JP5572812B2 - Gateway device and communication block allocation changing method - Google Patents

Description

  The present invention relates to a gateway device that connects a plurality of networks, and more particularly to a gateway device that increases the number of logical nodes whose number of connections is increasing.

  A gateway device exists in a communication system between a network terminal and a user terminal connected to a wireless communication network, and can convert data with different communication media and protocols to communicate with a connection destination desired by the user. It is a device for making.

  One type of gateway device is an integrated gateway device. The integrated gateway device has a function capable of connecting to a plurality of communication networks. For example, the integrated gateway device can be connected to both the LTE network and the EVDO network, which are wireless communication networks.

  The LTE network and EVDO network will be described.

  LTE and EVDO are names of wireless communication standards. The EVDO network is a wireless communication standard that is currently used, and the LTE network is a wireless communication standard that is scheduled to start service in the future. In the LTE network, a communication speed of 100 Mbit / s or more can be realized, and higher-speed communication is possible than in the EVDO network. In the future, it is assumed that users will migrate from the currently used EVDO network to the LTE network.

  The integrated gateway device includes a P-GW block that is a functional unit for connecting to the LTE network, and an HA block that is a functional unit for connecting to the EVDO network. These are collectively referred to as a communication block.

  When the number of users of the LTE network has increased, the integrated gateway device needs to newly prepare a P-GW block and increase the number that can be accommodated in the LTE network.

  For this reason, it is necessary to determine when to newly prepare a P-GW block based on the rate of increase in the number of users in the LTE network. It is also important to improve the quality of service by introducing a new device without interrupting the service when updating the device.

  As described above, the integrated gateway device controls communication of a plurality of communication systems. In this case, it becomes a problem to have an appropriate configuration for the number of CPUs used by each communication block.

  If the ratio of the CPU used by a certain communication block and the CPU used by another communication block is far from the ratio of calls connected to the integrated gateway device, there is a margin in the capacity that can be accommodated in the performance of the entire integrated gateway device. Even in such a case, the number of CPUs used by either the LTE block or the HA block may actually be insufficient, and a call that cannot be connected to the integrated gateway device may occur.

  For this reason, it is important to appropriately set the ratio of CPUs used in the communication system of the integrated gateway device based on the ratio of calls that are connected.

  In order to adjust the ratio of CPUs used by the communication blocks in the integrated gateway apparatus based on the ratio of calls connecting to the integrated gateway apparatus, accommodation switching is performed between the CPUs used by the respective communication blocks.

  The accommodation change is to transfer a user accommodated in a certain CPU so as to be accommodated in another CPU. The function of the CPU is converted from one communication block to another communication block by accommodation change. As a result, the ratio of the CPU used by the communication block is adjusted from the ratio of calls connected to the integrated gateway device, and the number of accommodated integrated GW devices can be used without surplus.

  A plurality of logical nodes exist inside a node device such as an integrated gateway device. In the existing system, the configuration of the logical node for each CPU provided in the apparatus has not been dynamically and autonomously changed.

  For this reason, when a certain logical node is used up to the upper limit of the connectable number, if there is still room for the connectable number in the integrated gateway device as a whole, the user cannot access the logical node that is not used up to the upper limit of the connectable number. Users can connect, but users cannot connect to logical nodes that are used up to the maximum number of connections. Therefore, the integrated gateway device cannot be operated efficiently in a situation where connections are concentrated on some logical nodes.

  Here, the transition of the number of users when a service of a new wireless communication system is started will be considered. At the time when a new wireless communication system service is started, it is considered that users of the old communication system that have been operating so far accounted for a large part of the total number of users, and the number of users of the new communication system is not so large.

  For this reason, it is considered that it is not necessary to obtain a very powerful performance for the communication block for the new communication system when the service of the new communication system is started.

  As time passes from the start of service of the new wireless communication system, it is expected that users who have used the old communication system will start switching to the new communication system. Therefore, since the number of users of the new communication system increases, the equipment for the new radio communication system is increased from the state at the start of service, or the old radio communication system is replaced with a new radio communication system, and a new radio There is a need to increase the number of users of communication systems that can be accommodated.

  However, in this case, it is very expensive to add a device. Therefore, it is not just a matter of adding a device, but by checking the number of connected users of each wireless communication system and when to change which device equipment. It is necessary to make a plan and add equipment. It is necessary to have an operator who investigates the number of connected users, decides when to change the equipment of which apparatus, and changes the actual equipment.

  Devices for changing accommodation have been developed so far. For example, when a device that accommodates user information is abolished, a method is known in which accommodation change is performed by moving the accommodated user information to another device. In this method, it is necessary to determine in advance which device is to be replaced. There is also known a method of changing accommodation by moving user information accommodated in a certain apparatus to an arbitrary apparatus by human operation.

  In addition, when an exchange is abolished and the exchange is actually abolished, or when an exchange is exchanged and replaced with a new exchange, a large number of subscribers who are semi-fixedly accommodated in the exchange A method is known in which distributed accommodation is simultaneously switched to other exchanges by simultaneous path switching within a time division switch (see Patent Document 1). The processing load of a plurality of exchanges can be equalized by distributed accommodation switching.

  When changing accommodation, by changing the subscriber accommodation between multiple nodes according to a centralized maintenance instruction for only the accommodation destination node, the accommodation change can be performed for the user accommodated in the accommodation change source. A method is known (see Patent Document 2). Since accommodation was performed at the accommodation source node after investigating whether accommodation can be changed to the accommodation destination node in advance, it was necessary to operate both the accommodation source node and the accommodation destination node. However, according to the method described in Patent Document 2, accommodation can be changed only by operating the accommodation destination node.

JP-A-9-36956 JP-A-10-117232

  When a new wireless communication system service is started by an integrated gateway device having both functions of the old and new wireless communication systems, it is necessary to change the CPU function by carrying out accommodation change. There is a fourth problem.

  The first problem is that it is necessary to manually check the number of users connected to each communication system and the usage status of the memory in order to determine whether or not to change the accommodation. is there.

  The second problem is that, when it is determined that the accommodation is to be changed, the maintenance person must make an expropriation plan as to which CPU is to be changed from which CPU, and the burden on the maintenance person is large.

  The third problem is that, when changing accommodation, the operation of changing accommodation must be performed manually, which places a heavy burden on the maintenance personnel.

  The fourth problem is that since the service of the wireless communication system has already been started, it is necessary to change accommodation without affecting the user receiving the service.

  In both methods described in Patent Document 1 and Patent Document 2, it is necessary for the maintenance person to make a plan for accommodation change, and for the maintenance person to actually perform the accommodation change operation. There are challenges.

  The present invention reduces the workload of maintenance personnel by causing the integrated gateway device to automatically and autonomously execute the accommodation change determination and the accommodation change operation so as not to affect the user receiving the service. For the purpose.

In a typical example of the present invention, a first communication network to which a communication terminal is connected and a second communication network to which the communication terminal is connected, which are different from the first communication network. In the gateway device, the gateway device includes at least one processor and a memory connected to the processor, and a resource of the processor is a first communication block for managing the first communication network, and the second communication network. Assigned to a control block that manages the assignment of the processor resources to the first communication block and the second communication block, and the control block is managed by the first communication block. The first number of connections of the communication terminal to the first communication network up to now A connection number counting unit that counts a second connection number that is the number of connections of the communication terminal to the second communication network managed by the second communication block up to the present time, and the first communication block; An allocation change unit that changes the allocation of the processor resources to the second communication block, and the allocation change unit subtracts the first connection number counted last time from the first connection number counted this time A first subtraction result is calculated, a second subtraction result is calculated by subtracting the second connection number counted last time from the second connection number counted this time, and the calculated first subtraction result or second subtraction is calculated. comparing the result with a first predetermined value, the resources of the processors assigned to the communication block corresponding to the first predetermined value or less and became subtraction result corresponding to the subtraction result becomes equal to or less than the first predetermined value Communication block And allocating the other communication blocks.

  According to the present invention, the integrated gateway device can automatically and autonomously execute the determination of accommodation change and the operation of accommodation change so as not to affect the user who receives the service. The purpose is to reduce.

It is a block diagram of the integrated gateway apparatus of embodiment of this invention. It is explanatory drawing of ARP request of embodiment of this invention. It is explanatory drawing of unnecessary ARP of embodiment of this invention. It is explanatory drawing of the method of grasping | ascertaining the number of connections of a HA block and a P-GW block (henceforth a communication block generically) using each block of the integrated gateway apparatus of the 1st Embodiment of this invention using unnecessary ARP. It is a sequence diagram of the function conversion process of the 1st Embodiment of this invention. It is explanatory drawing of the session number message of the 1st Embodiment of this invention. It is explanatory drawing of the function conversion message of the 1st Embodiment of this invention. It is a flowchart of the function conversion determination process performed by CPU allocated to the control block of the 1st Embodiment of this invention. It is explanatory drawing of CPU which changed to the pre-blocking state of the 1st Embodiment of this invention. It is explanatory drawing of CPU newly allocated to the P-GW block of the 1st Embodiment of this invention. It is a sequence diagram of the power-off process of the 1st Embodiment of this invention. It is a flowchart of the power-off determination process performed by CPU allocated to the control block of the 1st Embodiment of this invention. It is explanatory drawing of the state by which the power-off process was performed with CPU allocated to the HA block of the 1st Embodiment of this invention. It is a flowchart of the increase cause function conversion determination process performed by CPU allocated to the control block of the 1st Embodiment of this invention. It is a sequence diagram of an increase cause function conversion process when a CPU assigned to a communication block different from a communication block assigned to a CPU conforming to the increase condition of the first embodiment of the present invention performs function conversion. It is a sequence diagram of increase cause function conversion processing when the CPU whose power supply is stopped according to the first embodiment of the present invention performs function conversion. It is explanatory drawing when a function conversion process is performed by CPU111 which is allocated to the HA block of the 1st Embodiment of this invention, and the power supply has stopped. It is a flowchart of the increase cause function conversion determination process performed by CPU allocated to the control block of the modification of the 1st Embodiment of this invention. It is explanatory drawing of the multi-core CPU with which the integrated gateway apparatus of the 2nd form of this invention is equipped.

(First embodiment)
Hereinafter, embodiments of the present invention will be described with reference to FIGS.

  FIG. 1 is a block diagram of an integrated gateway device according to an embodiment of the present invention.

  The integrated gateway device is a device that is arranged between a network network and a base station that accommodates a wireless communication terminal (such as a mobile phone) and transmits and receives data.

  When the wireless communication terminal transmits data to a predetermined computer in the network, the base station receives the data transmitted by the wireless communication terminal and transmits the received data to the integrated gateway device. The integrated gateway device performs data processing on the received data and transmits the data to a computer with a network.

  When a predetermined computer of the network transmits data to the wireless communication terminal, the integrated gateway device transmits data received from the predetermined computer to the base station. Then, the base station transmits the received data to the wireless communication terminal.

  The integrated gateway device holds user information related to the user of the wireless communication terminal, refers to the user information, and transmits data to the base station or the network.

  The integrated gateway device can be connected to networks having different communication methods, for example, can be connected to an LTE network and an EVDO network. In this case, the integrated gateway device can accommodate and communicate users of the LTE network and the EVDO network.

  In the integrated gateway device, a plurality of logical nodes are generated. In this embodiment, a control block, a P-GW block, an HA block, a redundant block, and the like are generated.

  The P-GW block controls communication of a wireless communication terminal using the LTE network, and the HA block controls communication of a wireless communication terminal using the EVDO network. The control block controls each block, and the redundant block is a block used as a substitute block when any block does not operate.

  In this embodiment, the integrated gateway device includes five CPUs 110 to 114 and five memories 130 to 134, and a logical node is generated by being assigned to each block in units of CPU and memory.

  The CPU 110 and the memory 130 are assigned to the control block. The CPUs 111 and 112 and the memories 131 and 132 are assigned to the HA block. The CPU 113 and the memory 133 are assigned to the P-GW block. The CPU 114 and the memory 134 are assigned to redundant blocks.

  Data relating to the wireless communication terminal connected to the HA block is stored in the memories 131 and 132 allocated to the HA block, and the wireless communication terminal connected to the P-GW block is stored in the memory 133 allocated to the P-GW block. Is stored.

  When the number of CPUs allocated to the P-GW block is large, it indicates that the processing capability of the P-GW block is large and the number of wireless communication terminals that can be accommodated in the LTE network is large. On the other hand, when the number of CPUs allocated to the HA block is large, it indicates that the processing capacity of the HA block is large and the number of wireless communication terminals that can be accommodated in the EVDO network is large.

  It is assumed that the number of wireless communication terminals using the LTE network increases and the number of wireless communication terminals using the EVDO network decreases.

  It is necessary to determine the timing for allocating the CPU to the P-GW block from the increase rate of the number of wireless communication terminals using the LTE network. Note that the integrated gateway device is not allowed to interrupt the service even when the allocation of CPUs to blocks is changed.

  In the present embodiment, the control block can grasp the number of wireless communication terminals connected to the P-GW block and the HA block by unnecessary ARP (Gratuitous ARP).

  The unnecessary ARP will be described with reference to FIGS.

  First, the HA block transmits an unnecessary ARP (Address Resolution Protocol) when a wireless communication terminal is connected to the HA block.

  First, before describing unnecessary ARP, ARP will be described.

  ARP is a protocol for obtaining a MAC address from an IP address in an Ethernet (registered trademark) environment.

  FIG. 2 is an explanatory diagram of an ARP request according to the embodiment of this invention.

  In FIG. 2, a case where the wireless communication terminal obtains the MAC address of the HA block 1 will be described.

  In order to obtain the MAC address of the HA block 1, the wireless communication terminal transmits an ARP request by broadcast. The ARP request includes the IP address of the transmission source (wireless communication terminal), the MAC address of the transmission source (wireless communication terminal), and the IP address of the transmission destination (HA block 1).

  Whether all terminals (HA blocks 1 to 3) that have received the ARP Request transmitted by broadcast match the IP address of the transmission destination (HA block 1) included in the received ARP Request with their own IP address Determine whether.

  If both IP addresses match, the terminal (HA block 1) transmits an ARP reply including its own MAC address to the wireless communication terminal that is the transmission source.

  Note that the ARP reply includes the MAC address of the destination (HA block 1) in addition to the information included in the ARP request.

  As described above, in order for the terminal A to obtain the MAC address of the terminal B, the ARP request including the IP address of the terminal B may be transmitted by broadcast. As a result, the terminal B transmits an ARP reply including its own MAC address to the terminal A, so that the terminal A can obtain the MAC address of the terminal B.

  FIG. 3 is an explanatory diagram of the unnecessary ARP according to the embodiment of this invention.

  Unnecessary ARP is a kind of ARP packet. The unnecessary ARP has its own IP address registered in the destination IP address included in the ARP Request.

  Usually, since the IP address of the destination of the unnecessary ARP does not match the IP address of the terminal that received the unnecessary ARP, unless the IP address overlaps with the terminal of the source, the terminal that received the unnecessary ARP Do not send ARP reply. For this reason, the unnecessary ARP can be used to detect duplication of IP addresses.

  Here, as shown in FIG. 3, the HA block 1 broadcasts an unnecessary ARP when a wireless communication terminal is connected. Accordingly, the HA block can notify the other HA blocks 2 and 3 that the wireless communication terminal is connected to itself.

  FIG. 4 is a diagram illustrating a method of grasping the number of connections of HA blocks and P-GW blocks (hereinafter collectively referred to as communication blocks) using unnecessary ARP for each block of the integrated gateway device according to the first embodiment of this invention. FIG.

  In FIG. 4, when the wireless communication terminal is connected to the CPU 111 assigned to the HA block, the CPU 111 transmits unnecessary ARP by broadcast. The unnecessary ARP transmitted by broadcast is received by the CPUs 110, 112 to 114 and the external device.

  In the unnecessary ARP, as described in FIG. 3, the transmission destination IP address and the transmission source IP address are set to be the same. Therefore, the CPUs 110 and 112 to 114 that receive the unnecessary ARP can determine that the unnecessary ARP has been received if the transmission destination IP address and the transmission source IP address included in the received packet match.

  When receiving unnecessary ARPs, the CPUs 110 to 114 count the number of unnecessary ARPs received for each source IP address. Thereby, it is possible to grasp the number of wireless communication terminals connected to the CPUs 111 and 112 assigned to the HA block and the CPU 113 assigned to the P-GW block.

  When the number of wireless communication terminals connected to the CPU assigned to the communication block meets a predetermined reduction condition, a process (function conversion process) for assigning the CPU to the other communication block will be described with reference to FIGS. explain.

  FIG. 5 is a sequence diagram of function conversion processing according to the first embodiment of this invention.

  First, the CPU 112 (HA block 2) assigned to the HA block detects that a wireless communication terminal is connected to itself (501), and transmits an unnecessary ARP by broadcast (502).

  Each of the CPUs 110 to 113 periodically executes a process of determining whether or not the number of wireless communication terminals connected to the CPUs 111 to 113 meets a predetermined decrease condition based on the unnecessary ARP.

  Specifically, each of the CPUs 110 to 113 specifies the total number of wireless communication terminals connected to each CPU based on the received source IP address of the unnecessary ARP. Then, each of the CPUs 110 to 113 specifies the total number of connections (the previous total number of connections) of each of the CPUs 110 to 113 specified when the previous determination process is executed. Then, the CPUs 110 to 113 subtract the previous total number of connections from the total number of connections for each CPU, and determine whether or not the subtracted value is equal to or less than a predetermined lower limit value.

  The CPU 110 can execute the determination process for all the CPUs 111 to 113 assigned to the communication block by the above-described process. However, the CPUs 111 to 113 assigned to the communication block have their own wireless communication terminals based on the received unnecessary ARP. It cannot be determined whether or not the number of connections meets a predetermined decrease condition.

  For this reason, the CPUs 111 to 113 specify the total number of wireless communication terminals connected to themselves based on the transmitted unnecessary ARP, and the total number of connections specified when the previous determination process is executed (the previous total) Number of connections). Then, the CPUs 111 to 113 subtract the previous total number of connections from the total number of connections, and determine whether the subtracted value is equal to or less than a predetermined lower limit value.

  Here, the CPUs 110 to 113 determine that the number of connected CPUs 112 meets a predetermined decrease condition (503).

  Next, when it is determined in step 503 that the number of connections of the CPU 112 meets a predetermined reduction condition, the CPUs 111 to 113 assigned to the communication block determine the number of sessions established between itself and the wireless communication terminal. The number-of-sessions message to be included is notified to the CPU 110 assigned to the control block (504). Details of the session number message will be described with reference to FIG. The number of sessions indicates the number of wireless communication terminals actually connected to each of the CPUs 111 to 113.

  The CPU 110 receives the session number message from the CPUs 111 to 113, refers to the received session number message, and determines that the number of sessions of the CPU 112 determined to satisfy the predetermined decrease condition is equal to or less than the predetermined value. It is determined to perform function conversion for assigning the CPU 112 to the other communication block (P-GW block) (505).

  Then, the CPU 110 notifies the maintenance person of the integrated gateway device that the function conversion of the CPU 112 is executed (506), and transmits a function conversion message to the CPU 112 (507). The function conversion message will be described in detail with reference to FIG.

  When the CPU 112 receives the function conversion message, the CPU 112 puts itself into a pre-blocking state in which a new connection of the wireless communication terminal is not accepted, and when the number of sessions between the CPU 112 and the wireless communication terminal becomes 0, the received function conversion The block is assigned to the block registered in the post-conversion system name 703 included in the message (508).

  FIG. 6 is an explanatory diagram of a session number message according to the first embodiment of this invention.

  The session number message includes a message type 601, a message length 602, and a connection session number 603.

  In the message type 601, an identifier indicating the type of message is registered. Here, an identifier indicating a session number message is registered. In the message length 602, the bit length of the session number message is registered. In the connection session number 603, the number of sessions established between the CPUs 111 to 113 assigned to the communication block and the wireless communication terminal is registered.

  FIG. 7 is an explanatory diagram of a function conversion message according to the first embodiment of this invention.

  The function conversion message includes a message type 701, a message length 702, a post-conversion system name 703, and a communication destination card 704.

  In the message type 701, an identifier indicating that it is a function conversion message is registered. In the message length 702, the bit length of the function conversion message is registered.

  In the post-conversion system name 703, an identifier of a communication block assigned to the CPU after function conversion is registered. In the communication destination card 704, an identifier of a CPU that is a transmission destination of the function conversion message is registered.

  FIG. 8 is a flowchart of a function conversion determination process executed by the CPU 110 assigned to the control block according to the first embodiment of this invention.

  This function conversion determination process is repeatedly executed by the CPU 110 at a predetermined cycle.

  First, the CPU 110 refers to the received unnecessary ARP and determines whether or not the number of wireless communication terminals connected to the CPUs 111 to 113 assigned to the communication block satisfies a predetermined decrease condition (801). Since the determination process has been described with reference to FIG.

  If it is determined in step 801 that at least one of the CPUs 111 to 113 assigned to the communication block meets a predetermined decrease condition, the CPU 110 receives a signal from a CPU determined to satisfy the predetermined decrease condition. With reference to the session number message, it is determined whether or not the number of sessions established between the CPU and the wireless communication terminal determined to meet the predetermined decrease condition is equal to or less than a predetermined value (802).

  Since the number of connections based on unnecessary ARP is determined in the process of step 801, the number of unnecessary ARP receptions, that is, the total number of connections. For this reason, since it is not possible to specify whether or not the communication block is actually connected, determination is made based on the number of sessions indicating the number of actually connected to the communication block in the process of step 802. This is executed in order to prevent the CPU assigned to the communication block from being assigned to another communication block even though the number of actually connected wireless communication terminals is large.

  If it is determined in step 802 that the number of sessions established between the CPU determined not to meet the predetermined decrease condition and the wireless communication terminal is equal to or less than the predetermined value, the CPU 110 preliminarily determines the CPU. A function conversion message for executing the function conversion process in the closed state is transmitted to the CPU (803), and the process ends.

  If it is determined in step 801 that all the CPUs 111 to 113 assigned to the communication block do not satisfy the predetermined decrease condition, and the CPU determined in step 802 that the predetermined decrease condition is met. CPU 110 determines that the number of sessions established between the wireless communication terminal and the wireless communication terminal is greater than the predetermined value, CPU 110 ends the process without executing the process of step 803.

  As a result, the communication blocks of the CPU whose number of connections tends to decrease can be automatically assigned to other communication blocks.

  Next, the state in which the CPU 112 assigned to the HA block receives the function conversion message and transitions to the pre-blocking state and then is assigned to the P-GW block will be described with reference to FIGS. 9 and 10.

  FIG. 9 is an explanatory diagram of the CPU 112 that has transitioned to the pre-blocking state according to the first embodiment of this invention.

  In the process of step 507 shown in FIG. 5, the CPU 112 that has received the function conversion message transmitted from the CPU 110 assigned to the control block shifts to the pre-blocking state.

  When the CPU 112 assigned to the HA block transitions to the pre-blocking state, it does not accept a new connection of the wireless communication terminal. In order to show this state, in FIG. 9, it has shown with the dotted line between CPU112 and EVDO network.

  Accordingly, the CPU 112 does not increase the number of sessions established between itself and the wireless communication terminal until the session becomes 0, that is, until there is no wireless communication terminal connected to the CPU 112. It becomes a pre-occlusion state.

  When the CPU 112 enters a pre-blocking state, the CPU 114 assigned to the redundant block functions as an HA block. Therefore, the wireless communication terminal that has requested connection to the CPU 112 is connected to the CPU 114 assigned to the redundant block.

  FIG. 10 is an explanatory diagram of the CPU 112 newly assigned to the P-GW block according to the first embodiment of this invention.

  The CPU 112 that has transitioned to the pre-blocking state allocates itself to the P-GW block when the session established between itself and the wireless communication terminal becomes 0.

  Specifically, the function of the CPU 112 is obtained by loading the program data of the P-GW block from the storage device (not shown) of the integrated gateway device into the memory 132 and replacing the program data of the HA block loaded in the memory 132 so far. Is changed from the HA block to the P-GW block.

  When the CPU 112 is assigned to the P-GW block, the CPU 112 transmits a function conversion end message to the CPU 114 assigned to the redundant block.

  When the CPU 114 receives the function conversion end message, the CPU 114 moves the information related to the wireless communication terminal connected to the CPU 114 to the memory 131 of the CPU 111 assigned to the HA block. Connecting. Thereafter, information regarding the wireless communication terminal connected to the CPU 114 is stored in the memory 131.

  When the storable capacity of the memory 131 of the CPU 111 allocated to the HA block is less than the capacity of information related to the wireless communication terminal connected to the CPU 114, the CPU 114 pre-blocks the CPU 111 allocated to the HA block. After the state is changed and the capacity of the memory 131 becomes equal to or larger than the capacity of information related to the wireless communication terminal, the information related to the wireless communication terminal is moved to the memory 131.

  Next, when the number of wireless communication terminals connected to the CPU assigned to the communication block meets a predetermined decrease condition, the process of stopping the power supply to the CPU (power cut-off process) is shown in FIGS. Will be described.

  FIG. 11 is a sequence diagram of power-off processing according to the first embodiment of this invention.

  Of the power-off processing, the same processing as the function conversion processing shown in FIG. 5 is assigned the same reference numeral as in FIG.

  The process of step 1101 will be described.

  The CPU 110 assigned to the control block refers to the received number of sessions of each of the CPUs 111 to 113, and the CPU 113 assigned to a communication block (P-GW block) that is different from the CPU 112 determined to meet the predetermined reduction requirement. When the number of sessions is less than the predetermined value, it is determined to stop the power supply to the CPU 112 (1101).

  Then, the CPU 110 notifies a maintenance person of the integrated gateway device that power supply to the CPU 112 is stopped (506), and transmits a power-off message to the CPU 112 (1102).

  When the CPU 112 receives a power-off message, it puts itself in a pre-blocking state in which a new connection of the wireless communication terminal is not accepted, and when the number of sessions between the CPU 112 and the wireless communication terminal becomes zero, Supply is stopped (1103).

  FIG. 12 is a flowchart of the power-off determination process executed by the CPU 110 assigned to the control block according to the first embodiment of this invention.

  The power-off determination process is repeatedly executed by the CPU 110 at a predetermined cycle.

  Of the power-off determination process, the same process as the function conversion determination process shown in FIG.

  This power-off determination process is repeatedly executed by the CPU 110 at a predetermined cycle.

  If it is determined in step 801 that at least one of the CPUs 111 to 113 assigned to the communication block satisfies the predetermined decrease condition in step 801, the CPU 110 determines that the predetermined decrease condition is satisfied. It is determined whether the number of sessions of the CPU assigned to a communication block different from the communication block to which the assigned CPU is assigned is equal to or greater than a predetermined value (1201).

  If it is determined in step 1201 that the number of sessions of the CPU assigned to the different communication block is equal to or greater than the predetermined value, the CPU 110 determines that the CPU determined to meet the predetermined decrease condition is assigned to the different communication block. A function conversion message for conversion is transmitted (803), and the power-off determination process is terminated.

  When the number of sessions of CPUs of different communication blocks is equal to or greater than a predetermined value and there is a high possibility that different communication blocks cannot accommodate wireless communication terminals, a CPU that meets the reduction condition is newly assigned to the communication block. Occurrence of a situation where a communication terminal cannot be accommodated is prevented.

  On the other hand, if it is determined in step 1202 that the number of sessions of CPUs assigned to different communication blocks is less than the predetermined value, the CPU 110 puts the CPU determined to meet the predetermined decrease condition into a pre-blocking state. Then, a power-off message for executing the power-off process is transmitted to the CPU determined to meet the predetermined decrease condition (1203), and the power-off determination process is terminated.

  When the number of sessions of CPUs of different communication blocks is less than a predetermined value and there is a low possibility that different communication blocks cannot accommodate wireless communication terminals, power supply to CPUs that meet the reduction condition is stopped, which is useless. Power consumption can be reduced.

  If it is determined in step 801 that all the CPUs 111 to 113 assigned to the communication block do not satisfy the predetermined decrease condition, the power-off determination process is terminated.

  FIG. 13 is an explanatory diagram showing a state in which the power-off process is executed by the CPU 112 assigned to the HA block according to the first embodiment of this invention.

  In FIG. 13, the CPU 112 assigned to the HA block is determined to be a CPU that meets a predetermined decrease condition. Then, it is determined that the number of sessions of the CPU 113 assigned to the P-GW block which is a communication block different from the HA block assigned to the CPU 112 is less than a predetermined value.

  For this reason, the P-GW block cannot accommodate the wireless communication terminal, and it is unlikely that a situation in which other CPUs need to be allocated will occur. Therefore, supply of power to the CPU 112 that meets the predetermined reduction condition is stopped. doing.

  Next, if a CPU assigned to a certain communication block meets a predetermined increase condition, the CPU is likely to be unable to accommodate a wireless communication terminal. Processing will be described with reference to FIGS.

  FIG. 14 is a flowchart of an increase cause function conversion determination process executed by the CPU 110 assigned to the control block according to the first embodiment of this invention.

  The increase cause function conversion determination process is repeatedly executed by the CPU 110 at a predetermined cycle.

  First, the CPU 110 refers to the received unnecessary ARP and determines whether or not the number of wireless communication terminals connected to the CPUs 111 to 113 assigned to the communication block satisfies a predetermined increase condition (1401).

  Specifically, CPU 110 identifies the total number of wireless communication terminals connected to each CPU based on the received source IP address of unnecessary ARP. Then, the CPU 110 specifies the total number of connections (the previous total number of connections) of each CPU specified when the previous increase cause function conversion determination process is executed. Then, the CPU 110 subtracts the previous total number of connections from the total number of connections for each CPU, and determines whether or not the subtracted value is equal to or greater than a predetermined upper limit value.

  The CPU 110 determines that a CPU whose subtracted value is equal to or greater than a predetermined upper limit value is a CPU that meets a predetermined increase condition.

  If it is determined in step 1401 that there is a CPU that satisfies the increase condition, it is highly likely that a CPU that satisfies the increase condition will not be able to accommodate the wireless communication terminal. It is necessary to assign a new CPU.

  The CPU 110 executes the processing of steps 1402 to 1405 in order to specify the CPU to be newly assigned to the communication block assigned to the CPU that meets the increase condition.

  First, the CPU 110 determines whether there is a CPU whose power supply is stopped (1402).

  If it is determined in step 1402 that there is a CPU whose power supply is stopped, the CPU 110 assigns the CPU whose power supply is stopped to the communication block assigned to the CPU that satisfies the increase condition, A return message for starting the power supply of the CPU whose power supply is stopped is transmitted to the CPU (1403), and the process is terminated.

  On the other hand, if it is determined in step 1402 that there is no CPU whose power supply is stopped, the CPU 110 performs a session with a CPU assigned to a communication block different from the communication block assigned to the CPU that satisfies the increase condition. It is determined whether there is a CPU whose number is equal to or smaller than a predetermined value (1404).

  When it is determined in step 1404 that there is a CPU having a session number equal to or less than a predetermined value in a different communication block, the CPU 110 puts a CPU having a session number equal to or less than the predetermined value into a pre-occlusion state and meets the increase condition. A function conversion message for executing a function conversion process for converting into a function of the communication block assigned to is transmitted to the CPU (1405), and the process ends.

  If there are a plurality of CPUs having the number of sessions equal to or smaller than the predetermined value in the process of step 1404, the CPU 110 selects the CPU with the smallest number of sessions among these CPUs as the function conversion target CPU.

  On the other hand, if it is determined in step 1404 that there is no CPU whose number of sessions is equal to or less than a predetermined value in a different communication block, there is no CPU that converts the function of the communication block assigned to the CPU that meets the increase condition. The process is terminated as it is.

  If it is determined in step 1401 that there is no CPU that satisfies the increase condition, the process ends.

  FIG. 15 is a sequence diagram of an increase cause function conversion process when a CPU assigned to a communication block different from a communication block assigned to a CPU meeting the increase condition of the first embodiment of the present invention performs function conversion. .

  First, the CPU 113 assigned to the P-GW block detects that a wireless communication terminal is connected to the P-GW block (1501), and transmits an unnecessary ARP by broadcast (1502).

  Here, the CPUs 110 to 113 determine that the number of connections of the CPU 113 meets a predetermined increase condition (1503).

  The specific process of the increase condition determination process in the process of step 1503 is the same as that of step 1401 shown in FIG.

  Next, when it is determined in step 1503 that the number of connections of the CPU 113 meets the predetermined increase condition, the CPUs 111 to 113 assigned to the communication block establish a session established between itself and the wireless communication terminal. The CPU 110 assigned to the control block is notified of a session number message including the number (1504).

  When the CPU 110 receives the number of sessions from the CPUs 111 to 113, the CPU 110 selects a CPU (function conversion target CPU) to be newly assigned to the P-GW assigned to the CPU 113 whose number of connections meets a predetermined increase condition. Specifically, the CPU 110 executes the processing of steps 1402 and 1404 shown in FIG. 14, and the number of sessions is less than or equal to a predetermined value among the CPUs whose power supply is stopped or the CPUs assigned to the HA block. The CPU with the smallest number of sessions is selected as the function conversion target CPU.

  In FIG. 15, the CPU 110 selects the CPU 112 assigned to the HA block as the function conversion target CPU (1505).

  Then, the CPU 110 notifies the maintenance person of the integrated gateway device that the function conversion of the CPU 112 is to be executed (1506), and transmits a function conversion message to the CPU 112 (1507). The P-GW block is designated by the post-conversion system name 703 of the function conversion message.

  When the CPU 112 receives the function conversion message, the CPU 112 puts itself into a pre-blocking state in which a new connection of the wireless communication terminal is not accepted, and when the number of sessions between the CPU 112 and the wireless communication terminal becomes 0, the received function conversion Based on the message, the P-GW block is allocated (1508).

  FIG. 16 is a sequence diagram of an increase cause function conversion process when the CPU whose power supply is stopped according to the first embodiment of the present invention performs function conversion.

  Of the increase cause function conversion processing shown in FIG. 16, the same processing as in FIG.

  In FIG. 16, the CPU 111 assigned to the HA block does not receive the unnecessary ARP transmitted in the process of step 1502 because the power supply is stopped, and is assigned to the P-GW block in the process of step 1503. Further, it is not detected that the number of connections of the CPU 113 meets a predetermined increase condition. Further, the CPU 111 does not transmit a session number message including the number of sessions of the own node in the process of step 1504.

  Since there is a CPU 111 whose power supply is stopped, the CPU 110 selects the CPU 111 as a function conversion target CPU (1601).

  Then, the CPU 110 proceeds to the processing of step 1506, notifies the maintenance person of the integrated gateway device that the power supply to the CPU 111 is started, and outputs a return command to the CPU 111. When the power supply is started, the CPU 111 notifies the CPU 110 that the power supply is started. When notified that the power supply has been started, the CPU 110 transmits a return message including information that can specify a communication block (P-GW) to be assigned to the CPU 111 to the CPU 111 (1602).

  When receiving the return message, the CPU 111 assigns itself to the communication block specified by the return message (1603).

  FIG. 17 is an explanatory diagram when the function conversion process is executed by the CPU 111 that is assigned to the HA block according to the first embodiment of this invention and whose power supply is stopped.

  In FIG. 17, the CPU 113 assigned to the P-GW block is determined to be a CPU that meets a predetermined increase condition. Since there is a CPU 112 whose power supply is stopped, the CPU 110 selects the CPU 111 as a function conversion target CPU, restarts the power supply to the CPU 111, and allocates a P-GW block to the CPU 111.

  As described above, when there is a CPU that satisfies the increase condition among the CPUs assigned to the communication block, the number of sessions is less than or equal to a predetermined value among the CPUs that have stopped supplying power or the CPUs assigned to other communication blocks. The CPU having the smallest number of sessions is newly assigned to the communication block assigned to the CPU that meets the increase condition. As a result, even when the number of wireless communication terminals connected to a certain communication block is increasing, a CPU is newly allocated to the communication block, so that a situation in which the communication block cannot accommodate the wireless communication terminal can be prevented. .

(Modification of the first embodiment)
In the first embodiment, the CPU that is the function conversion target transitions to the pre-blocking state and executes the function conversion processing after the number of sessions becomes 0. However, in this modification, the function that does not shift to the pre-blocking state Perform the conversion process.

  A specific example will be described with reference to FIG.

  FIG. 18 is a flowchart of an increase cause function conversion determination process executed by the CPU 110 assigned to the control block according to the modification of the first embodiment of this invention.

  Of the increase cause function conversion determination processing shown in FIG. 18, the same processing as in FIG.

  If it is determined in step 1402 that there is no CPU whose power supply is stopped, the CPU 110 detects the latest communication from a CPU assigned to a communication block different from the communication block assigned to the CPU that satisfies the increase condition. Two CPUs with a small average value of the memory usage rate for a predetermined time are selected in ascending order (1801).

  Next, the CPU 110 determines whether or not the communication blocks assigned to the two CPUs selected in step 1801 can be integrated (1802).

  Here, integration refers to the data used by the CPU with the smaller average memory usage rate among the CPUs selected in step 1801 for communication with the wireless communication terminal. It is to move to the memory used by the CPU with the larger value.

  Specifically, CPU 110 specifies the number of sessions of each CPU selected in the process of step 1801. Then, the CPU 110 determines whether or not an addition value obtained by adding the number of sessions of each CPU is equal to or less than a predetermined value. If the addition value is equal to or less than the predetermined value, the two CPUs selected in the process of step 1801 It is determined that the communication blocks assigned to can be integrated.

  The CPU assigned to the communication block uses a memory in proportion to the number of sessions.

  If it is determined in step 1802 that the communication blocks assigned to the two CPUs selected in step 1801 can be integrated, CPU 110 determines the average memory usage rate among the CPUs selected in step 1801. The CPU with the lower value moves the data used for communication with the wireless communication terminal to the memory used by the CPU with the higher average memory usage rate, and the lower average memory usage rate A function conversion message for converting into a communication block assigned to the CPU satisfying the increase condition is transmitted to the CPU (1803).

  On the other hand, if it is determined in step 1802 that the communication blocks assigned to the two CPUs selected in step 1801 cannot be integrated, the CPU 110 uses the memory usage rate among the CPUs selected in step 1801. The CPU having the smaller average value is put into a pre-blocked state, and when the number of sessions of the CPU becomes 0, a function conversion message for converting into a communication block assigned to the CPU satisfying the increase condition is transmitted (1804). ).

  As described above, when a CPU that performs function conversion can be integrated with another CPU, it is not necessary to cause the CPU to transition to the pre-blocking state, so that function conversion can be performed quickly.

  Note that the processing in steps 1801 to 1804 can also be applied to FIGS. 8 and 12.

(Second Embodiment)
A second embodiment of the present invention will be described with reference to FIG.

  In the first embodiment, each block is assigned in units of CPUs. In the second embodiment, each block is assigned in units of cores of a multi-core CPU.

  FIG. 19 is an explanatory diagram of the multi-core CPU 1801 provided in the integrated gateway device according to the second embodiment of this invention.

  The multi-core CPU 1801 includes a plurality of cores 120 to 123 in one CPU, and performs processing as if each core is one CPU.

  In FIG. 19, for example, the core 120 is assigned to the control block, the core 121 is assigned to the HA block, the core 121 is assigned to the HA block, the core 122 is assigned to the P-GW block, and the core 123 is the redundant core. Is assigned.

  Since the function conversion process in units of cores can be realized by using the CPU of the first embodiment as a core, the description thereof will be omitted.

(Third embodiment)
In the second embodiment, each block is allocated in units of cores. However, in the third embodiment, the ratio in which the CPU resource of one CPU is allocated to each block (HA block and P-GW block) is changed. Is.

  When CPU resources in one CPU are allocated to the HA block and the P-GW block, the data format used in the HA block is different from the data format used in the P-GW block, so data is stored in the memory for each communication block. The area must be secured in advance.

  Here, when the number of connectable communication blocks is increased, the ratio of CPU resources allocated to other communication blocks must be decreased to increase the ratio of CPU resources allocated to the communication block.

  For example, when increasing the percentage of CPU resources allocated to the P-GW block and decreasing the percentage of CPU resources allocated to the HA block, the number of sessions in the HA block is reduced to the number of sessions that can be accommodated in the HA block after the CPU resources are reduced. It is necessary to let

  In other words, when the ratio of CPU resources allocated to the P-GW block is increased, the number of sessions that can be accommodated in the HA block is equal to the number of sessions that can be accommodated at the ratio currently allocated to the HA block (current maximum accommodated number). ) Is reduced to a value obtained by subtracting the number of sessions that can be accommodated (the number of objects to be removed) from the ratio to be reduced, and then the percentage of CPU resources allocated to the P-GW block is increased.

  Here, a sequence for changing the ratio of the CPU resources allocated to the P-GW block and the ratio of the CPU resources allocated to the HA block in the same CPU will be described.

  The control block notifies the communication block (reduction target block, for example, HA block) that reduces CPU resources of a message including the number of reduction targets.

  When a message is notified from the control block, the reduction target block accepts new connections from the wireless communication terminal until its own session count decreases to the value obtained by subtracting the reduction target number from the current maximum capacity. Transition to no pre-occlusion state. Then, if the number of sessions of the reduction target block decreases to the value obtained by subtracting the reduction target number from the current maximum capacity, the reduction target block is reduced in the storage area of the memory reserved for the reduction target block. Since there are as many free areas as there are objects, the free areas are allocated to the used areas of the P-GW block, and the pre-occlusion state is released.

  Thereby, the function conversion process can be executed without interrupting the connection to the wireless communication terminal currently connected to the reduction target block.

(Fourth embodiment)
Further, in the second embodiment, it is described whether or not to perform function conversion or the like based on the number of sessions and reception of unnecessary ARP from each communication block for each core assigned to each communication block. However, in the fourth embodiment, it is determined whether or not to perform function conversion or the like based on the transaction amount by communication with the core wireless communication terminal assigned to each communication block.

  In order to reduce the load on the core of a communication block with a large transaction amount, the control block assigns a core with a small transaction amount to a core assigned to a communication block different from the communication block assigned to the core. Assign to communication block.

  Specifically, the control block determines, for each core assigned to the communication block, whether or not there is a core whose transaction amount by communication with the wireless communication terminal is equal to or greater than a first predetermined value.

  When it is determined that there is a core whose transaction amount is determined to be greater than or equal to the first predetermined value, the control block includes a transaction amount among the cores allocated to communication blocks other than the communication block allocated to the core. It is determined whether or not there is a core having a value equal to or smaller than a second predetermined value. Here, the second predetermined value is set to a value smaller than the first predetermined value.

  When it is determined that there is a core whose transaction amount is equal to or smaller than the second predetermined value, the control block determines that the core is a communication block assigned to the core whose transaction is determined to be equal to or larger than the first predetermined value. assign.

  On the other hand, when it is determined that there is no core whose transaction amount is equal to or smaller than the second predetermined value, the control block is a communication block assigned to the core whose transaction amount is determined to be equal to or larger than the first predetermined value. Do not assign a new core. As a result, when a core with a transaction amount larger than the second predetermined value is assigned to a communication block assigned to a core whose transaction amount is equal to or greater than the first predetermined value, processing can be performed with the communication block assigned to the assigned core. Since the transaction amount is reduced, it is possible to prevent a situation in which the wireless communication terminal cannot connect to the communication block before allocation.

  The maintenance person of the integrated gateway device can set the first predetermined value and the second predetermined value according to the number of cores assigned to each communication system.

  Thus, function conversion can be executed in units of cores based on the transaction amount of each core.

  For example, when it is determined that the transaction amount by the LTE user is greater than or equal to the first predetermined value in the core allocated to the P-GW block, the control block is determined by the EVDO user among the cores allocated to the HA block. Assume that it is determined that there is a core whose transaction amount is equal to or less than the second predetermined value. In this case, the control block is set so that the core whose transaction amount is equal to or smaller than the second predetermined value is allocated to the P-GW block.

  As a result, the number of cores allocated to the P-GW block increases, and the amount of transactions that can be processed by the P-GW block increases. Therefore, the P-GW block can cope with an increase in transaction amount.

  Note that the communication block assigned to the core may be changed by changing the setting of the distribution core. In this case, the distribution core changes the communication block assigned to the change target core after the process for the user's call being processed in the change target core is completed.

  In the first to fourth embodiments, whether or not the integrated gateway device automatically changes accommodation can be set in the gateway device.

  When the integrated gateway device automatically changes accommodation, various predetermined values used for determining whether to change accommodation are set in advance by a maintenance person.

  Furthermore, based on the number of wireless communication terminals connected to each communication block, etc., it is possible to predict the trend of the communication block that is the connection destination of the wireless communication terminal and to change accommodation. For example, when the number of wireless communication terminals connected to a certain content increases and there is a possibility of further increase in the future, or when the number of connections suddenly increases with the start of the content, the connection to the content is made. The accommodation is changed to increase the connectable number of communication blocks.

  In addition, it is possible for the maintenance person to change the accommodation at a specified time in anticipation of an increase in the number of connections due to the introduction of a wireless communication terminal of a new communication method. Thereby, it is possible to secure the number of connections of wireless communication terminals of a new communication method.

110-114 CPU
130 to 134 Memory 601 Message type 602 Message length 603 Number of connected sessions 701 Message type 702 Message length 703 System name after conversion 704 Communication destination card

Claims (16)

In a gateway device for connecting a first communication network to which a communication terminal is connected and a second communication network to which the communication terminal is connected, which is different from the first communication network.
The gateway device includes at least one processor and a memory connected to the processor,
The processor resources include a first communication block that manages the first communication network, a second communication block that manages the second communication network, and the processor resources to the first communication block and the second communication block. Assigned to the control block that manages the allocation of
The control block is
A first connection number that is the number of connections of the communication terminal to the first communication network managed by the first communication block, and the communication to the second communication network managed by the second communication block; A connection number counting unit that counts the number of second connections that is the number of connections to the terminal so far;
An allocation changing unit that changes the allocation of the processor resources to the first communication block and the second communication block;
The allocation change unit
Calculating a first subtraction result obtained by subtracting the first connection number counted last time from the first connection number counted this time;
Calculating a second subtraction result obtained by subtracting the second connection number counted last time from the second connection number counted this time;
Comparing the calculated first subtraction result or second subtraction result with a first predetermined value ;
Allocating a processor resource assigned to a communication block corresponding to a subtraction result that is equal to or less than the first predetermined value to another communication block that is not a communication block corresponding to a subtraction result that is equal to or less than the first predetermined value. A gateway device characterized. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second communication block transmit connection signals to other blocks,
The first connection number is the number of times the control block has received the connection signal from the first communication block,
The second connection number is the number of times the control block has received the connection signal from the second communication block,
Each communication block manages the number of sessions, which is the number of communication terminals currently connected to itself, and transmits the number of sessions to the control block at a predetermined timing.
The allocation change unit
When it is determined that the first subtraction result or the second subtraction result is less than or equal to the first predetermined value, the other communication block corresponding to the subtraction result that is less than or equal to the first predetermined value based on the number of received sessions Extract the number of sessions in the communication block,
Comparing the number of extracted sessions with a second predetermined value;
If the extracted number of sessions is equal to or greater than the second predetermined value, the processor resource allocated to the communication block corresponding to the subtraction result equal to or less than the first predetermined value is allocated to the other communication block;
If the number of sessions extracted is smaller than the second predetermined value, the number of sessions of the processor assigned to the communication block corresponding to the subtraction result that is less than or equal to the first predetermined value becomes 0, and then the resources of the processor The gateway apparatus according to claim 1, wherein the supply of power is stopped. The allocation change unit
Determining whether the calculated first subtraction result or second subtraction result is greater than a third predetermined value set to a value greater than the first predetermined value;
When it is determined that the first subtraction result or the second subtraction result is greater than the third predetermined value, it is determined whether there is a processor resource whose supply of power is stopped;
3. The gateway device according to claim 2, wherein when there is a processor resource whose supply of power is stopped, the processor resource is allocated to a communication block corresponding to a subtraction result larger than the third predetermined value. . The allocation change unit
When there is no processor resource whose power supply is stopped, the processor resource allocated to the other communication block that is not the communication block corresponding to the subtraction result larger than the third predetermined value is set to be less than the third predetermined value. 4. The gateway device according to claim 3, wherein the gateway device is assigned to a communication block corresponding to a large subtraction result. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second communication block transmit connection signals to other blocks,
The first connection number is the number of times the control block has received the connection signal from the first communication block,
The second connection number is the number of times the control block has received the connection signal from the second communication block,
Each communication block manages the number of sessions, which is the number of communication terminals currently connected to itself, and transmits the number of sessions indicating the number of sessions to the control block at a predetermined timing.
The allocation change unit
When it is determined that the first subtraction result or the second subtraction result is less than or equal to the first predetermined value, the number of sessions of the communication block corresponding to the subtraction result that is less than or equal to the first predetermined value based on the received session number Extract
Comparing the extracted number of sessions with a fourth predetermined value;
If the extracted number of sessions is less than or equal to a fourth predetermined value, the processor resources allocated to the communication block corresponding to the number of sessions less than or equal to the fourth predetermined value are less than or equal to the fourth predetermined value. 2. The gateway device according to claim 1, wherein the gateway device is assigned to the other communication block that is not a communication block corresponding to the number of sessions. The gateway device includes at least three processors.
The processor resources are allocated to the first communication block, the second communication block, and the control block in units of the processor,
The gateway apparatus according to claim 1, wherein the allocation changing unit changes the allocation of the processor resources to the first communication block and the second communication block in units of the processors. The processor comprises at least three cores;
The communication block allocation unit allocates resources of the processor in the core unit to the first communication block, the second communication block, and the control block,
The gateway apparatus according to claim 1, wherein the allocation changing unit changes the allocation of the processor resources to the first communication block and the second communication block in units of the core. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second block transmit unnecessary ARP by broadcast,
In the IP address of the transmission destination of the unnecessary ARP, the IP address of the first communication block or the second communication block that detects the connection of the communication terminal is registered,
The block that has received the unnecessary ARP transmits an ARP Reply only when the IP address of the destination of the received unnecessary ARP matches its own IP address,
The connection number counting unit
Counting the number of times the control block has received the unnecessary ARP from the first communication block as the first connection number,
The gateway device according to claim 1, wherein the control block counts the number of times the unnecessary ARP is received from the second communication block as the second connection number. A communication block for managing each network in a first communication network to which a communication terminal is connected and a gateway device for connecting the second communication network to which the communication terminal is connected, which is different from the first communication network. in the communication block allocation changing method for changing a resource allocation pulp processor to,
The gateway apparatus comprises at least one of said processor, and a memory coupled to said processor,
The processor resources include a first communication block that accommodates the first communication network, a second communication block that accommodates the second communication network, and the processor resources to the first communication block and the second communication block. Assigned to the control block that manages the allocation of
The method
The control block has a first connection number that is the number of connections of the communication terminal to the first communication network managed by the first communication block, and the second communication managed by the second communication block. A connection number counting step of counting a second connection number that is the number of connections of the communication terminal to the network to the present time;
The control block comprises an assignment changing step for changing the assignment of the processor resources to the first communication block and the second communication block;
In the allocation change step,
Calculating a first subtraction result obtained by subtracting the first connection number counted last time from the first connection number counted this time;
Calculating a second subtraction result obtained by subtracting the second connection number counted last time from the second connection number counted this time;
Comparing the calculated first subtraction result or second subtraction result with a first predetermined value;
Allocating a processor resource assigned to a communication block corresponding to a subtraction result that is equal to or less than the first predetermined value to another communication block that is not a communication block corresponding to a subtraction result that is equal to or less than the first predetermined value. A communication block allocation changing method as a feature. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second communication block transmit connection signals to other blocks,
The first connection number is the number of times the control block has received the connection signal from the first communication block,
The second connection number is the number of times the control block has received the connection signal from the second communication block,
Each communication block manages the number of sessions, which is the number of communication terminals currently connected to itself, and transmits the number of sessions to the control block at a predetermined timing.
In the allocation change step,
When it is determined that the first subtraction result or the second subtraction result is less than or equal to the first predetermined value, the other communication block corresponding to the subtraction result that is less than or equal to the first predetermined value based on the number of received sessions Extract the number of sessions in the communication block,
Comparing the number of extracted sessions with a second predetermined value;
If the extracted number of sessions is equal to or greater than the second predetermined value, the processor resource allocated to the communication block corresponding to the subtraction result equal to or less than the first predetermined value is allocated to the other communication block;
If the number of sessions extracted is smaller than the second predetermined value, the number of sessions of the processor assigned to the communication block corresponding to the subtraction result that is less than or equal to the first predetermined value becomes 0, and then the resources of the processor The communication block allocation changing method according to claim 9, wherein the power supply is stopped. In the allocation change step,
Determining whether the calculated first subtraction result or second subtraction result is greater than a third predetermined value set to a value greater than the first predetermined value;
When it is determined that the first subtraction result or the second subtraction result is greater than the third predetermined value, it is determined whether there is a processor resource whose supply of power is stopped;
11. The communication block according to claim 10, wherein when there is a processor resource whose supply of power is stopped, the processor resource is allocated to a communication block corresponding to a subtraction result larger than the third predetermined value. Assignment change method. In the allocation change step,
When there is no processor resource whose power supply is stopped, the processor resource allocated to the other communication block that is not the communication block corresponding to the subtraction result larger than the third predetermined value is set to the third predetermined value. 12. The communication block allocation changing method according to claim 11, wherein allocation is performed to a communication block corresponding to a large subtraction result. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second communication block transmit connection signals to other blocks,
The first connection number is the number of times the control block has received the connection signal from the first communication block,
The second connection number is the number of times the control block has received the connection signal from the second communication block,
Each communication block manages the number of sessions, which is the number of communication terminals currently connected to itself, and transmits the number of sessions indicating the number of sessions to the control block at a predetermined timing.
In the allocation change step,
When it is determined that the first subtraction result or the second subtraction result is less than or equal to the first predetermined value, the number of sessions of the communication block corresponding to the subtraction result that is less than or equal to the first predetermined value based on the received session number Extract
Comparing the extracted number of sessions with a fourth predetermined value;
If the extracted number of sessions is less than or equal to a fourth predetermined value, the processor resources allocated to the communication block corresponding to the number of sessions less than or equal to the fourth predetermined value are less than or equal to the fourth predetermined value. 10. The communication block allocation changing method according to claim 9, wherein the allocation is made to the other communication block that is not the communication block corresponding to the number of sessions. The gateway device includes at least three processors.
The processor resources are allocated to the first communication block, the second communication block, and the control block in units of the processor,
10. The communication block allocation changing method according to claim 9, wherein, in the allocation changing step, the allocation of the processor resources to the first communication block and the second communication block is changed in units of the processor. The processor comprises at least three cores;
Before Symbol first communication block, wherein the second communication block, and the control block, resource of the processor are allocated in the core unit,
10. The communication block allocation changing method according to claim 9, wherein, in the allocation changing step, allocation of the processor resources to the first communication block and the second communication block is changed in units of the core. When the first communication block and the second communication block detect the connection of the communication terminal, the first communication block and the second block transmit unnecessary ARP by broadcast,
Wherein the required ARP destination IP address, IP address of the detected first copies Shinbu lock or the second communication block connections of said communication terminal is registered,
The block that has received the unnecessary ARP transmits an ARP Reply only when the IP address of the destination of the received unnecessary ARP matches its own IP address,
In the connection number counting step,
Counting the number of times the control block has received the unnecessary ARP from the first communication block as the first connection number,
10. The communication block allocation changing method according to claim 9, wherein the number of times the control block receives the unnecessary ARP from the second communication block is counted as the second connection number.

Images