KR0162764B1 - Method for building index directory system agent - Google Patents

Abstract

The present invention relates to a method of constructing an index directory system processor using a partial replication method. The present invention relates to a method of constructing an index directory system processor that is hierarchically composed in an X.500 directory system standardized by ISO or ITU-T. A method of constructing an index directory system processor for indexing to enable a response to a query even when a sequence is not known, comprising: a first step of configuring a directory system processor that is a local color to provide a country level indexing function; The first step is to configure the national local index directory system handler and then the second step is to configure the master index directory system handler to provide global indexing. The hierarchical structure provides efficient indexing for global directories. As well as partial replication In order to improve query response performance through the present invention, the present invention enables fast query processing by allowing a user to respond to a query even if they do not know the sequence of information organized hierarchically in the X.500 directory system. That has an effect.

Description

How to configure the index directory system handler

1 is a directory function model diagram.

2 is a diagram showing the components of a conventional DIT.

3 is an exemplary diagram of a configuration of a conventional DIT.

4 is a diagram of a conventional index DSA using alternate names.

5 is a block diagram of an index DSA using partial replication according to the present invention.

6 is a structural diagram of an index DSA layer.

7 is a flow chart of configuring a local index DSA.

8 is a flowchart of a process of configuring a master index DSA.

The present invention relates to a method of configuring an index directory system processor, and more particularly, to a method of configuring an index directory system processor using a partial replication method.

In general, the directory function is:

1 is a directory function model diagram.

Referring to Figure 1, the model of the directory function is described as follows.

An X.500 directory system (1) exists in the form of interconnected one or more Directory System Agents (DSAs), where users (3) access these DSAs through a Directory User Agent (DUA) 4 It performs actions such as searching, reading and writing information.

Information stored in the directory has a hierarchical structure, which is a directory information tree (hereinafter, abbreviated as DIT).

2 is a diagram showing the components of a conventional DIT.

Referring to FIG. 2, the components of the conventional DIT will be described.

The hierarchical directory entries (represented as 'nodes') 6 starting from the root 5 are again composed of a set of multiple 'attribute type-values' 7.

The alias name entry 8 represents an entry that refers to another entry.

Due to the advancement of the information and communication environment, the conventional centralized structure in which one system manages all the information as in the past has faced fundamental limitations in storing, providing and managing various information.

Therefore, the need for a directory with a decentralized structure has been raised. X.500, which is standardized by international standardization organizations such as the Internal Organization for Standardization (ISO) / Internal Telecommunication Union-Telecommunication Standardization Sector (ITU-T), is taking an open decentralized structure. Directories are considered a good alternative to meet these needs.

In other words, attempts are being actively made to manage various objects using the advantages of the X.500 directory, which has an object-oriented and distributed paradigm. A white-page service that provides various information of users is a representative example. to be.

The X.500 directory has even become the foundation of directory applications in the Internet, which is a typical open architecture, along with the Open Systems Interconnection (OSI) architecture.

However, although the hierarchical structure adopted in the conventional X.500 directory is good for the distributed environment management or the overall structure, problems may arise in the case of searching for query processing.

In other words, in the current X.500 directory, country names, global organization names, or region names (eg locality = Europe) are at the first level from the top level root, organization names at the second level, and the names of individuals below them. Or hierarchical structure to locate common names.

Thus, for example, to search for objects located on leaf nodes, the current technique uses the sequence of parent nodes (other than the name of the object in the directory name (Distinguished name)) or those You need to know some of the information before you can navigate.

In other words, in order to retrieve the information of a person named Hong Gil-dong, in most cases, at least a country name and an organization name belonging to it are required.

The user who wants to search for the relevant information often knows this additional information, so it may not be a big problem, but it is not necessarily the case.

In particular, users who are used to using Internet services such as Whois (++) may have a desire to enter additional information in addition to the name of the person they are looking for, but if they do not really know the additional information This method has the disadvantage that it is impossible or practical even if it is impossible to search.

3 is an illustration of the configuration of a conventional DIT.

An example of the configuration of a conventional DIT will be described with reference to FIG. 3.

This is an example of DIT composition consisting of three organizations (10) existing in the Korean DIT called 'c = KR' (9) and a unit called 'ou = Department of Electronic Engineering' (11).

Below the name of the organizational unit is the name of another unit or member.

In this case, if you want to find someone in an electronic engineering department at a university, you must know the country and organization name to search the directory.

This is because you need to determine which subtrees you need to search in order to effectively search for information in the DIT which can be huge.

However, as mentioned earlier, there was a limitation that users who did not know this information could not use the directory.

To overcome this drawback, the prior art constructs one index directory system handler (hereinafter referred to as DSA) with multiple alternative names using alternative names provided by the directory standard, or one large index DSA. All entries that are distributed and managed are stored in.

However, all of these methods have a problem in that the performance is accompanied from a practical point of view.

First, the first approach makes it easy to construct an index DSA, but using an alternative name only combines the relevant entries logically into the index DSA, which is very useful for query processing when managed by more than one DSA. It takes a lot of time.

On the other hand, the full duplication of all directory entries is not suitable for the X.500 architecture with distributed directory model, so the benefits offered by X.500 cannot be announced.

4 is a diagram of a conventional indexed DSA using alternate names.

Referring to FIG. 4, a conventional method using alternative names among several methods of configuring the index DSA 12 is described as follows.

This is the addition of separate nodes to provide indexing to the original DIT, depending on the type of attribute you want to configure, in this case a replacement pointing to all organization entries that exist within the Korean DIT according to all organization names (13) in Korea. To make a name (14).

In this case, note that in the case of the original DIT, if the parent entry is different through the hierarchical structure, even if the entry has the same name, uniqueness is guaranteed, but if you change to a flat structure using an alternative name, this property is not guaranteed. It may not.

Thus, there is a need for a naming system that can distinguish when more than one of the same name exists on the original DIT.

In addition, the configuration of the index DSA using alternative names has a problem that it takes a lot of time since the queries must be distributed to multiple DSAs to process a user's query if the alternative names refer to information belonging to different DSAs. do.

In addition, as the number of DSAs soared, it may not be possible to process the query itself.

In the meantime, the search of the actual entry through the alternative name requires about two to three times more time than the direct search.

One way to solve the above problem is to duplicate the contents of all entries to promote a single, unified indexing system.

Full replication can be easily reconfigured in an existing environment and is suitable when there is no strong replication mechanism such as the '88 DSA.

In addition, the DSA that does not implement the replication mechanism can be implemented by using an 'out-fo-band' replication technique such as FTP (File Transfer Protocol).

However, the full duplication of all directory entries is fundamentally placed in the structure of X.500 with a distributed directory model, which points out problems such as management and maintenance difficulties, directory capacity, and limitation of multiple query processing.

Accordingly, an object of the present invention to solve the above problems is to provide a method of configuring an index directory system processor that enables efficient indexing by partial replication of entries existing in a directory information tree (DIT).

A feature of the present invention for achieving the above object is that it is possible to answer a query even if the sequence of information organized hierarchically in the X.500 directory system standardized by ISO or ITU-T, which is an international standardization organization. A method of constructing an indexed directory system processor for indexing, comprising: a first step of providing a national level indexing function by configuring a local directory system processor for each country, and a country, local, index, directory as the first step; After configuring the system handler, the second step is to configure the master index directory system handler to provide global indexing. The hierarchical configuration provides efficient replication of the global directory as well as partial replication. Method to improve the performance of the question and answer.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In order to solve the problem of the alternative name method or the full replication method and to enable fast query processing, the present invention proposes a method of constructing an index DSA using partial replication entries.

5 is a block diagram of an index DSA using partial replication according to the present invention.

Referring to Figure 5 describes the configuration of the index DSA using a partial replication entry according to the present invention.

It can be seen from FIG. 4 that the partially duplicated entry is located in the alternate name entry portion.

This method of constructing the index DSA 15 through partial replication entries overcomes the shortcomings of the alternative name method, which provides only the information needed to navigate to the entry the user wants to see, and mitigates the inefficiency through full replication as above. At the same time, the performance can be improved.

The partial replication entry 16 referred to here refers to an entry which takes a subset of the attribute value pairs of the original entry and constructs the index when the index of the entry existing in the original DIT is to be made.

The reason for not simply duplicating the name here is to get the information you want to find even by querying the index DSA.

Additional attribute values to store may include things like your telephone number, fax number, e-mail address, uniform resource locator (URL), etc.

The new Recommendation of X.500, established by ITU-T in 1993, defines a robust replication mechanism that enables 'subtree replication' or 'incremental updating', so that the partial replication index DSA is constructed according to a standardized mechanism. can do.

6 is a structural diagram of an index DSA layer.

Referring to FIG. 6, the structure of the index DSA layer will be described.

From the perspective of the global DIT, including the entire world, to manage the partially replicated index DSA, we can consider the national local DSA 17 as shown in Figure 6 and the master index DSA 18 that integrates it. .

That is, the master index DSA is placed at the same level as the national or international organization, and the information stored there is periodically retrieved from the national local DSA and updated.

7 is a flow chart of configuring a local index DSA.

7 is a flowchart illustrating a method of configuring a DSA that is a local color according to a country.

First, in S1, a local index DSA is designated among countries.

In S2, the DSA constructs a Local Searchabel Attributes Table (LSAT) that contains information on which attribute types are stored in the index DSA.

In S3, a filter indicating which belonging attribute type is stored in each attribute type belonging to the S2 is specified.

S4 checks for any DSAs that have not been discovered.

If it is determined that there is a DSA that has not been searched for in S4, the index DSA in turn connects 22 to the DSA existing in the country.

In S6, the attribute type of the DSA in the LSAT is selected.

If the attribute type of the DSA is selected as the organization name in S6, S7 prepares to apply a filter corresponding to the organization name.

If the attribute type of the DSA is selected as the organizational unit name in S6, S8 prepares to apply a filter corresponding to the organizational unit name.

If the attribute type of the DSA is selected as the personal name in S6, S9 prepares to apply a filter corresponding to the personal name.

If the attribute type of the DSA is selected as the region name in S6, S10 prepares to apply a filter corresponding to the region name.

If the attribute type of the DSA is selected as an attribute type other than the organization name, the organization unit name, the personal name, or the region name in S6, then in S11, a filter corresponding to the other attribute type is ready to be applied. Do

In S12, a filter is applied to the corresponding filter prepared in S7 to S11 according to the attribute type in the LSAT.

In S13, each result is stored in the partial table of the attribute type to be searched, and the process proceeds to S6.

If the attribute type of the DSA is selected as the organization name, the organization unit name, the personal name, the region name, and other attribute types in S6, in step S14, the connection with the DSA is released and the process proceeds to S4. .

If it is determined that there is no DSA not searched at S4, it is determined at S15 whether the generated partial table exists.

If it is determined in step S15 that the partial table does not exist, the configuration of the local index DSA is completed, and thus the process ends.

If it is determined in S15 that the partial table exists, then in S16, a suitable common name field for containing the extraction information is created.

In S17, in order to exclude redundancy between the extracted names, a unique attribute value according to the name rule is assigned to each node of the DIT existing in the corresponding partial table.

In S18, the remaining information in the partial table is stored.

In S19, if there is an attribute value necessary for management, it is added.

In S20, the corresponding partial table is deleted.

After the configuration of the DSA, which is the local color of each country, the integrated master index DSA can be configured as shown in FIG.

8 is a flowchart of a process of configuring a master index DSA.

Referring to Figure 8 describes the process of configuring a master index DSA as follows.

In S31, the master index DSA is specified among the DSAs.

In S32, the DSA constructs a global searchable attribute table (GSAT) having information on which attribute types are stored in the index DSA from a global standpoint.

In S33, a filter indicating which attribute type is to be stored in each attribute type belonging to it is specified.

In S34, a check is made to see if there is an unsearched index DSA.

If it is determined in S34 that there is no unsearched index DSA, an additional attribute value necessary for management is assigned or an existing attribute value is appropriately modified and terminated.

If it is determined in S34 that there is no search index DSA, then in S35, the master index DSA connects to the DSA which is the local color of each country.

In S36, an attribute type of the DSA in the LSAT is selected.

If the attribute type of the DSA is selected as the organization name in S36, S37 prepares to apply a filter corresponding to the organization name.

If the attribute type of the DSA is selected as the organizational unit name in S36, S38 prepares to apply a filter corresponding to the organizational unit name.

If the attribute type of the DSA is selected as the personal name in S36, S39 prepares to apply a filter corresponding to the personal name.

If the attribute type of the DSA is selected as the region name in S36, S40 prepares to apply a filter corresponding to the region name.

If the attribute type of the DSA is selected as an attribute type other than the organization name, the organization unit name, the personal name, or the region name in S36, then in S41, a filter corresponding to the other attribute type is ready to be applied. Do

In S42, the index information is checked for existence.

If it is determined in step S42 that there is no index information, the flow proceeds to step S36.

If it is determined in S42 that the index information exists, S43 applies the filter according to the attribute type in the LSAT to the corresponding filter prepared in S37 to S41.

In S44, a suitable common name field is created in the master index DSA to contain information extracted from the local index DSA.

In S45, each result is separately stored in the corresponding field of the master index DSA, and the flow proceeds to S34.

In this case, it is assumed that each index entry is guaranteed global uniqueness as well as national uniqueness according to a naming convention that gives a single name throughout the world.

In step S36, if the attribute type of the DSA is selected as the organization name, the organization unit name, the personal name, the region name, and other attribute types, in S46, the connection with the corresponding DSA is released and the process proceeds to S34. do.

As described above, the present invention has an effect in enabling quick query processing by allowing a user to respond to a query even when the sequence of information hierarchically organized in the X.500 directory system is unknown.

Claims (4)

In a method of constructing an index directory system processor for enabling a response to a query even when the sequence of information hierarchically organized in the X.500 directory system standardized by the international standardization organization ISO or ITU-T is unknown. A first step of providing a national level indexing function by configuring a local directory system processor for each country; And a second step of configuring a master index directory system processor for providing a global indexing function after configuring a local index directory system processor for each country in the first step; A method of constructing an index directory system processor, which not only provides an efficient indexing function for a worldwide directory through a hierarchical structure but also improves query response performance through a partial replication method. The index directory system handler of claim 1, wherein the index directory system handler is not in violation of the X.500 model and can be created and periodically updated using a replication mechanism provided by the X.500 directory system standard. How to configure. The method of claim 1, wherein when configuring the local index directory system processor of the first step and configuring the master index directory system processor of the second step, a method of connecting a plurality of directory system processors by attribute types is provided. Rather, the method of constructing an index directory system processor characterized by extracting searchable attribute types and attribute values with a single connection. 2. The index directory system processor of claim 1, wherein a partial searchable categorical type table, a global searchable attribute type table, and an associated filter are designated to enable the construction of an effective index directory system processor and to avoid storing unnecessary information. How to configure DSA).