Object-oriented computer environment and related method

Abstract

An object-oriented computer environment is managed by storing, in a plurality of repositories accessible during the life cycles of objects, information required to initiate object operations. The repositories can contain location information identifying the storage location of system objects or information indicating preferences for a preselected manner of performing object operations. In another preferred form, the repositories can be assigned to different levels of a hierarchy to control the distribution of stored information.

Claims

What is claimed is:

1. A method of managing an object-oriented computer environment for use by a plurality of users comprising the steps of:

providing a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle including object creation, object activation and object deletion;

providing a plurality of object brokers to manage said objects in performance of said object operations;

storing, in a plurality of location repositories separate from said object implementation repositories, location information identifying the storage locations of respective ones of said object implementation repositories;

causing the object brokers to access the information in at least one of said repositories during the life cycles of one or more of said objects; and

using the information within the repositories during the life cycles of said objects to manage the use of system resources by said objects.

2. The method of claim 1 which further comprises:

accessing the object by retrieving said location information from one of the repositories and thereafter using the retrieved information to locate the object; and

changing said location information in response to a change in said storage location for an object.

3. The method of claim 3 wherein:

said location information is changed in the course of object operations.

4. The method of claim 2 wherein:

new objects are created, as needed, and said location information is changed to assign locations within the environment at which said new objects are stored.

5. A method of managing an object-oriented computer environment for use by a plurality of users comprising the steps of:

providing a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle including object creation, object activation and object deletion;

providing a plurality of object brokers to manage said objects in the performance of said object operations;

storing, in a plurality of profile repositories separate from said object implementation repositories, information indicating preferences for a preselected manner of performing said object operations;

causing the object brokers to access the information in at least one of said repositories during the life cycles of one or more of said objects; and

using the information within the repositories during the life cycles of said objects to manage the use of system resources by said objects.

6. The method of claim 5 which further comprises:

initiating object operations by retrieving said preference information from one of the profile repositories and thereafter using said preference information in the performance of said operations.

7. The method of claim 6 wherein:

said preference information is changed in the course of object operations.

8. The method of claim 5 wherein:

the information within at least one of said profile repositories includes information as to preferences of at least one user for a preselected manner of performing object operations, said preference including system resource information; and

said profile repository provides said user preference information in response to user identification.

9. The method of claim 8 wherein:

said user identification messages include information as to a type of object to be implemented and a type of operating system to be used.

10. The method of claim 5 wherein:

at least one of said profile repositories contains search information and information specifying default preferences for a preselected manner of performing object operations within a preselected service category; and

said at least one profile repository provides said default preferences in response to messages including search criteria and identifying said service category.

11. The method of claim 10 wherein:

said identifying messages further identify a type of object to be implemented and a type of operating system to be used.

12. The method of claim 5 wherein:

at least one of said profile repositories contains search information and information specifying default preferences for a preselected manner of performing object operations within a preselected geographic service category; and

said at least one profile repository provides said default system preferences in response to messages including search criteria and identifying said geographic service category.

13. The method of claim 5 wherein:

the contents of said at least one profile repository further specifies restrictions on which of the object brokers can be used to manage a particular object in the performance of object operations.

14. A method of managing an object-oriented computer environment for use by a plurality of users comprising the steps of:

providing a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle for performing object operations, each of said objects having a life cycle;

providing a plurality of object brokers to manage said objects in the performance of said object operations;

storing, in a plurality of system profile repositories, separate from said object implementation repositories, information to perform object operations;

assigning said implementation repositories to different levels of a hierarchy, said levels including at least a parent level, a peer level and a child level;

accessing information in at least one of said implementation repositories to activate one or more of said objects; and

the information stored in said system profile repositories including access information required to access the contents of said implementation repositories, said access information including information for at least one of said parent level, said peer level or said child level of said hierarchy.

15. The method of claim 14 wherein:

the system profile repositories are themselves assigned to different levels of said hierarchy and said access information includes information required to access other system profile repositories.

16. The method of claim 14 wherein:

the repositories at different levels of said hierarchy pass queries to each other to locate requested information.

17. The method of claim 14 wherein:

the repositories perform operations to replicate said information within and between said levels of said hierarchy, as required to make the information accessible throughout the computer environment.

18. The method of claim 17 wherein:

a preselected one of the repositories at each level of the hierarchy replicates changes in said information to the other repositories at the same level.

19. The method of claim 17 wherein:

at least one of the repositories at each level of the hierarchy replicates changes in said information to the repositories at the next lower level.

20. The method of claim 14 wherein:

one of said repositories passes a query to another of said repositories at a different level of the hierarchy when said one repository does not contain information requested by a computing entity of said environment.

21. The method of claim 20 wherein:

said other of said repositories is located above said one of said repositories in the hierarchy.

22. The method of claim 20 wherein:

said other of said repositories queries an additional repository at another level of said hierarchy if said other repository does not contain the requested information.

23. The method of claim 22 wherein:

the repository at which the requested information is ultimately found communicates the requested information or its address to said one of said repositories; and

said one of said repositories, in turn, communicates the requested information or its address to the computing entity requesting it.

24. The method of claim 23 wherein:

said one of said repositories replicates the requested information for storage therein.

25. An object-oriented computer environment for use by a plurality of users comprising:

a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle including object creation, object activation and object deletion;

a plurality of object brokers capable of managing said objects in the performance of said object operations; and

a plurality of location repositories separate from said object implementation repositories which contain location information indicating the storage locations within the environment of respective ones of said object implementation repositories, the information within said repositories being useable during the life cycles of said objects to manage the use of system resources by said objects.

26. The object-oriented computer environment of claim 25 wherein:

said location information is changed in response to a change in said location at which said object is stored.

27. The object-oriented computer environment of claim 26 wherein:

said location information is changeable in the course of object operations.

28. The object-oriented computer environment of claim 26 wherein:

said location information is changeable to assign a location within the environment at which a new object is stored when it is created.

29. An object-oriented computer environment for use by a plurality of users comprising:

a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle including object creation, object activation and object deletion;

a plurality of object brokers capable of managing said objects in the performance of said object operations; and

a plurality of profile repositories separate from said object implementation repositories which contain location information indicating the preferences for a preselected manner of performing object operations, the information within said repositories being useable during the life cycles of said objects to manage the use of system resources by said objects.

30. The object-oriented computer environment of claim 29 wherein:

said preferences are stored after an object type associated with said object operations is installed in said computer environment.

31. The object-oriented computer environment of claim 30 wherein:

said preference information is changeable in the course of object operations.

32. The object-oriented computer environment of claim 29 wherein:

the information within at least one of said profile repositories includes information as to preferences of at least one user for a preselected manner of performing said object operations, said preferences including system resource information, said user preference information being provided by said repository in response to user identification messages.

33. The object-oriented computer environment of claim 32 wherein:

said user identification messages include information as to a type of object to be implemented and a type of operating system to be used.

34. The object-oriented environment of claim 29 wherein:

at least one of said profile repositories contains search information and information specifying default preferences for a preselected manner of performing object operations within a preselected service category, said default preferences being provided in response to messages including search criteria and identifying said service category.

35. The object-oriented computer environment of claim 34 wherein:

said identifying messages further identify a type of object to be implemented and a type of operating system to be used.

36. The object-oriented computer environment of claim 29 wherein:

at least one of said profile repositories contains search information and information specifying default preferences for a preselected manner of performing object operations within a preselected geographic service category, said default characteristics being provided in response to messages including search criteria and identifying said geographic service category.

37. The object-oriented computer environment of claim 29 wherein:

the contents of said at least one profile repository further specifies restrictions on which of the object brokers can be used to manage a particular object.

38. An object-oriented computer environment for use by a plurality of users comprising:

a plurality of objects stored in object implementation repositories for performing object operations, each of said objects having a life cycle, said implementation repositories being assigned to different levels of a hierarchy including at least a parent level, a peer level and a child level;

a plurality of object brokers capable of managing said objects in the performance of said object operations; and

a plurality of system profile repositories repositories separate from said object implementation repositories which themselves are processing entities that process queries for information, said system profile repositories containing access information required to access the information in said implementation repositories, said access information including information for at least one of a parent level, a peer level and a child level.

39. The object-oriented computer environment of claim 38 wherein:

the system profile repositories are themselves assigned to different levels of said hierarchy and said access information includes information required to access other system profile repositories.

40. The object-oriented computer environment of claim 38 wherein:

the repositories at different levels of said hierarchy pass queries to each other to locate requested information.

41. The object-oriented computer environment of claim 38 wherein:

the repositories are capable of replicating information within and between said levels of said hierarchy, as required to make the information accessible throughout the computer environment.

42. The object-oriented computer environment of claim 41 wherein:

a preselected one of the repositories at each level of the hierarchy is capable of replicating changes in said information to the other repositories at the same level.

43. The object-oriented computer environment of claim 41 wherein:

at least one of the repositories at each level of the hierarchy is capable of replicating changes in said information to the repositories at the next lower level.

44. The object-oriented computer environment of claim 38 wherein:

one of said repositories is capable of querying another of said repositories at a different level of said hierarchy when said one repository does not contain information requested by a computing entity of said environment.

45. The object-oriented computer environment of claim 44 wherein:

said other of said repositories is located above said one of said repositories in the hierarchy.

46. The object-oriented computer environment of claim 44 wherein:

said other of said repositories is capable of querying an additional repository at another level of said hierarchy if said other repository does not contain the requested information.

47. The object-oriented computer environment of claim 46 wherein:

the repository at which the requested information is ultimately found is capable of communicating the requested information or its address to said one of said repositories; and

said one of said repositories, in turn, communicates the requested information or its address to the computing entity requesting it.

48. The object-oriented computer environment of claim 47 wherein:

said one of said repositories is capable of replicating the requested information for storage therein.

Description

BACKGROUND OF THE INVENTION

The present invention relates to computer and communications systems and, more particularly, to an object-oriented computer environment (for example, a comprehensive composite of communication and computing facilities) and method for managing the environment.

Recent advances in computer technology have brought about the proliferation of relatively low cost, yet extremely powerful computers. Despite the reduced cost of these computers, it is still impractical in many situations to install all of the information, computer software, computer processing power and other computing resources that a given user or group of users may need on a single computer. Thus, in many computing environments, the computing resources available to users are distributed among several machines. As the number of machines in a distributed environment grows, the problems associated with sharing resources grow as well.

A variety of networking solutions have been proposed to facilitate the sharing of resources among different computers. However, due to the large number of different microprocessors, operating systems and other computer components that may be used on a given computer in a distributed computing environment, managing resources in such an environment can be quite difficult.

In addition, as the number and variety of computers in a distributed environment increase, the administration of the environment becomes more difficult since more computers and more configurations must be maintained. Furthermore, since computers are often assigned to individual users in a distributed environment, the task of administering a computer may fall upon the user. This is troublesome because users often are neither capable of nor interested in administration tasks.

Finally, as specific computing resources become increasingly popular in a distributed computing environment, the demand for such resources increases. In the past, popular resources have often been duplicated to make them more widely available. However, as a distributed computing environment increases in size, the complexity of the network connecting the computers increases, as well. This, in turn, makes it more difficult to duplicate resources without undesirably complicating and slowing computer operations.

One method for reducing the administrative burden associated with complex computer systems, such as a distributed computing environment, involves the use object-oriented programming. In object-oriented programming, an object is defined as a combination of data and operations to be performed on that data. Such objects are invoked, as needed, to perform a given task. For example, if a user wants to draw an icon on a screen, the user invokes an object capable of drawing the icon. object-oriented programming has the advantage that a routine calling the object ("calling function") does not need to "know" how the object performs its operations. The object is defined solely by its interface, which consists of a set of named operations along with any parameters required to perform those operations. Thus, the calling function "knows" only the operation performed by the object and how to invoke it.

An object can be changed at any time to perform its function differently and the change is transparent to the calling function so long as the interface to the object remains unchanged. Thus, an object-oriented system can be upgraded or maintained with minimal changes to applications already present in the system. Programming code in an object-oriented system is therefore more modular and maintainable than that written using more traditional programming techniques. As a result, object-oriented programming reduces the overhead associated with changing program code or installing new code to perform old functions.

An industry consortium called the Object Management Group ("OMG") has developed a standard for object-oriented systems known as the Common Object Request Broker Architecture ("CORBA"). The CORBA standard calls for three basic system elements, that is, an object broker, an implementation repository and an interface repository. The object broker of CORBA provides a mechanism by which objects or other applications can make requests to and receive responses from other objects managed by the system. The interface repository is used by the object broker to determine which objects are being invoked by an object operation request and to create an interface to new objects as they are created. The implementation repository holds information that allows the object broker to access the actual executable code required to perform the object operations.

As currently employed, the object broker handles object requests from "client applications", which can be programs implemented by a user or machine-based requests stemming from operational events such as function calls. Here, a client application passes to the object broker an object reference identifying the desired object operation along with any required parameters. After receiving an object operation request from the client application, the object broker uses the object reference to retrieve the location of the executables from the implementation repository. The object broker then activates the executables, which causes the desired operation to be performed. Thereafter, the implementation passes any results back to the object broker, which passes them, in turn, to the client application. This completes the object operation sequence.

The CORBA approach makes use of multiple object brokers, with the various components of the system distributed, as needed, to accommodate growth or the reassignment of system resources. Each of these multiple object brokers is capable of accessing information in corresponding implementation and interface repositories. Object requests are transparent to the users in that an object operation can be performed on an array of several computing devices and the entity invoking the object operation need not be aware of where the object operation resides or is executed. Consequently, using the CORBA approach the individual user is freed from the responsibility of administrating where specific programs reside or where they are executed since these details are handled by the administrators of the CORBA-based system. In addition, by defining a fixed set of operations, the CORBA approach enables object brokers from different vendors to communicate with each other, causing the system to be more universally compatible.

Since the information on how to interface to an object and execute the object are contained in the interface and implementation repositories, respectively, new objects can be added to a CORBA-based system simply by updating the information in the respective repositories. Consequently, such a system can be upgraded easily to accommodate new or modified objects.

Because the object-oriented approach allows the actual implementation of an object's operations to be transparent to the user, existing applications need only be modified in their interface in order to function in a CORBA based system. Consequently, existing applications need not be completely rewritten in order to work with the system.

In spite of these advantages, however, CORBA does not address problems in a number of areas. First, CORBA does not provide a mechanism for moving objects after an object is created. Second, CORBA does not make an individual user's information accessible throughout the network or permit the system resources to be administered at either the user level or the object level. In addition, CORBA does not support the use of multiple versions of a given object type. CORBA also does not provide for an effective means of duplicating highly used resources.

Accordingly, it is desirable in many applications to provide an improved object-oriented computer environment permitting greater flexibility and adaptability in its operation and reducing the burden of system administration.

SUMMARY OF THE INVENTION

The present invention relates to an object-oriented computer environment managed by storing information required to perform object operations in a plurality of changeable repositories accessible to object brokers. The stored information can relate, for example, to preferences of users for the manner in which object operations are performed or to default preferences of users in a specific group or geographic area. The information is retrieved on a selective basis when object operations are requested, permitting objects to be customized to the needs of a user or a group of users during an object activation sequence.

In one embodiment, information identifying the storage location of an object is kept in a location repository reserved for that purpose and updated periodically. This is a form of "indirect addressing" which enables an object to be moved, or one object to be substituted for another, without altering the location reference utilized by client application programs that activate the object. Only the contents of the location repository itself need be changed. The objects invoked in any given situation therefore can be changed without affecting the object's client application.

Administration of the object-oriented environment is further simplified in another embodiment of the invention by organizing repository information in hierarchies and selectively replicating information within the hierarchies. The hierarchial arrangement facilitates growth by allowing information to be distributed throughout a system to any extent required. Information frequently used in one part of the hierarchy but not in another can be installed only in those parts where it is actually used, drastically reducing information storage. Furthermore, since the depth and width of any hierarchy can be increased as the system of the disclosed embodiment grows, the system reduces "bottlenecking" that often occurs when users access information from a common source.

In one embodiment, the invention provides a mechanism for each level of a hierarchy to access other levels. This is done by storing at each level the information needed to access repositories at that particular level, at higher levels or at lower levels of the repository.

Although the system of the invention is useful in a wide variety of circumstances to distribute and process information, it is particularly suited to the needs of "distributed computing systems" where system resources, such as networks, computing resources, computing applications and transmission facilities, are distributed over a geographical area or divided among several entities. A system constructed according to the invention can easily accommodate widespread growth and reallocation of resources from one location or entity to another.

The primary components of the disclosed system are: (1) objects written to perform specific tasks; (2) object brokers capable of handling requests to have object operations performed; and (3) changeable repositories containing information required to perform object operations and administer the larger system. The system performs an object operation when a client application (such as a user at a computer terminal, another object operation or a system administrator) sends a request to perform the operation to the object broker. The object broker then uses information retrieved from the repositories to activate the object and alter the manner in which the object operation is performed. After the operation is complete, any results may be sent back through the object broker to the client application that requested the operation.

By using information retrieved from changeable repositories to affect object operations, the invention permits a computer-based information processing and distribution system to expand, as needed, with relative ease and minimal administrative overhead. It also permits the components of the system to be distributed over as wide an area and among as many segments of the system as required. In addition, it enables the performance of object operations to be customized on the basis of user preferences, group preferences, default preferences or preferences based on the hardware and software on which an object is to run.

When repositories are accessible on a system-wide basis, all upgrades can be performed by a system administrator rather than a local user, significantly reducing the user's administrative workload. For example, the present development allows software upgrades to be made by an administrator merely changing the contents of appropriate information repositories. Thus, software can be upgraded or new software can be installed with minimal effort by the user.

Moreover, since the present invention can work within a distributed network, a user can utilize his or her personal computer or take advantage of resources available elsewhere in the network. Because the user's personal preferences and home operating environment are stored in information repositories, the user can access his or her home environment at any computer throughout the system.

In sum, the present invention provides a computer system or environment wherein the manner of processing or transmitting information can be altered in the course of object operations, and where the resources required to perform object operations can be distributed throughout the system, as needed. Furthermore, the invention provides this flexibility while simplifying administration of the system.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other features of the present invention may be more fully understood from the following detailed description, taken together with the accompanying drawings, wherein similar reference characters refer to similar elements throughout and in which:

FIG. 1 is a high-level system block diagram of an object-oriented computer environment constructed according to one embodiment of the present invention;

FIG. 2 is a block diagram of a portion of the system of FIG. 1 illustrating the flow of information in the course of object operations;

FIG. 3 is a block diagram showing the hierarchical arrangement of information repositories constructed in accordance with the invention;

FIG. 4 is a high-level flowchart illustrating a sequence of steps performed in the course of object operations;

FIGS. 5A, 5B and 5C together constitute a flowchart illustrating a sequence of steps performed in creating an instance of an object in accordance with the teachings of the invention;

FIGS. 6A, 6B and 6C together constitute a flowchart illustrating a sequence of steps performed in activating an object in accordance with the teachings of the invention;

FIGS. 7A, 7B and 7C together constitute a flowchart illustrating a sequence of steps performed in deleting an instance of an object in accordance with the teachings of the invention;

FIG. 8 is a flowchart illustrating a sequence of steps performed in searching through a repository to retrieve its contents in accordance with the teachings of the invention;

FIGS. 9A, 9B and 9C together constitute a flowchart illustrating a sequence of steps performed in locating a repository in accordance with the teachings of the invention;

FIGS. 10A and 10B together constitute a flowchart illustrating a sequence of steps performed in replicating the contents of a repository in accordance with the teachings of the invention;

FIG. 11 is a flowchart illustrating a sequence of steps performed in replicating the contents of a repository on-demand in accordance with the teachings of the invention; and

FIGS. 12A and 12B together constitute a flowchart illustrating a sequence of steps performed in accessing the computer environment in accordance with the teachings of the invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

For purposes of illustration and explanation, the object-oriented environment 20 of the present invention is discussed in terms of a CORBA-based system. The CORBA standard is described in a document entitled The Common Object Request Broker: Architecture and Specification, published in 1992 by the Object Management Group, Inc. and X/Open Company Limited, the teachings of which are hereby incorporate d by reference for all purposes. It will be understood, however, that the features of the invention may be applicable broadly to any object-oriented system or network, regardless of how it is implemented.

Referring now to the drawings, specifically FIG. 1, an object-oriented environment 20 constructed according to one embodiment of the present invention coordinates the storage, creation, invocation and other functions required to manage a system of software objects. The environment 20 has a plurality of object brokers 22-22n (lower left) capable of communicating with location repositories 24-24n and profile repositories 26-26n through at least one communication channel 28 to initiate object operations. Implementation repositories 36-36n also are provided.

The initiation process is begun by a user (not shown) acting through one or more client applications 30 by sending an object operation request message to the object broker 22 for example. The storage location(s) of the object, and any restrictions as to which object brokers 22-22n are entitled to access the particular object , are obtained by addressing one of the location repositories 24-24n with a message identifying the object to be activated. In the illustrated embodiment, this may be accomplished by the object broker 22. Any established preferences for the manner in which the user or a group of users wishes to implement objects of this particular type are then obtained from the profile repositories 26-26n in response to user identity and object type information. Specifically, such preferences are obtained from a form of the profile repositories 26-26n, i.e., user profile repositories 34 (FIG. 2). To account for all contingencies, a set of default preferences contained in the profile repositories 26-26n also may be obtained by the object broker 22 in response to a message identifying the applicable operating system, object type and geographic or other service area. Armed with this information, the object broker 22 retrieves the executable code and data libraries required to perform object operations and configures, or otherwise customizes, the object to perform in the intended manner. Note that the environment of FIG. 1 also includes one or more administration tools 32 and one or more miscellaneous services 40 as will be described later.

The repositories (FIG. 1) of the invention are preferably arranged in hierarchies, as illustrated in FIG. 3. In order to provide access to repositories throughout the hierarchy, each branch of the hierarchy preferably has a separate repository (not shown) containing information on how to access the repositories in a hierarchy. In addition, information within the repositories is selectively replicated throughout the hierarchy depending on the demand that exists for that information.

A high-level description of the program flow described above is depicted in FIG. 4. For purposes of ease of reference, the blocks of this flow chart will be preceded with the prefix "HL". As represented by Blocks HL1-HL4, a client application sends an object operation request message to the object broker 22 (FIG. 1) which, in turn, retrieves the location of the object and any applicable preferences from the location and profile repositories, 24-24n and 26-26n, respectively. As indicated at Blocks HL5-HL7, the object broker 22 uses this information to search an implementation repository 36-36n in order to retrieve the location of the executable and other information required to activate an object implementation. When this information is found, the object broker 22 retrieves it and loads it into the appropriate computing device. The object broker 22 then starts the object implementation and passes back any results from the implementation to the client application, as represented by Blocks HL8 and HL9.

A typical client application can take a wide variety of forms in the context of the present invention including, for example, a computer interface that enables the computer user to invoke object operation requests or a computer program that invokes an object operation request based on an occurrence in its program flow. Similarly, a client application can take the form of one object operation that invokes another object operation or any other computer application seeking to access information managed by the system.

The client application 30 may invoke an object operation by sending a message to the object broker 22 for example (FIG. 1) through the communications channel 28. The message identifies the object operation to be performed and includes any other parameters required. After the object broker 22 processes the message from the client application 30 and invokes the desired object operation, the object broker 22 passes any results from the object operation back to the client application 30.

The communications channel 28 typically is implemented using intra-process communications, e.g., subroutine calls, inter-process communications (through the operating system), or network communications (through a remote device over a network) depending on system and user requirements. For example, applications requiring high-speed communications to other applications on the same computer typically use intra-process communications. In contrast, applications communicating with geographically remote applications typically use some form of network communications. The implementation of these communication techniques is well known in the art and will not be described in detail here.

Administration tools 32 (FIG. 1, upper central) are used to administer the object-oriented environment 20. This use includes adding new objects to the system, upgrading the software in the system, changing the contents of the profile repositories 26-26n, adding new client applications and other basic system administration tasks. The administration tools 32 typically contain a user interface allowing a system administrator to access and alter the information in the system.

Miscellaneous services 40 (FIG. 1, upper right) typically include such services as billing services, security services, statistics services and delayed delivery services. These services are preferably used in any system where access to the system must be restricted, where users of the system must be billed for their usage, where the activity within the system is monitored in order to keep the system functioning smoothly, or where other specialized services are needed.

A typical implementation of the environment 20 includes numerous instances of each of the components illustrated in FIG. 1. For example, a particular system typically contains multiple client applications 30 and, as indicated, multiple profile repositories 26-26n. Also, as indicated, several object brokers 22-22n along with various administration tools 32 can be used depending on system resource demands and how the components of the system are distributed.

An examination of the object broker 22 and the information in the various repositories will illustrate how the object broker 22 uses information in the repositories to control the manner in which object operations are performed. Thus, the structure and operation of the object broker 22, the profile repositories 26-26n and the repository hierarchies are discussed hereinafter in detail.

    ______________________________________
    TABLE OF CONTENTS
    ______________________________________
    The Basic System      15
    Object Broker         15
    Repositories          16
    Location Repository   17
    Service Profile Repository
                          19
    User Profile Repository
                          22
    Implementation Repository
                          27
    Interface Repository  31
    Object Operation      31
    Object Creation       32
    Object Activation     42
    Object Deletion       46
    Repository Hierarchy  46
    System Profile Repository
                          48
    Locating a Repository 50
    Repository Replication
                          56
    Automatic Replication 57
    On-demand Replication 60
    Example               62
    Log-in Procedure      70
    Creating the Video Mail Message
                          73
    Sending the Video Mail Message
                          76
    Viewing the Video Mail Message
                          79
    ______________________________________

                  TABLE 1
    ______________________________________
    Location Repository
    ______________________________________
    Object #1:
           Data Storage Locations:
              Location #1
                .
                .
                .
              Location #n
           Accessor Object Brokers:
              Object Broker #1
                .
                .
                .
              Object Broker #n
           Object Broker Restriction Flag
    Object #2:
    .
    .
    Object #n:
    ______________________________________

                  TABLE 2
    ______________________________________
    Service Profile Repository
    ______________________________________
    Object Type #1:
    Search Parameters:
    Object Type Name
    Object Type Version
    Replication Control
    Last Update Time
    Last Access Time
    Object Resource Assignments:
    Object Storage Information
    Available Object Brokers
    Location Object Storage Information
    Default Attribute Values
    Operating System Implementation Defaults:
    Operating System #1:
             Implementation #1
               .
               .
               .
             Implementation #n
             Default Implementation
    Operating System #2:
            .
            .
            .
    Operating System #n:
    User Profile Required Flag
    Object Type #2:
    .
    .
    Object Type #n:
    ______________________________________

                  TABLE 3
    ______________________________________
    User Profile Repository
    ______________________________________
    User #1:
    User Top Container:
    Object #1
            .
            .
            .
    Object #n
    User System Resource Assignments:
    Home Implementation Repository
    Home Interface Repository
    Billing and Statistics Information
    Object Type Entries:
    Object Type #1:
             User Object Resource Assignments:
                Object Storage Information
                Available Object Brokers
                Location Object Storage Information
                Default Attribute Values
             Operating System Implementation Preferences:
                Operating System #1:
                   Implementation #1
                     .
                     .
                     .
                   Implementation #n
                   Default Implementation
                Operating System #2:
                  .
                  .
                  .
                Operating System #n:
             Object Destination Container
    Object Type #2:
            .
            .
            .
    Object Type #n:
    User Demographics
    User #2:
    .
    .
    User #n:
    ______________________________________

                  TABLE 4
    ______________________________________
    Implementation Repository
    ______________________________________
    Module #1:
    Object Type Entries:
    Object Type #1:
             Search Parameters:
                Operating System
                Implementation Preference
                Network Protocol
                Hardware Platform
                Object Type Name
                Object Type Version
                Object Broker Vendor
                Replication Control
                Last Update Time
                Last Access Time
             Executable Information:
                Executable Name
                Executable Path
                Executable Storage Location
                Executable Storage Type
                Library Path
                Runtime Library #1:
                   Library Name
                   Library Storage Location
                   Library Storage Type
                Runtime Library #2:
                  .
                  .
                  .
                Runtime Library #n:
    Object Type #2:
            .
            .
            .
    Object Type #n:
    Module #2:
    .
    .
    Module #3:
    ______________________________________

                  TABLE 5
    ______________________________________
    System Profile Repository
    ______________________________________
    Local Name
    Entry Name #1:
    Parent Level Name
    Parent System Profile Repository
    Parent Security Information
    Peer Master System Profile Repository
    Peer Replication System Profile Repositories:
    Peer Replication System Profile Repository #1
            .
            .
            .
    Peer Replication System Profile Repository #n
    Peer Security Information
    Child Level Names
    Child System Profile Repositories:
    Child System Profile Repository #1
            .
            .
            .
    Child System Profile Repository #n
    Child Security Information
    Implementation, Interface, and Service Repository
    Configuration:
    Parent Repository
    Master Peer Repository
    Peer Replication Repositories:
             Peer Replication Repository #1:
                Address Information
                Child Master Repositories:
                   Child Master Repository #1
                     .
                     .
                     .
                   Child Master Repository #n
             Peer Replication Repository #2:
               .
               .
               .
             Peer Replication Repository #n:
    Entry Name #2:
    .
    .
    Entry Name #n:
    ______________________________________