JP3092349B2 - Communication equipment with database - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device provided with a database, for example, a communication device for accessing a database storing management data of a communication network at a high speed.

[0002]

2. Description of the Related Art In a communication network, functions of performance management such as utilization rate and throughput of various communication devices constituting the network and a fault management function of detecting and notifying a failure of a network component are important. As these network management technologies, a method has been proposed in which a database is constructed in a form in which the components and functions of communication devices are abstracted, and management information is obtained by accessing the database. Generally, a database provided for such management purpose is a MIB (Management Information B).
ase).

[0003] In network management, a technique for quickly and accurately extracting a large number of items and a large amount of information relating to the performance and failure of various communication devices stored in the MIB is important. For this reason, the accumulated information of the MIB is structured, for example, such as a “tree structure” in which lower detailed information can be abstracted and grasped by a higher concept.

[0004] References regarding this type of database access method include, for example, N.I. Wirth,
"Algorithms: data Structur
es = Programs ", Prentice-Hal
1 (1976) (Japanese translation, Katayama: "algorithm + data structure = program", Japan Computer Association (197
8)), which shows, as an example of a method of accessing a database, a method called hashing or a method of accessing by an index using a multi-ary tree.

FIG. 3 shows an example of an MIB having a tree structure.
In this example, under the generic term "network"
There is a lower concept of "devices" that constitute a network, and as further lower concepts, there are a "cooling unit" and each "card". In this database, conceptual management units such as devices and their components are "objects"
It is expressed in functional units called. Information for managing each device is stored, for example, as a value of an object serving as a node in the tree structure or an object at the end of the tree structure. Each object is given a non-negative integer or character string as an identifier for identifying the object. Therefore, in order to access specific management information (object) in a database having a tree structure, an operation of following the tree structure is required. Usually, a numeric string or a character string in which identifiers of objects on the way from the root to the target object are sequentially connected is given as an information identifier of each management information.

For example, in FIG. 3, the information identifier of “alarm details” is “network.device.card.alarm.line alert.alarm content.alarm details” or “9.18.alarm details”.
Such as a character string or a number string separated by a period, such as "216.12.3.1.4".

[0007] The information identifier for accessing the MIB is encoded according to a predetermined rule to maintain compatibility. FIG. 4 shows an example of the encoding rule.

In network management, a variable-length numerical representation encoding is used to reduce the amount of communication when accessing an MIB provided in a remote communication device from a higher-level device (management device). For example, as shown in FIG.
A value obtained by dividing a binary number by 3 bits is defined as an expression unit 402. For a numerical value that cannot be expressed by 3 bits, “1” is set in a flag (f) bit 401 located at the head of each expression unit. This indicates that the expression unit is followed by a 1-bit flag and a 3-bit numerical value.

FIG. 5 shows an example of an information identifier encoding according to the above encoding rule. In this example, "9"
Is encoded as “10010001” and “18” is encoded as “101100010”. Also,
The entire information identifier is `` 10010001101000101011101100001001010000110001010
It is represented by a bit string like "0".

FIG. 6 shows an example in which the management database of FIG. 3 is constructed in a "multi-stage table" system. In the multi-level table system, a tree structure is represented by connecting a plurality of tables with pointers. For example, the table 503 for the “device” object is a pointer Ptr of the identifier “18” in the table 501 for the “network” object.
Is indicated by In the above database, the attributes of the target object can be known by sequentially following the table according to the pointer from the "network" object located at the highest level. For example,
The management information on the “alarm details” is obtained from the table 501 on the “network” object, the table 503 on the “device” object, the table 504 on the “card” object, the table 506 on the “alarm” object, and the table 507 on the “line alarm” object. "Alarm contents" via each
Is stored in the column indicated by the identifier “4” in the table 508 relating to

[0011]

However, the database of the multi-stage table system shown in FIG. 6 has a problem that it is difficult to speed up database access because it is necessary to sequentially follow pointers of each table by software processing. . Also, in each communication device, if the structure of the database or the number of elements in the table changes due to, for example, the addition of a card, a complicated pointer rewriting process is required, and it becomes difficult to manage the free space. is there. Also, when a multi-level table is mapped to a normal table and used, if the depth of the branches in the tree structure is different, the number of pointers to be passed before reaching the target information is different depending on the information. Becomes non-uniform. In addition, when the amount of accumulated information differs depending on the node position of the tree structure, the required table size differs depending on the node, so the table capacity becomes insufficient at the time of adding information, and additional processing such as table reconfiguration and overall movement is performed. Inconveniences such as necessity and increase in required time due to complicated access occur.

An object of the present invention is to speed up access to a management database.

It is another object of the present invention to speed up access to a database in which each information has an information identifier of a relatively long bit size.

Another object of the present invention is to provide a database particularly suitable for speeding up access to network management information and a device using the database.

[0015]

In order to achieve the above object, according to the present invention, when a search key is given to a management database storing management information, position information stored in advance corresponding to the search key is provided. Is provided, and a search key to be given to the associative storage device is generated in accordance with the content of the designated information identifier, and the search key is given to the associative storage device. The management database is accessed in accordance with the position information output from the device or the address specified by the position information and the information identifier. A management database storing management information for management, a processor, and a program to be executed by the processor are stored. A communication device for communicating control information between the storage device and another communication device, and when a search key is given, outputs position information stored in advance corresponding to the search key. Associative storage means, wherein the processor, which has received designation of an information identifier from another device via the communication interface means, provides a search key generated in accordance with the content of the information identifier to the associative storage means, By accessing the management database in accordance with the position information output from the associative storage means or the address specified by the position information and the information identifier, the management information corresponding to the information identifier is read or written. It is characterized by the following.

In the network management system, it is necessary for a management device located above each communication device to specify all management information of a large number of communication devices included in the network by a centralized identifier. Therefore, when a notification request or update request of specific management information in the database is made from the upper management device to each communication device, the identifier for specifying the target management information has a relatively large bit size. Will be. However, since the range of management information under the control of each communication device is limited, for example, in a database in which management information is structured like a tree structure, a specific range of information identifiers specified by a higher-level device is specified. The management information can be identified even if the bits are excluded.

In a preferred embodiment of the present invention, a search key is generated based on a specified portion of a specified information identifier bit, and position information is obtained from an associative storage device. The output of the associative storage device may be directly used as a database address, or a database address may be generated from the obtained position information and a predetermined portion of the information identification bits.

[0018]

In the database of the present invention, a database address is obtained by using an associative storage device. The associative memory device can store output information (position information) in an arbitrary address space without being restricted by the bit length of the search key, and can output information corresponding to the search key at high speed. As a result, access to the database can be sped up.

Further, according to the present invention, it is possible to cope with a change in the search key only by rewriting the content of the content addressable storage device. Therefore, it is possible to easily change the structure of the database, such as the names and identifiers of the nodes in the tree structure, the number of branches, and the depth of the branches. In addition, even when data is stored in a database in a manner that the data amount becomes non-uniform depending on the position on the tree structure, the search key is simply input to the associative memory as a bit string, and the storage position of the management information in the database is determined. Can be specified regardless of the structure of the search key. Also,
Even if the number of attributes required for access or the value of the attribute differs depending on the purpose information, the input to the associative storage device can be treated as a simple bit string,
A fixed access speed can be guaranteed regardless of the number of attributes.

In the present invention, when a method of generating a search key for an associative storage device from a part of a given information identifier is used, the bit length of each tag information to be stored in the associative storage device in advance corresponding to the search key is reduced. Therefore, the memory capacity required for associative storage can be reduced.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 shows an associative memory: index CAM (Contents Address) applied to a communication device according to the present invention.
1 shows a basic configuration of a database using a sable memory. Communication network management is performed by giving commands for setting or obtaining management information to individual communication devices located in various parts of the network from a higher-level management device that manages the entire network, and responding to each communication device. It is realized by doing. Each of the communication devices has an information identifier 10 that specifies network management information managed by the management database (DB) 2 when the command received from the upper management device is used.
When the management database 2 is included, the management database 2 is accessed based on this.

The communication apparatus of the present invention has a configuration in which the management database includes an associative memory (index CAM) 1, a search key extracting unit 3, and an information storage position determining unit 4. The search key extraction unit 3 receives the information identifier 1
A search key 102 is generated from 01 and passed to the index CAM1. The index CAM1 corresponds to a given search key 102, and corresponds to a management D set in advance.
In B, a pointer to the management information storage location, or location information 1 such as a procedure name for obtaining the management information storage location
03 is output. The position information 103 and one part 104 of the information identifier 101 output from the index CAM 1 are input to the information storage position determination unit 4 and converted into an address 105 indicating the management information storage position in the management database 2.
The management database 2 receives an access for writing or reading the management information 106 based on the output address of the determination unit 4.

The search key extraction unit 3 uses a part (or all) of the information identifier 101 to obtain a search key 102 to be given to the index CAM 1 and partial information 104 to be input to the information storage position determination unit 4. This function can be realized by, for example, an encoder that encodes the identifier of each clause of the MIB included in the input information identifier 101, and a selector that extracts a part of the output of the encoder. If the information identifier 101 is given as a numeric string, the search key extracting unit 3 only needs to have a function of selectively extracting a predetermined field of the input information.

The information storage position determining unit determines whether the index CA
Based on the position information 103 output from M1 and a part 104 of the input information identifier, for example, a functional process or a table indexing operation is performed to obtain a storage location 105 of the management information. In this case, the position information 103 may provide a pointer to the management database 2 or a process name or an identifier for obtaining a storage location. Also, location information 1
The storage position 105 may be obtained by giving a base address in the management database 2 as 03 and adding a numerical value indicated by the information 104 or another numerical value derived from the numerical value.

The above-described search key extracting unit 3 and information storage position determining unit 4 may be constituted by hardware circuits, or their functions may be replaced by software processing.

The reason why a part of the information identifier input by the search key extracting unit 3 is given to the index CAM is as follows. In a network management system, a higher-level management device necessarily requires an identifier having a long bit size to identify a large amount of management information under its jurisdiction. However, each communication device under the jurisdiction of the higher-level device handles management information limited to a certain range. Therefore, when the management information is managed by, for example, identification information having a hierarchical tree structure,
Looking at the individual communication devices or the management data involved,
The upper bits of the identification information are common, and only relatively lower identification information often has an information identification function. Even if the upper bits have an identification function, the significant bits are often limited specific bits. Further, among the lower bits having the identification function, the bits in a specific range have, for example, only a detailed item identification function in a table structure common to each data identified by higher bits. Often it is not something.

In the configuration shown in FIG. 1, an unnecessary bit portion which does not have an identification function in each communication device is excluded from information identifiers given by the host device, and identification of management information managed in each database is performed. Is given as a search key to the index CAM, thereby reducing the memory capacity required for the associative memory. If the information identifier 101 given from the higher-level device can be directly applied as the search key 102 in terms of the type and number of information to be managed and the capacity of the associative memory, the search key extracting unit 3 101 may be directly given to the index CAM1.
Also, the position information 1 output from the index CAM1
03 can be directly applied as the management information storage location 105, the information storage location determination unit 4
Is output to the management database 2 as it is.

FIG. 2 shows an index CA according to the present invention.
FIG. 4 is an explanatory diagram of a database access method using M1.

The index CAM1 is composed of the tag 11 and the R
It comprises an AM unit 12. The RAM unit 12 stores position information 103 to be a CAM output, and stores the position information 103 on a one-to-one basis.
Is stored in the tag unit 11 corresponding to the "tag". That is, the tag and the position information 103 are stored in the CAM1.
And stored as a pair. The tag corresponds to the given search key 102. If the bit length of the search key is short, the size of each tag can be short. Search key is CAM
When given to the tag 1, the tag matching the search key is searched, and the position information 103 paired with the tag is read from the RAM unit. That is, an index CAM (associative memory) is different from a normal memory in which information is read and written by designating an address. When a search key having a certain meaning is given, data (position information )
Can be automatically read.

When the bit length of the search key is long (that is, when the range of the search key is large), in order to access the memory area corresponding to each search key, in the case of a normal memory, the range of the search key is A proportional address space is required. On the other hand, when the associative memory is employed, the tag corresponding to the search key is automatically searched, and the R corresponding to the tag is searched.
Since the storage contents of the AM unit can be output, even when the range of the search key is large, data (position information) can be read at high speed with a relatively small-capacity memory corresponding to the number of data.

FIG. 7 shows a communication device according to the present invention to which the above-mentioned database access method using an associative memory is applied.
It is a block diagram showing an example.

In the figure, an internal bus 601, a microprocessor 61, a ROM 62, a RAM 63, and a communication interface unit (CU) 66 constitute basic functions of a communication device. In this example, the communication device is C
It is connected to another communication device, for example, a higher-level device of a network management system via the U66 and the line 609, and receives a control command from the higher-level device. The RAM 63 stores the CU 6
6, an input / output buffer 70 for temporarily storing a control command communicated with another communication device via the communication device 6, and a database 2 for storing network management information.
In addition, a memory area that is used as a work area by the microprocessor 61 is formed, and various programs for controlling the operation of the communication device executed by the microprocessor 61 are prepared in the ROM 62. In this example, the functions of the search key extraction unit 3 and the information storage position determination unit 4 shown in FIG. I / O controller 6 connected to internal bus 601
4 and an index CAM1 connected to the I / O controller 64 are additional device parts for accessing the database 631 at high speed. When accessing the database 2, the microprocessor 61
A search key is given to the controller 64. The search key is obtained by extracting from a management information identifier included in a control command received from another communication device via the CU 66.

The I / O controller 64 has a search key 60
2 and a control signal 603 for starting CAM are given to the index CAM1. When the search control signal 603 is enabled, the index CAM1 stores the search key 60
Data (positional information) 604 having a tag that matches 2;
The coincidence signal 605 is output. The match signal 605 is for indicating whether or not data having a tag matching the given search key exists in the index CAM1. When the match signal 605 has a value of “false”, the search signal This means that the data corresponding to the key did not exist in the management database 2, and the output data 604 in this case becomes invalid.

FIG. 8 shows an embodiment of the database access operation according to the present invention. In this example, the structure of the management database 2 has the tree structure described in FIG. 3, the information identifier 101 is encoded according to the rule described in FIG. 4, and the management information to be operated is shown in FIG. The case of “alarm details” represented by the encoding rule is shown.

A part 102 of the bit string of the encoded information identifier 101 is extracted as a search key and given to the index CAM1. From the index CAM1, the position information 103 of the purpose management information corresponding to the search key 102 is read. In this example, location information 1
Assume that the value of 03 was "$ b".

The position information 103 includes an information identifier 101
Of the prefix portion 104-1 that is a part of
Along with “a” and the value “$ c” of the suffix 104-2, the information storage location determination unit 4 calculates the storage address “$ z” 105 of the purpose management information in the database 2. In this example, the microprocessor performs a function processing F (#a, #b,
By executing $ c), the address "$ z" is obtained. The microprocessor accesses the address $ z of the database 2 to obtain the contents of the "alarm details" as the purpose management information (in this case, "1
(Signal interruption) ”).

In the above embodiment, the index CAM
The position information 103 obtained from 1 may be a function name. When the position information 103 is the function name f, the information storage position determination unit 4 obtains the storage address 105 by executing, for example, a function process f ($ b, $ c). Also,
When the position information 103 obtained from the index CAM1 indicates the storage address of the function, a subroutine is executed using the storage address as an argument, thereby storing the storage address 105 of the objective management information in the database.
Can be requested.

The information storage position judging section 4 can be realized by a software function executed by the microprocessor 61, but this can be replaced by dedicated hardware. In addition, this function may recursively use the process using the CAM.

As shown in the above embodiment, the management database 2 can store various management information belonging to each node of the tree structure. For example, the address $ (z-1) contains the name of the manufacturer of the card's individual information, and the address $ (z-1)
(Z + 1) stores line alarm occurrence location information. This is because there is no difference in access efficiency even if a single structure that does not reflect the structural meaning of the information identifier 101 is used because the tag portion of the index CAM1 is simply a bit string. That is, according to the present invention, there is no particular restriction on the structure of the database storing the management information, and even if the database structure is freely set, the access efficiency hardly changes. Therefore, even when the depth of the nodes in the tree structure and the amount of stored information are greatly different for each branch, the CAM
By automatically performing the structure conversion, the management database can be efficiently designed.

FIG. 9 shows another embodiment of the database access method in which the information identifier 101 encoded according to the rule shown in FIG. 4 is applied to the management database having the tree structure shown in FIG. In this example, the identifier of the purpose management information is “alarm detail” in the encoding example of FIG. 5, and shows a case where the information identifier 101 is used as the search key 102 as it is.

When the search key 102 is given, the index CAM1 retrieves the storage address of the target management information.
The position information 103 corresponding to z is read. This example shows a case where the position information 103 matches the management information storage address 105 in the database, and the CAM
The objective management information (“alarm details”) is stored in the storage location indicated by the output “$ z” of the above. In this case, CAM output 10
By accessing the database based on No. 3, the management information "1 (signal interruption)" can be referred to.

FIG. 10 is a block diagram showing another embodiment of the communication device according to the present invention. In this embodiment, the management database 2 is formed on an external storage device connected to the I / O controller 602, and the database is directly accessed by the output address of the index CAM1. The operation of reading the management information in the database only needs to give a control command including a search key from the microprocessor 61 to the I / O controller 64. In this case, the external storage device 65 is accessed based on the position information 604 read from the index CAM1,
Purpose management information 608 read from the external storage device 65
Is passed to the microprocessor 61 through the I / O controller 64. The operation of writing the management information to the database can be performed in the same manner as the above-described reading operation by providing the management information to the I / O controller by the microphone and the processor.

FIG. 11 is a block diagram showing still another embodiment of the communication device according to the present invention. In this embodiment, the index CAM1 is directly connected to the internal bus 601 via the bus interface 13, and the microprocessor 61 directly applies the search key to the index CAM1 without passing through the I / O interface to obtain the objective management information. Is obtained, and the database in the RAM 63 is accessed based on the obtained position information. In this embodiment, the index CAM1 is connected to one internal bus 601.
May be connected to a plurality of internal buses via a plurality of bus interfaces 13 so that the plurality of buses can access the buses independently. In this case, for example, it is possible to update and read the management information of the database from different control systems, so that the processing speed of network management can be further increased by a plurality of control systems.

The communication device provided with the management database has been described above. However, the above-described database access method is also applicable to devices other than communication devices that need to access management data with an input address having a relatively long bit size. can do.

[0046]

According to the present invention, since the database is accessed using the position information (database address) read from the associative memory, a relatively long bit size identifier is required for identifying management data. In this case, the address space of the database can be compressed. Also, the search key input to the associative memory is
High-speed access to the database is possible with a small-capacity associative memory by limiting to a specific bit portion of the information identifier provided from the host device and omitting the meaningless identification bit portion in the database of each communication device.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of a database access method applied in the present invention.

FIG. 2 is a block diagram showing a relationship between an index CAM and a management database, which are main parts of the present invention.

FIG. 3 is a diagram showing an example of a data structure of a management database provided in the communication device of the present invention.

FIG. 4 is a diagram showing an example of an information identifier encoding rule.

FIG. 5 is a view showing an example of encoding of an information identifier based on the encoding rule shown in FIG. 4;

FIG. 6 is a view showing a relationship between tables when the database shown in FIG. 3 is realized using a conventional technique.

FIG. 7 is a block diagram showing an embodiment of a communication device according to the present invention.

FIG. 8 is a diagram showing an example of an operation of accessing a management database in the communication device.

FIG. 9 is a diagram showing another embodiment of the operation of accessing the management database in the communication device.

FIG. 10 is a block diagram showing another embodiment of the communication device according to the present invention.

FIG. 11 is a block diagram showing still another embodiment of the communication device according to the present invention.

[Explanation of symbols]

1 ... Index CAM, 2 ... Management database, 3 ...
Search key extraction unit, 4 information storage position determination unit, 11 index CAM tag unit, 12 index CAM RAM unit, 13 internal bus interface, 61 microprocessor, 62 ROM, 63 RAM, 64 ...
I / O controller, 65: external storage device, 66: communication unit (CU), 70: input / output buffer, 101: information identifier, 102: search key, 103: position information, 10
4 part of information identifier, 105 storage location of management information,
106: management information, 401: a flag indicating that there is a subsequent bit, 402: a value obtained by expressing a search key in a binary number and divided into three bits, 501 to 508: a table representing each node of a database having a tree structure, 601 internal bus,
602: search key, 603: search enable signal, 60
4: Position information, 605: Search success / failure signal, 608
... management information, 609 ... communication path.

Continued on the front page (72) Inventor Setsuo Takahashi 216 Totsuka-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Within the Hitachi, Ltd. Information and Communication Division (72) Inventor Yusuke Mishina 1-280, Higashi-Koigabo, Kokubunji-shi, Tokyo Hitachi, Ltd. Inside the Central Research Laboratory (72) Inventor Shiro Tanabe 1-280 Higashi Koikekubo, Kokubunji-shi, Tokyo Inside the Central Research Laboratory, Hitachi, Ltd. (56) References JP-A-4-35439 (JP, A) IEICE Technical Report SSE92-65, "How to realize the management function in ATM transmission equipment" (1999-9-18) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04L 12/24 H04L 12/26 G06F 15/40

Claims (10)

(57) [Claims] An index CAM means for outputting, when a search key is given, position information stored in advance corresponding to the search key, in a communication device having a management database in which management information for network management is stored. Search key extracting means for generating a search key to be given to the index CAM means from at least a part of the specified information identifier; and the management based on the position information output from the index CAM means and the information identifier. Means for generating an address indicating a storage location of the target information in the database, wherein the management database is accessed by the address. 2. A management database (2) storing management information for network management, and a processor (61).
A storage device (62) for storing a program to be executed by the processor, and communication interface means (6) for communicating control information with another communication device.
6), and associative storage means (1) for outputting, when given a search key, position information stored in advance corresponding to the search key, information from another device via the communication interface means. The processor that has received the designation of the identifier gives the search key generated in accordance with the content of the information identifier to the associative storage means, and the position information output from the associative storage means,
Alternatively, the communication device is configured to read or write the management information corresponding to the information identifier by accessing the management database according to the address specified by the position information and the information identifier. 3. The communication device according to claim 2, wherein said processor (61) and said associative storage means (1) are connected to each other by an I / O.
A communication device connected via O interface means (64), wherein said search key and position information are exchanged between said associative storage means and said processor via said I / O interface means. . 4. A processor (61) and a storage device (6) for storing a program to be executed by the processor.
2), communication interface means (66) for communicating control information with another communication device, I / O interface means (64), and a search key connected to the I / O interface means. When given, the associative storage means (1) for outputting position information stored in advance corresponding to the search key, and connected between the I / O interface means and the associative storage means, A management database (2) storing management information for network management accessed using the output position information of the means as an address, wherein the processor receives designation of an information identifier from another device via the communication interface means Gives a search key having a value corresponding to the content of the information identifier to the associative storage means via the I / O interface means, A communication device for accessing a management database. 5. The communication device according to claim 1, wherein said management database has a tree structure in which an identifier and management information are defined for each node. . 6. The communication device according to claim 1, wherein the management database has a tree structure in which an identifier is provided for each section and management information is defined in a terminal section. A communication device comprising: 7. The communication device according to claim 5, wherein the management database has an identifier for each node,
A communication device having a tree structure in which management information is defined in a terminal node. 8. The communication device according to claim 5, wherein the information identifier is obtained by converting an identifier of at least one node for specifying the corresponding management information into a bit string according to a predetermined rule. A communication device characterized by being given as a device. 9. A communication device according to claim 1, wherein said index CAM means outputs information specifying a procedure for calculating an address of said management database as said position information. . 10. The communication device according to claim 2, wherein the associative storage means outputs information for specifying a procedure for calculating an address of the management database as the position information. Communication device.