Electrical computers and digital processing systems: multicomput – Distributed data processing – Processing agent
Reexamination Certificate
1996-12-03
2002-11-12
Courtenay, III, St. John (Department: 2151)
Electrical computers and digital processing systems: multicomput
Distributed data processing
Processing agent
C709S241000, C709S217000, C707S793000
Reexamination Certificate
active
06480881
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a distributed information system for distributing and locating a plurality of information sources, and more particularly to an information access apparatus for searching required information such as an object, etc. within the system.
2. Description of the Related Art
Due to environmental changes in computer systems today, a computer environment in which an information management system is employed is changing from an intensive-type environment centering on database servers to a distributed object management system running on a plurality of workstations connected via a high-speed network.
Conventional database systems are intended to store and provide mainly numerical and character data. With the variety of information required, the distributed object management system needs to store and provide not only data such as numerals, characters, etc., but also various types of objects such as procedures, programs, data structures, etc.
These objects are distributed and stored in object servers mutually connected via a network, and users can satisfy their needs by requesting an information providing service of an object server storing the information that they require.
However, as the scale of the distributed information system in such a form grows, it becomes more difficult for the users to learn which object server stores the information that they require.
One of solutions to this problem is an agent technique employing an agent on behalf of a user. Here, the agent means, in the widest definition, a capability of performing an intelligent operation instead of it being performed by a user. In most cases, the agent is generated as a process on a computer.
By way of example, an agent described in telescript, which is the language based on remote programming, can perform a programmed process by autonomically moving within a network. (Japanese patent laid-open H07-182174, Priority Data: U.S. Ser. No. 08/090,521). There is also a method for delegating an artificial agent based on a user request to information sources in order to obtain necessary information, by taking advantage of this feature (Japanese patent laid-open H06-301577).
In addition, there is another method where an delegated agent allows information exchange between another agent and itself using a communication section (Japanese patent laid-open H07-225798, Priority Data: GB 9315448:2). Furthermore, an agent which does not move, but asynchronously performs a process while communicating with another agent, can be considered.
With the above described techniques, the users can obtain the necessary information by delegating an agent, making one agent and communicate with another, even if the users themselves do not know where the information sources are located.
However, the conventional agent techniques described above have the following problems.
According to the conventional agent techniques, an agent delegated by a user must obtain a path to information sources from one particular server storing path information about that path, or search for the path using a heuristic method. Accordingly, it takes a lot of time for the user to obtain an answer from the agent to the inquiry, or it is difficult for the user to estimate the amount of time until he or she can obtain an answer.
With a large-scale distributed system consisting of a plurality of information sources which are autonomically updated, it is difficult to know which information source is updated when. Therefore, if the particular server providing an agent with path information is introduced, there is great difficulty in collecting the path information stored in that particular server. As a result, the possibility that the latest path information could not be obtained increases significantly.
SUMMARY OF THE INVENTION
The object of the present invention is to provide an information access apparatus and its method for allowing required information to be obtained efficiently and quickly from various types of information sources, put into an electronic form and distributed over an extensive range, by taking advantage of information about the information sources, such as path information possessed by an agent.
In first and second aspects of the present invention, an information access apparatus, which corresponds to an information processing device on a client side being an access source, comprises an agent generating unit and an output unit.
In the first aspect, the agent generating unit generates a first agent section that obtains knowledge about at least one information source accessed by a second agent section autonomically, collecting information using a network, among a plurality of information sources, and obtains information from the at least one information source based on the knowledge. The output unit outputs the information collected by the first agent section.
The agent section used here is, for example, a process having a capability of performing an intelligent operation instead of a user. It includes a movable agent described in a distributed processing description language, and an immovable agent fixed in a computer. The knowledge about an information source indicates, for example, identification information identifying an information source, and utility information indicating a degree of utilization of the information source. The name of an information source, identifier, etc. are used as the identification information, while a score (to be explained later) of the information source, etc. is used as the utility information.
According to such a configuration, the first agent section can learn of the existence of the information source without making any direct access, and collect the necessary information from the information source. Therefore, a plurality of agent sections effectively share the knowledge about their information sources. In addition, an unidentified information source from which the required information could be obtained with high probability, can be found efficiently, by referencing the utility information about each information source.
In the second aspect, the agent generating unit generates a third agent section that autonomically accesses at least one information source among a plurality of information sources, obtains knowledge about the at least one information source, and notifies a fourth agent section of the knowledge using a network. An output unit outputs the information that the third agent section collects.
According to such a configuration, the fourth agent section can learn of the existence of the information source without making any direct access, and collect the required information from the information source, in a similar manner as in the first aspect. Therefore, a plurality of agent sections can share the knowledge about their information sources effectively.
Furthermore, an information access apparatus in the third aspect of the present invention, which corresponds to an information processing device located at an information source on a server side, comprises an agent management unit, information storing unit, and knowledge storing unit.
In the third aspect, the information storing unit stores information, while the knowledge storing unit stores knowledge about at least one information source accessed by a fifth agent section using a network, among a plurality of information sources.
The agent management unit provides a sixth agent section with information obtained from the information storing unit based on an access request made by the sixth agent section, stores the knowledge about any of the plurality of information sources in the knowledge storing unit after receiving it from the sixth agent section, and provides the sixth agent section with the knowledge about the at least one information source after extracting it from the knowledge storing unit.
According to such a configuration, the sixth agent section can learn of the existence of the information source without making any direct access, and collect the required information fr
Ishikawa Hiroshi
Kubota Kazumi
Bullock, Jr. Lewis A.
Courtenay III St. John
Fujitsu Limited
Staas & Halsey , LLP
LandOfFree
Information access apparatus and method for cooperatively... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Information access apparatus and method for cooperatively..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Information access apparatus and method for cooperatively... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2923384