in the following supplementary notes, but is not limited to the following.

(Additional remark 1) It is determined as a node which can collect the information in each packet transfer path | route corresponding to the said flow, and the node which is the collection object of the identification information of the flow of an open flow, or its part and the said flow When the base node information storage means for storing the correspondence with the base node which is one or a plurality of nodes and the flow identification information or part thereof are designated, the flow identification information or part thereof is A flow information collection system comprising: an information acquisition unit that specifies a corresponding base node and acquires flow identification information that matches the flow identification information or a part thereof from the base node.

(Supplementary Note 2) Base point node determining means for determining a base point node corresponding to the flow based on the flow identification information and storing the correspondence between the flow identification information and the base point node in the base point node information storage means. The flow information collection system according to supplementary note 1 provided.

(Supplementary note 3) The flow information collecting system according to supplementary note 2, wherein the base node determination unit determines the base node by designating one or more nodes on the packet transfer path for each packet transfer path corresponding to the flow.

(Supplementary Note 4) The base node determination means determines the base node by designating one or more nodes from among the nodes through which all packet transfer paths corresponding to the flow pass. system.

(Supplementary note 5) The flow information collecting system according to supplementary note 2, wherein the base node determination means determines, as a base node, a node that first transfers a packet in a packet transfer path corresponding to the flow.

(Supplementary note 6) The base node determination means determines the node that will transfer the packet first and the node that outputs the packet last in the packet transfer path corresponding to the flow as the base node. system.

(Supplementary note 7) The flow point information collection system according to supplementary note 2, wherein the base point node determination unit determines the base point node based on a communication amount with the flow information collection system.

(Supplementary Note 8) A flow entry that sets a flow entry indicating flow identification information and an operation when a packet matching the flow identification information is received for each node on the packet transfer path corresponding to the flow The flow entry setting means instructs each node to notify that the flow entry is deleted when the flow entry is deleted, and the information acquisition means The flow information collection system according to any one of Supplementary Note 1 to Supplementary Note 7, which determines that the flow is deleted when it is a base node.

(Supplementary Note 9) The flow entry setting means designates passage of a predetermined time as a flow entry deletion condition for each node on the packet transfer path corresponding to the flow, and sets the predetermined time for the base node as a base point The flow information collection system according to attachment 8, wherein the flow information is set shorter than the predetermined time for nodes other than the node.

(Supplementary Note 10) When a flow group is designated, the information acquisition unit specifies a base node corresponding to the flow identification information for each flow belonging to the group, and converts the base node to the flow identification information. The flow information collection system according to any one of Supplementary Note 1 to Supplementary Note 9, which acquires information on a suitable flow.

(Supplementary Note 11) For each flow identification information, the correspondence between the flow identification information and the node on each packet transfer path corresponding to the flow is stored in the base node information storage means, and the flow group is designated. In this case, an overlapping portion of identification information of each flow belonging to the group is used as the identification information of the group, a base node is determined from among the overlapping nodes in each packet transfer path corresponding to each flow, and the group The flow information collection system according to appendix 1, further comprising base point node determination means for storing a correspondence relationship between the identification information of the base point and the base point node in base point node information storage means.

(Supplementary Note 12) A flow entry that sets a flow entry indicating flow identification information and an operation when a packet matching the flow identification information is received for each node on the packet transfer path corresponding to the flow The flow entry setting means instructs each node to notify that the flow entry is deleted when the flow entry is deleted, and the information acquisition means If it is a base node, identify identification information of the group that matches the deleted flow entry, request information that matches the identification information of the group from the base node corresponding to the identification information of the group, The flow information collection system according to supplementary note 11, wherein the group is determined to be deleted when the information is not received.

(Supplementary note 13) The flow information collecting system according to any one of Supplementary note 1 to Supplementary note 12, wherein the information acquisition means acquires traffic statistical information as flow information.

  This application claims the priority on the basis of the JP Patent application 2012-196281 for which it applied on September 6, 2012, and takes in those the indications of all here.

  Although the present invention has been described with reference to the embodiments, the present invention is not limited to the above-described embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

  The present invention is preferably applied to information collection regarding an open flow.

10, 30 Control device (flow information collection system)
DESCRIPTION OF SYMBOLS 11 Topology management part 12 Path | route calculation part 13 Flow information acquisition part 14 Flow entry calculation part 15 Flow entry setting part 131 Base node determination part 132 Information acquisition part 133 Base point node information storage part

Claims (10)

Base node information storage means for storing the correspondence between the flow identification information or a part thereof and a base node that is a node from which the flow information is collected;
A flow information collection system comprising: information acquisition means for acquiring the flow information from a base node corresponding to the flow identification information or a part thereof. A base node determination unit that determines a base node corresponding to the flow based on the flow identification information and stores a correspondence relationship between the flow identification information and the base node in a base node information storage unit. Flow information collection system described in 1. For each node on the packet transfer path corresponding to the flow, there is provided flow entry setting means for setting a flow entry indicating flow identification information and an operation when a packet matching the flow identification information is received. ,
The flow entry setting means instructs each node to notify that the flow entry is deleted when the flow entry is deleted,
The flow information collection system according to claim 1, wherein the information acquisition unit determines that the flow has been deleted when a transmission source of the notification is a base node. Flow entry setting means
The flow information collection system according to claim 3, wherein elapse of a predetermined time is specified as a flow entry deletion condition, and the predetermined time for a base node is set shorter than the predetermined time for a node other than the base node. Information acquisition means
The base point node corresponding to the flow identification information is specified for each flow belonging to the flow group, and the flow information that matches the flow identification information is acquired from the base point node. The flow information collection system according to any one of the above. For each flow identification information, the correspondence between the flow identification information and the node on each packet transfer path corresponding to the flow is stored in the base node information storage means, and the identification information of each flow belonging to the flow group is stored. The overlapping part is used as the group identification information, a base node is determined from among the overlapping nodes in each packet transfer path corresponding to each flow, and the correspondence between the group identification information and the base node is used as a base point. The flow information collection system according to claim 1, further comprising a base node determination unit that is stored in the node information storage unit. For each node on the packet transfer path corresponding to the flow, there is provided flow entry setting means for setting a flow entry indicating flow identification information and an operation when a packet matching the flow identification information is received. ,
The flow entry setting means instructs each node to notify that the flow entry is deleted when the flow entry is deleted,
The information acquisition means, when the transmission source of the notification is a base node, specifies a group that matches the deleted flow entry, requests information on the group from the base node corresponding to the group, The flow information collection system according to claim 6, wherein if the information is not received, it is determined that the group is deleted. The flow information collection system according to any one of claims 1 to 7, wherein the information acquisition means acquires traffic statistical information as flow information. Storing the correspondence between the identification information of the flow or a part thereof and the base node that is the node from which the flow information is collected;
The flow information collection method characterized by acquiring the flow information from a base node corresponding to the flow identification information or a part thereof. A flow information collection program mounted on a computer comprising base node information storage means for storing a correspondence relationship between flow identification information or a part thereof and a base node that is a node from which the flow information is collected. ,
In the computer,
A flow information collection program for executing an information acquisition process for acquiring the flow information from a base node corresponding to the flow identification information or a part thereof.

Images