JP2007067667A - Telephone exchange main device, telephone exchange system, and telephone directory data management method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a telephone exchange system capable of easily adding or erasing a main device relative to a network. <P>SOLUTION: A telephone exchange main device 10 is connected to the network so as to be connected to other telephone exchange main devices 20, 30 by peer-to-peer. Telephone directory data held by each main device are managed by distribution through the use of a DHT (a distribution hash table) 16 and a DHT control program 18. Consequently, each main device has no necessity to hold the telephone directory data held by another main device. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a telephone exchange system in which a plurality of telephone exchange main units are connected to each other via a network, and more particularly to a management method for the telephone directory data.

  A telephone exchange system is composed of a telephone exchange main unit and a plurality of telephones connected thereto, and a plurality of telephone exchange systems are connected to each other via a network to constitute a larger-scale telephone exchange system. Can do.

  For example, as shown in FIG. 6, a conventional telephone exchange system in which a plurality of telephone exchange systems are connected via a network includes a plurality of telephone exchange main devices 100, 200, 300 connected via the network, It consists of a plurality of telephones (only one of each is shown) connected to the telephone exchange main unit.

  The telephone exchange main apparatus 100 includes a CPU 101, an IP interface 102, a VoIP CODEC 103, a memory 104, and a telephone interface 108.

  The memory 104 includes a memory space 107 for storing management data such as telephone directory data, a memory space 105 for setting information of other systems (telephone exchange main apparatus 200 or 300 and a telephone connected thereto), and each call. Includes a memory space 106 for managing the name and telephone number (extension number) of the other party.

  When the telephone exchange main unit 100 receives a call from the telephone via the telephone interface 108, it searches the telephone directory data stored in the memory space 107 to recognize the connection destination. Then, the telephone exchange main apparatus 100 issues a connection request to the recognized connection destination, and sets a call if a response is made.

  The telephone exchange main devices 200 and 300 are each configured in the same manner as the telephone exchange main device 100 and operate in the same manner.

  Such a telephone exchange system is described in Patent Document 1, for example.

  On the other hand, in the field of computer network systems, various network systems have been proposed that use a distributed hash table for peer-to-peer networks, enable high-speed searching and high-speed routing, and can also support frequent node join / leave. (For example, refer to Patent Document 2).

JP 2004-31440 A Japanese Patent Application Laid-Open No. 2004-266796

  In the conventional telephone exchange system, when a new telephone exchange main unit is connected to the network, not only the setting of the newly connected telephone exchange main unit but also the setting change of the telephone exchange main unit already connected to the network Also need to do. Also, when any of the telephone exchange main apparatuses connected to the network is disconnected from the network, it is necessary to change the setting of the other telephone exchange main apparatus. This is because each telephone switch main unit does not have a mechanism to refer to the data held by other telephone switch main units, so all the telephone switch main units connected to the network are updated to have the same phone book data. Because you have to do that.

  As described above, the conventional telephone exchange system has a problem in that complicated work is required with the addition / deletion of the telephone exchange main unit to / from the network. Further, since the setting work of each telephone exchange main unit is performed manually, there is a problem that erroneous setting is likely to occur.

  Further, the conventional peer-to-peer network has a problem that application to a telephone exchange system is not considered at all.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a telephone exchange system that can easily add or delete a main device from a network.

  The present invention relates to a telephone exchange main apparatus connectable to an IP network, network connection means for peer-to-peer connection with one or more other telephone exchange main apparatuses connected to the IP network, and telephone book data to the other And a data management means having a distributed hash table for distributed management with the telephone switch main unit.

  The telephone exchange system according to the present invention is characterized in that a plurality of the telephone exchange main devices are connected to the IP network.

  Furthermore, the present invention provides a telephone book data management method for a telephone exchange system in which a plurality of main devices are connected to each other via an IP network. The phone book data held by each of the plurality of main devices is distributed using a distributed hash table. The plurality of main apparatuses are distributedly managed.

  According to the present invention, a plurality of main devices are connected by peer-to-peer, and telephone book data is managed using a distributed hash table, so that addition and deletion of main devices can be easily performed.

  In addition, even if a failure occurs in any of the main devices, other system functions can be maintained.

  Hereinafter, a (large-scale) telephone exchange system according to an embodiment of the present invention will be described in detail with reference to the drawings.

  In FIG. 1, telephone switch main units (hereinafter simply referred to as main units) 10, 20, and 30 are connected to telephones 40, 50, and 60, respectively, and small-scale telephone switch systems (hereinafter simply referred to as systems are small). This refers to a large-scale telephone exchange system.) These main devices 10, 20, and 30 are also connected to the IP network 40 and constitute a large-scale telephone exchange system. Since the main devices 10, 20 and 30 have the same configuration, the main device 10 will be described below.

  The main apparatus 10 includes a CPU 11 that operates under program control, an IP interface 12, a CODEC 13 that converts voice into voice packets, a memory 14 that stores telephone book data and DHT (distributed hash table), and a DHT control program that controls the DHT. 18 and an interface 19 with a telephone.

  The CPU 11 functions as a network connection unit that realizes a via-to-peer connection with another main apparatus connected to the network, together with the IP interface and the control program stored in the memory 14. Further, the CPU 11 functions as a data management means for managing distributedly with other main devices and telephone book data together with the DHT and DHT control program 18 stored in the memory 14.

  The memory 14 includes a cache space 15 that primarily stores search results using the distributed hash table, a memory space 16 that stores DHT, and a memory space 17 that stores telephone directory data. The memory 14 also has a memory space (not shown) for storing a routing table, a skip list (not shown), and other necessary information. The memory 14 may store a DHT control program 18.

  Next, the operation of the telephone exchange system of FIG. 1 will be described with reference to FIGS. Here, the Chord algorithm is assumed.

  When activated (step S301), the main device 10 first generates its own node ID (unique ID) (step S302). The own node ID is generated by the CPU 11 executing the DHT control program 18 to obtain a hash value (= node ID) using a predefined hash function (for example, SHA-1).

  Next, the main apparatus 10 broadcasts its own node ID to the neighboring system via the IP interface 12 (step S303), and waits for a response from the neighboring system.

  When receiving the node ID by broadcast, the main device of the neighboring system generates a response signal according to a predetermined rule and transmits it to the main device 10. The response signal includes, for example, the node ID of the other system managed by the routing table or the skip list held by the main apparatus of the neighboring system.

  The main apparatus 10 waits for a response from the neighboring system until a predetermined time has elapsed after broadcasting its own node ID. If there is no response, the main device 10 shifts to the standby state as it is.

  The main apparatus 10 acquires a node ID of another system by receiving a response from the neighboring system (step S204). Then, the main apparatus 10 generates a DHT (distributed hash table) 16 based on the response from the neighboring system (step S205).

  Next, the main apparatus 10 specifies a system (node ID = X) in which the search order on the logical network is next to the own system by using the node ID of another system included in the response of the neighboring system. Then, the main apparatus 10 requests the system with the node ID = X to delegate a part of the hash data managed by the system to the own system (step S206).

  The main device having the node ID = X that has received the data delegation request transmits to the main device 10 the data that the main device 10 should manage among the hash data managed by itself.

  When the main apparatus 10 receives the data transmitted from the main apparatus having the node ID = X (step S207), the main apparatus 10 registers the received data in the DHT. Further, the telephone book data 17 managed by itself is converted into hash data, and data other than the hash data to be managed by itself is transmitted to the other main device corresponding thereto. Thereafter, the main apparatus 10 shifts to a standby state (step S208).

  The other main device that has received the hash data from the main device 10 additionally registers the received hash data in its own DHT.

  Addition of a system including the main device 10 to a large-scale telephone exchange system and distributed management of data are realized.

  Next, the above operation will be further described with reference to FIG.

In the Chord algorithm, a plurality of main devices are logically connected in a ring shape as shown in the upper part of FIG. Here, to simplify the explanation, the maximum number of nodes is 2 ⁴ = 16. Further, it is assumed that the main devices A, B, D, and E are connected to the nodes 11, 0, 13, and 9, respectively, and the other nodes are empty.

  Now, assume that the main apparatus C is connected to the network and generates node ID = 5 as its own node ID.

  As described above, the main apparatus C broadcasts the node ID = 5 on the network. Since the main devices A, B, D, and E exist on the network, responses from them can be obtained.

  Each main device A, B, D, E knows at least the node ID and IP address of the main device located on both sides of the logical network. For example, the main device A knows the IP addresses of the main device E located at the node ID = 9 and the main device D located at the node ID = 13. Therefore, in the main device C, based on the responses from the main devices A, B, D and E, the main device located next to the node ID = 5 on the logical network is the main device E having the node ID = 9. , And its location on the physical network (IP address). Further, it is possible to recognize that the main device located before the node ID = 5 on the logical network is the main device B having the node ID = 0 and the position (IP address) on the physical network.

  On the other hand, the main device B with the node ID = 0 is changed from the main device E with the node ID = 9 to the main device C with the node ID = 5 by the main device located next on the logical network by the broadcast of the main device C. Recognize that Similarly, the main device E with the node ID = 9 is transmitted from the main device B with the node ID = 0 to the main device C with the node ID = 5. Recognize that

  Further, since the hash data to be distributed and managed by each main device is generated using the same hash function as that for generating the node ID, it is classified into 16 groups that are the same as the maximum number of nodes in the logical network. That is, when the main device is connected to all nodes on the logical network, each main device manages one group of hash data. If there is a node to which the main device is not connected, the main device connected to the next node manages the data. If the main device is not connected to the next node, the main device connected to the next node manages data corresponding to these empty nodes. For example, in FIG. 3, before the main device C is connected, the main device E connected to the node ID = 9 has nine groups to be managed by the main device connected to the node ID = 1-9. Manage hash data.

  If the main device C recognizes that the next main device located on the logical network is the main device E with the node ID = 9, the main device C requests the transfer of the hash data before the node ID = 5. The main apparatus E delivers the hash data to be managed by the main apparatus connected to the node ID = 1-5 to the main apparatus C.

  Data held by the main device C is converted into hash data and registered and managed in the DHT of the corresponding main device.

  As described above, the main device C is added to the large-scale telephone exchange system.

  Next, data retrieval in the telephone exchange system will be described.

  The data search in the Chord algorithm is based on a linear search that is always performed clockwise, but a high-speed search is made possible by introducing the concept of a skip list.

  For example, when the main apparatus B with node ID = 0 refers to data managed by the main apparatus A with node ID = 11, in the simple linear search, node ID = 0 → node ID = 5 → node ID 9 → node ID 11 in this order. A search is performed. However, when the skip list is used, node ID = 0 → node ID 9 → node ID 11 is satisfied.

  Further, the data search operation of the main apparatus 10 will be described with reference to FIGS. 1 and 4.

  In the processing of the (small-scale) telephone exchange system (step S401), when data managed by another system is required, a data request is generated (step S402).

  When the data request is generated, the main apparatus 10 obtains a hash value of the requested data using the DHT control program (step S403).

  Then, the main apparatus 10 checks whether the obtained hash value exists in the cache memory space 15 (step S404).

  If the obtained hash value is already stored in the cache memory space 15, the related data stored in the cache memory space 15 is extracted in association with it and used for internal processing (step S409).

  If the obtained hash value is not stored in the cache memory space 15, the IP address of the main apparatus to be requested for retrieval is acquired based on the routing table and the skip list (step S405).

  Next, the main apparatus 10 requests the main apparatus having the acquired IP address to search for the previously obtained hash value (step S406). If the received hash value is not the data managed by the main device that has received the search request, the main device requests the next main device on the logical network according to the routing table. When the hash value that is the target of the search request is found, the related data associated with the hash value is sent to the main apparatus 10 along the path opposite to the search request (step S407).

  The main device 10 stores the received hash value and related data in the cache memory space 15 (step S408), and then uses them for internal processing (step S409).

  When the main device 10 receives a search request from another main device, if the received hash value is not data managed by itself, the main device 10 requests the next main device on the logical network according to the routing table. If the received hash value is data managed by itself, the DHT 16 is searched, and the corresponding related data is read out and transmitted to the requesting main apparatus.

  As described above, in the large-scale telephone exchange system according to the present embodiment, distributed data management and high-speed retrieval can be realized.

  According to the large-scale telephone exchange system according to the present embodiment, the telephone exchange main unit is connected to the network through the IP interface, and the telephone exchange main unit is connected by peer-to-peer connection. (ID setting) only needs to be performed, and there is no need for manual setting of the connection destination. Further, when deleting the system, there is no need to manually change the setting. That is, in the large-scale telephone exchange system according to the present embodiment, rewriting of DHT and the like accompanying the addition and deletion of the main device (small-scale telephone exchange system) is performed autonomously, so that the addition and deletion of the system can be easily performed. . In addition, since the system deletion DHT and the like are rewritten autonomously, by performing the same processing when a failure occurs, systems other than the system in which the failure has occurred can operate normally.

  Also, in the large-scale telephone exchange system according to the present embodiment, load distribution on the network and load reduction can be realized by using a distributed hash table.

  In addition, since the data once searched is stored in the cache memory space, search results can be obtained for data with a high search frequency without requesting other main devices to search. The load can be reduced.

  Next, a large-scale telephone exchange system according to another embodiment of the present invention will be described in detail with reference to FIG.

  FIG. 5 shows how a distributed hash table is managed using a CAN (Content-Addressable Network) algorithm. The configuration of each main apparatus is the same as that shown in FIG. 1, and the operation is also almost the same as the flowcharts shown in FIGS.

  In CAN, data managed by each system is represented as an n-dimensional torus space.

  Consider a (1,0) × (0,1) two-dimensional square space. A system connected to the network has hash values (x, y) (where 0 <x <1, 0 <y <1) in the logical space by hash functions called hash_x () and hash_y () Each space containing hash values is managed.

  In FIG. 5, the main device A has a hash value (a, b) and manages a space surrounded by (0, m) (u, t) (u, 1) (0, 1). Here, the hash value (a, b) corresponds to the node ID. Similarly, the main apparatus B has a hash value (e, f) and manages a space surrounded by (u, 0) (1, 0) (1, t) (u, t). The main device D has a hash value (g, h) (where u <i <1, t <j <s), and (u, t) (1, t) (1, 1) (u, 1 ). Further, the main apparatus E has a hash value (c, d) and manages a space surrounded by (0, 0) (u, 0) (u, t) (0, t).

  In the above state, if a system C having a new hash value (i, j) is connected to the network, the hash value (i, j) (where u <i <1, s <j <1) is , Included in the space managed by the system D. Therefore, the system C broadcasts to the neighboring system that it is connected to the network, and knows that the system D manages the space including itself.

  Thereafter, the systems C and D autonomously divide the space managed by the system D and manage the space surrounded by (u, s) (1, s) (1, 1) (u, 1). Delegate to C. Further, it notifies the adjacent system that management has been transferred to the system C for the space enclosed by (u, s) (1, s) (1, 1) (u, 1).

  When searching for data, the hash value (x, y) of the requested data is calculated and acquired using a predefined hash function. Based on the hash value, a request is transmitted to a system that manages a neighboring space and has a hash value closest to the value, and obtains data.

  For example, when the system E acquires data (q, r) managed by the system C, the requested point is not in its own management space, and therefore a request is transmitted to the system A. In the system A, since the point (q, r) is found in the space managed by the adjacent system C, a request for data is transmitted to the system C.

  As described above, even when this algorithm is used, a high-speed search is possible. Further, by using a cache memory, it is possible to obtain data at higher speed.

  Although the present invention has been described with reference to two embodiments, the present invention is not limited to the above embodiments.

  For example, the present invention can be used to connect VoIP telephones in the same telephone exchange system. It can also be applied to a network in which a telephone exchange system and a computer are mixed.

  In a network in which telephone exchange systems and computers are mixed, by sharing each other's data, changes in system data by each user and collection of information about each user by the telephone exchange (attended status and current status), etc. For example, and may be used in an application inside the telephone exchange system.

It is a block diagram which shows the example of 1 structure of the telephone switch system which concerns on one embodiment of this invention. 3 is a flowchart for explaining an operation when the main apparatus of the telephone exchange system of FIG. 1 is activated (when connected to a network). It is a figure for demonstrating the logical connection and data search operation | movement of the telephone switching system of FIG. 3 is a flowchart for explaining a data search operation of the telephone exchange system of FIG. It is a figure for demonstrating the logical space utilized for the data management in the telephone switching system which concerns on other embodiment of this invention. It is a block diagram which shows the example of 1 structure of the conventional telephone switching system.

Explanation of symbols

10, 20, 30 Telephone switch main unit 11 CPU
12 IP interface 13 CODEC
14 Memory 15 Memory space for cache 16 DHT
17 Phonebook Data Memory Space 18 DHT Control Program 19 Telephone Interface 40 IP Network 50, 60, 70 Telephone 100, 200, 300 Telephone Switch Main Unit 101 CPU
102 IP interface 103 CODEC
104 Memory 105 Memory space for connection destination information 106 Memory space for call control management 107 Memory space for telephone directory data 108 DHT control program

Claims (7)

In a telephone exchange main unit that can be connected to an IP network,
Network connection means for peer-to-peer connection with one or more other telephone switch main units connected to the IP network, and a distributed hash table for managing phone book data distributed among the other telephone switch main units And a data management means comprising: a telephone switching device main apparatus. In the telephone exchange main unit according to claim 1,
The main apparatus of a telephone exchange, wherein the data management means uses a Chord algorithm. In the telephone exchange main unit according to claim 1,
The telephone switch main unit characterized in that the data management means uses a CAN algorithm.   4. A telephone exchange system comprising a plurality of telephone exchange main apparatuses according to claim 1 connected to the IP network. In a telephone directory data management method for a telephone exchange system in which a plurality of telephone exchange main units are connected to each other via an IP network,
A telephone directory data management method characterized in that the telephone directory data possessed by each of the plurality of main apparatuses is distributed and managed by the plurality of main apparatuses using a distributed hash table. In the telephone directory data management method according to claim 5,
A telephone directory data management method characterized by using a Chord algorithm. In the telephone directory data management method according to claim 5,
A telephone directory data management method characterized by using a CAN algorithm.

Images