WO1999044134A1 - Method and system for facilitating access to a lookup service - Google Patents

Method and system for facilitating access to a lookup service Download PDF

Info

Publication number
WO1999044134A1
WO1999044134A1 PCT/US1999/004066 US9904066W WO9944134A1 WO 1999044134 A1 WO1999044134 A1 WO 1999044134A1 US 9904066 W US9904066 W US 9904066W WO 9944134 A1 WO9944134 A1 WO 9944134A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
stub
computer
client
receiving
Prior art date
Application number
PCT/US1999/004066
Other languages
French (fr)
Inventor
Robert Scheifler
Ann M. Wollrath
James H. Waldo
Original Assignee
Sun Microsystems, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US09/044,826 external-priority patent/US6832223B1/en
Application filed by Sun Microsystems, Inc. filed Critical Sun Microsystems, Inc.
Priority to KR1020007009230A priority Critical patent/KR20010034515A/en
Priority to JP2000533818A priority patent/JP2002505474A/en
Priority to EP99908444A priority patent/EP1058884A1/en
Priority to AU27878/99A priority patent/AU2787899A/en
Publication of WO1999044134A1 publication Critical patent/WO1999044134A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/28Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/42Loop networks
    • H04L12/427Loop networks with decentralised control
    • H04L12/433Loop networks with decentralised control with asynchronous transmission, e.g. token ring, register insertion
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28Databases characterised by their database models, e.g. relational or object models
    • G06F16/289Object oriented databases
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/465Distributed object oriented systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5055Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering software capabilities, i.e. software resources associated or available to the machine
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/54Interprogram communication
    • G06F9/547Remote procedure calls [RPC]; Web services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/54Interprogram communication
    • G06F9/547Remote procedure calls [RPC]; Web services
    • G06F9/548Object oriented; Remote method invocation [RMI]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00Indexing scheme relating to G06F9/00
    • G06F2209/46Indexing scheme relating to G06F9/46
    • G06F2209/462Lookup
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00Indexing scheme relating to G06F9/00
    • G06F2209/50Indexing scheme relating to G06F9/50
    • G06F2209/5015Service provider selection

Definitions

  • the present invention relates generally to data processing systems and, more particularly, to a method and system for facilitating access to a lookup service.
  • a number of personal computers, workstations, and other devices such as mass storage subsystems, network printers and interfaces to the public telephony system, are typically interconnected in one or more computer networks.
  • the personal computers and workstations are used by individual users to perform processing in connection with data and programs that may be stored in the network mass storage subsystems.
  • the personal computers/workstations, operating as clients typically download the data and programs from the network mass storage subsystems for processing.
  • the personal computers or workstations will enable processed data to be uploaded to the network mass storage subsystems for storage, to a network printer for printing, to the telephony interface for transmission over the public telephony system, or the like.
  • the network mass storage subsystems, network printers and telephony interface operate as servers, since they are available to service requests from all of the clients in the network.
  • the servers are readily available for use by all of the personal computers/workstations in the network.
  • Such a network may be spread over a fairly wide area, with the personal computers/workstations being interconnected by communication links such as electrical wires or optic fibers.
  • a client In addition to downloading information from servers for processing, a client, while processing a program, can remotely initiate processing by a server computer of particular routines and procedures (generally “procedures"), in connection with certain "parameter” information provided by the client. After the server has processed the procedure, it will provide results of its processing to the client, which the client may thereafter use in its processing operations.
  • the program will make use of a local "stub" which, when called, transfers the request to the server which implements the particular procedure, receives the results and provides them to the program.
  • the stub must be compiled with the program, in which case the information needed to call the remote procedure must be determined at compile time, rather than at the time the program is run.
  • the stub available to the client's programs is static, it may be at best the closest that can be determined should be provided for the program when it (the program) is compiled. Accordingly, errors and inefficiencies can develop due to mismatches between the stub that is provided to a program and the requirements of the remote procedure that is called when the program is run.
  • a new and improved system and method for facilitating the obtaining and dynamic loading of a stub is provided to enable a program operating in one address space to remotely invoke processing of a method or procedure in another address space, so that the stub can be loaded by the program when it is ran and needed, rather than being statically determined when the program is compiled.
  • the stub that is loaded can be obtained from the resource providing the remote method or procedure, and so it (the stub) can exactly define the invocation requirements of the remote method or procedure.
  • the stub can be located and dynamically loaded while the program is being run, rather that being statically determined when the program is compiled, run-time errors and inefficiencies which may result from mis-matches between the stub that is provided and the requirements of the remote method or procedure that is invoked can 5 be minimized.
  • the stub is obtained from a lookup service to provide access to a service defined in the lookup service.
  • a lookup service defines a network's directory of services and stores references to these services.
  • a client desiring use of a service on the network accesses the lookup service, which returns the stub information that facilitates the user's access of the service. The client uses the stub information to access the service.
  • FIG. 1 is a function block diagram of a computer network including an arrangement constructed in accordance with the present invention for facilitating the obtaining, dynamic loading and use of "stub" information to enable a program operating in one address space to invoke processing of a remote method or procedure in another address space;
  • FIGs. 2 and 3 are flow charts depicting the operations performed by the arrangement depicted in FIG. 1 , with FIG.2 depicting operations performed in connection with obtaining and dynamic loading of the stub information and FIG. 3 depicting operations performed in connection with use of the stub information to invoke processing of the remote method or procedure.
  • FIG. 4 is a diagram illustrating a lookup service consistent with the present invention
  • FIG. 5 is a flowchart illustrating a method of adding a stub to the lookup service consistent with the present invention.
  • FIG. 6 is a flowchart illustrating a method for retrieving a stub from a lookup service by systems consistent with the present invention.
  • FIG. 1 is a schematic diagram of a computer network 10 including an arrangement for facilitating dynamic loading of "stub" information to enable a program operating in one address space to remotely invoke processing of a method or procedure in another address space, where this method or procedure represents a network service.
  • computer network 10 includes a plurality of client computers 11(1) through 11 (N) (generally identified by reference numeral ll(n)), a plurality of server computers 12(1) through 12(M) (generally identified by reference numeral 12(m)), all of which are interconnected by a network represented by a communication link 14.
  • the network 10 may include at least one nameserver computer 13, which may also be connected to communication link 14, whose purpose will be described below.
  • the client computers 11 (n) are in the form of personal computers or computer workstations, each of which typically includes a system unit, a video display unit and operator input devices such as a keyboard and mouse (all of which are not separately shown).
  • the server computers 12(m) and nameserver computer 13 also typically include a system unit (also not separately shown), and may also include a video display unit and operator input devices.
  • the client computers 11 (n), server computers 12(m) and nameserver computer 13 are all of the conventional stored-program computer architecture.
  • a system unit generally includes processing, memory, mass storage devices such as disk and/or tape storage elements and other elements (not separately shown), including network interface devices 15(n), 16(m) for interfacing the respective computer to the communication link 14.
  • the video display unit permits the computer to display processed data and processing status to the operator, and an operator input device enables the operator to input data and control processing by the computer.
  • the computers 1 l(n) and 12(m) and 13 transfer information, in the form of messages, through their respective network interface devices 15(n), 16(m) among each other over the communication link 14.
  • the network 10 is organized in a "client-server" configuration, in which one or more computers, shown in FIG. 1 as computers 12(m), operate as servers, and the other computers, shown in FIG. 1 as computers 1 l(n) operate as clients.
  • one or more of the server computers 12(m) may, as "file servers,” include large-capacity mass storage devices which can store copies of programs and data which are available for retrieval by the 7 client computers over the communication link 13 for use in their processing operations.
  • a client computer 1 l(n) may also store data on the server computer 12, which may be later retrieved by it (the client computer that stored the data) or other client computers for use in their processing operations.
  • one or more of the server computers 12(m) may, as “compute servers,” perform certain processing operations in response to a remote request therefor from a client computer 1 l(n), and retum the results of the processing to the requesting client computer 1 l(n) for use by them (that is, the requesting client computers 1 l(n)) in their subsequent processing.
  • the server computers may be generally similar to the client computers 1 l(n), including a system unit, video display unit and operator input devices and may be usable by an operator for data processing operations in a manner similar to a client computer.
  • server computers may include only processing, memory, mass storage and network interface elements for receiving and processing retrieval, storage or remote processing requests from the client computers, and generating responses thereto.
  • client computer 1 l(n) may also perform operations described herein as being performed by a server computer 12(m), and similarly a server computer 12(m) may also perform operations described herein as being performed by a client computer 1 l(n).
  • the network represented by communication link 14 may comprise any of a number of types of networks over which client computers 1 l(n), server computers 12(m) and nameserver computers 13 may communicate, including, for example, local area networks (LANs) and wide area networks (WANs) which are typically maintained within individual enterprises, the public telephony system, the Internet, and other networks, which may transfer digital data among the various computers.
  • the network may be implemented using any of a number of communication media, including, for example, wires, optical fibers, radio links, and/or other media for carrying signals representing information among the various computers depicted in FIG. 1.
  • each of the computers typically includes a network interface which connects the respective computer to the communications link 14 and allows it to transmit and receive information thereover.
  • Systems consistent with the present invention facilitate the obtaining and dynamic loading of "stub" information to enable a program operating in one address space to invoke processing of a remote method or procedure in another address space, which may be located on 8 the same computer as the invoking program or on a different computer.
  • Java virtual machine in turn, is specified in the Lindholm andYellin, "The Java Virtual Machine Specification", Addison-Wesley, 1996, incorporated herein by reference.
  • programs in the Java programming language define "classes” and "interfaces.” Classes are used to define one or more methods or procedures, each of which may be invoked by reference to an interface.
  • a class may be associated with and extend a "super-class,” and in that regard will incorporate all of the interfaces and methods of the super-class, and may also include additional interfaces and/or methods.
  • a class may also have one or more sub-classes (and thus will comprise a super-class of each of its sub-classes), with each sub-class incorporating and possibly extending their respective super-classes.
  • An interface provides a mechanism by which a set of methods may be declared. In that connection, an interface identifies each method that is declared by the interface by, for example, a name, identities the data type(s) of argument(s) that are to be provided for the method, the data type(s) of retum values that are to be returned by the method, and identifiers for exceptions which can be thrown during processing of the method.
  • a class may indicate that it implements a particular interface, and in that connection will include the program code which will be used in processing all of the methods which are declared in the interface.
  • different classes may indicate that they implement the same interface, and each will have program code which will be used in processing all of the methods which are declared in the interface, but the program code provided in each class to for use in processing the methods may differ from the program code provided in the other classes which is used in processing the same methods; thus, an interface provides a mechanism by which a set of methods can be declared without providing an indication of the procedure which will be used in processing any of the methods.
  • An interface may be declared independently of the particular class which implements the method or methods which can be invoked using the interface.
  • a class that invokes the method and a class that actually implements the method will not need to share a common super-class.
  • a client computer 1 l(n) provides an execution environment 20 for interpreting the Java program.
  • the Java virtual machine includes a class loader 21 that, under control of a control module 19, can dynamically link instances of classes, generally identified in FIG. 1 by reference numeral 22, into the running program's execution environment while the program is being executed.
  • control module 19 effectively enables the class loader to retrieve uninstantiated classes, which generally identified by reference numeral 23, instantiate them and link them as class instances 22 into the execution environment's address space at the Java program's run time as the methods which the respective classes 23 implement are called.
  • class loader 21 can discard ones of the class instances 22 when they are not needed or to conserve memory. It will be appreciated that, if a class instance 22 has been discarded, it may be reloaded by the class loader 21 at a later point if it is then needed.
  • Systems consistent with the present invention provide an arrangement which facilitates the remote invocation, by a program executing in an execution environment 20 by a client computer 11 (n), of methods implemented by classes on a server computer 12(m).
  • the server computer 12(m) will also provide an execution environment 24 for processing, under control of a control module 28, the Java method.
  • the Java virtual machine which provides the execution environment 21 includes a class loader 25 (which may be similar to the class loader 21) that, under control of the control module 28, can dynamically link an instance of the class 26, to enable the method to be processed in the execution environment 24, and instances of other classes (also generally represented by reference numeral 26) which may be needed to process the remotely-invoked method.
  • control module 28 effectively enables the class loader 25 to retrieve an uninstantiated class for the method to be invoked, from a plurality of uninstantiated classes which are generally identified by reference numeral 27, instantiate it (that is, the uninstantiated class which provides the method to be invoked) and link it as a class instance 26 into the execution environment.
  • the class loader 25 can discard the class instances 26 when processing of the method has terminated. It will be appreciated that, if class instances 26 has been discarded, it may be reloaded by the class loader 25 at a later point if it is then needed. 10
  • nameserver computer 13 if provided, is generally similar to that of the server computer 12(m), and will not be separately described.
  • the control module 19 of the client computer's execution environment 21 makes use of one or more stub class instances generally identified by reference numeral 30 which are provided as part of the execution environment 21 in which the various class instances 22, including the class instance which is invoking the remote method, are being processed.
  • Each stub class instance 30 is an instance of an uninstantiated stub class 31, which the ser er computer 12(m) may maintain for the various class instances 26 and uninstantiated classes 27 which the server computer 12(m) has "exported," that is, which the server computer 12(m) makes available to client computers 1 l(n) for use in remote invocation of methods provided thereby.
  • An uninstantiated stub class 31 includes declarations for the complete set of interfaces for the particular remote uninstantiated class 27 which implements the remote method to be invoked, and also provides or invokes methods which facilitate accessing of the remote method(s) which are implemented by the remote class.
  • the a ⁇ angement by which the stub class instance may be provided to the execution environment 20 is similar to that described in the aforementioned Waldo, et al., patent application.
  • the server computer 12(m) provides a skeleton 32, which identifies the particular classes and methods which have been exported by the server computer 12(m) and information as to how it (that is, the server computer 12(m)) may load the respective classes and initiate processing of the particular methods provided thereby.
  • the server computer 12(m) contains a lookup service 400 for registering services on a network. The lookup service 400 will be discussed below.
  • a class instance invokes a remote method maintained by a server computer 12(m), it will provide values for various parameters to the stub class instance 30 for the remote method, 11 which values the remote method will use in its processing.
  • the remote method is implemented on the same computer as the invoking Java program, when the invoking Java program invokes a remote method, the computer may establish an execution environment, similar to the execution environment 20, enable the execution environment's class loader to load and instantiate the class which implements the method as a class instance similar to class instances 22, and process the remote method using values of parameters which are provided by the invoking class instance in the remote invocation.
  • the execution environment in which the remote method has been processed will provide the results to the stub class instance 30 for the remote method that was invoked, which, in turn, will provide to the particular class instance 22 which invoked the remote method.
  • client computer 11 (n) and server computer 12(m) are implemented on different physical computers.
  • the client computer l l(n) that is processing the invoking class instance 22, under control of the control module 19 for the execution environment 20 for the invoking class instance 22, will use the appropriate stub class instance 30 to communicate over the network represented by the communication link 14 with the server computer 12(m) which implements the remote method to enable it (that is, the server computer 12(m)) to establish an execution environment 24 for the class which implements the remote method, and to use the class loader 25 to load an instance of the class as a class instance 26.
  • the client computer 1 l(n), also using the appropriate stub class instance 30, will provide any required parameter values to the server computer 12(m) over the network 14. Thereafter, the server computer 12(m) will process the remote method using parameter values so provided, to generate result value(s) which are transferred over the network to the client computer 11 (n), in particular to the appropriate stub class instance 30. The client computer 1 l(n) will, after it receives the result value(s) from the network, provide them to the invoking class instance 22 for its processing.
  • control module 19 of the client computer's execution environment 20 determines that a reference to the remote object has been received, if it determines that the stub class instance 30 is not present when it receives the reference, it will attempt to obtain the stub class instance 30 from, for example, the server computer 12(m) which implements the remote method, and enable the stub class instance 30 to be dynamically loaded in the execution 12 environment 20 for the invoking class instance 22.
  • a reference to the remote object may be received, for example, either as a retum value of another remote method invocation or as a parameter that is received during another remote method invocation.
  • the stub class instance may be dynamically loaded into the execution environment in a manner similar to that used to load class instances 22 in the execution environment 22.
  • the execution environment 20 is provided with a stub class loader 33 which, under control of the control module 19, will attempt to find and load the stub class instances 30 as required by the class instances 22 processed in the execution environment.
  • the location of a particular server computer 12(m) that maintains the class that implements a method to be invoked remotely may be included in the call from the invoking class instance or may be made known to the stub class loader 33 through another mechanism (not shown) maintained by the client computer 1 l(n).
  • the stub class loader 33 may use the nameserver computer 13 to provide that identification.
  • the identification may comprise any identifier which may be used to identify a server computer 12(m) or other resource which is available on the network 14 and to which the server computer 12(m) can respond.
  • Illustrative identifiers include, for example, a network address which identifies the server computer and/or resource, or, if the network 14 is or includes the Internet, an identifier to, for example, a World Wide Web resource which may provide the identification or a "uniform resource locator" ("URL") which provides a uniform mechanism for identifying resources that are available over the Internet.
  • the server computer 12(m) which implements the remote method, in response to a request from the client computer 1 l(n) will provide stub class instance 30 which the client computer l l(n) may load into the execution environment 21 to thereafter enable the remote invocation to be initiated.
  • the stub class loader 33 may, under control of the control module 19, obtain the identification from the nameserver computer 13. In that operation, the stub class loader 33 may use a previously-provided default stub class which is provided for use in such cases.
  • the default class stub when used by the invoking Java program, enables the 13 computer that is processing the invoking Java program to communicate with the nameserver computer 13 to obtain information which can be used in invoking the remote method.
  • This operation is essentially the same as the invocation of a remote method to be processed by the nameserver computer 13 , with the remote method including a parameter identifying the class and method to be remotely invoked, and enabling the nameserver computer 13 to provide the identification of a server computer 12(m) which can process the method to the requesting client computer 1 l(n) and other information which may be helpful in communicating with the server computer 12(m) and invoking the particular method.
  • the nameserver computer 13 will maintain a table (not separately shown) of "exported" resources, that is, resources, such as classes and methods, that are available to client computers 1 l(n) connected to the network 14, and information, such as the identifications of the particular server computers 12(m) which provide those resources, which will be useful to the client computers l l(n) in making use of the exported resources.
  • “exported” resources that is, resources, such as classes and methods, that are available to client computers 1 l(n) connected to the network 14, and information, such as the identifications of the particular server computers 12(m) which provide those resources, which will be useful to the client computers l l(n) in making use of the exported resources.
  • the nameserver computer 13 may create and maintain the exported resource table in a number of ways that are known in the art.
  • the nameserver computer 13 may periodically broadcast requests for exported resource information over the network 14, to which the various server computers 12(m) which maintain exported resources may respond; in that case, the nameserver computer 13 may establish its exported resource table based on the responses from the server computers 12(m).
  • each of the various server computers 12(m) which maintains an exported resource may periodically broadcast information as to the exported resources which it maintains, and the nameserver computer 13 can update its exported resource table based on the broadcasts from the server computer.
  • the nameserver computer's exported resource table may be established by a system operator and may be fixed until he or she updates it.
  • the information provided by the nameserver computer 13 in response to a request initiated by the default stub would include such information as, for example, the identification of a computer 12(m) which can provide a class which implements the remote method to be invoked, particular information which the computer (that is, the computer which implements the remote method) will require to provide the required stub class code, and the like.
  • the computer 1 l(n) that is 14 processing the invoking Java program may, under control of the control module 19, use the info ⁇ nation communicate with the computer (that is, the computer which implements the remote method) to obtain the stub class, and may thereafter invoke the method as described above.
  • client computer l l(n), server computer 12(m) and, if necessary, nameserver 13 in connection with obtaining and dynamic loading of a stub class instance when a reference to a remote method is received will be described in connection with the flow chart depicted in FIG. 2.
  • client computer l l(n) and server computer in connection with remote invocation of a method using the stub class instance will be described in connection with the flow chart depicted in FIG. 3.
  • the execution environment control module 19 will, when it receives a reference to a remote method, will initially determine whether an appropriate stub class instance is present in the execution environment 20 to facilitate invocation of the remote method (step 100).
  • control module 19 determines that such a stub class instance 30 for the remote method is present in the execution environment, it may continue other operations (step 101). However, if the control module 19 determines in step 101 that such a stub class instance is not present in the execution environment 20 for the remote method, the control module 19 will use the stub class loader 33 to attempt to locate and load a stub class instance 30 for the class to process the remote method.
  • control module 19 will initially determine whether the invocation from the class instance 22 included a resource locator to identify the server computer 12(m) or other resource which maintains the class for the method to be invoked, or whether it (that is, the control module 19) or the stub class loader 33 otherwise are provided with such a resource locator (step 102). If the control module 19 makes a positive determination in that step, it will sequence to step 103 to enable the stub class loader 33 to initiate communications with identified server computer 12(m) to obtain stub class instance for the class and method to be invoked (step 103).
  • the stub class loader 33 When the stub class loader 33 receives the stub class instance 30 from the server computer 12(m), it will load the stub class instance 30 into execution environment 20 for the class instance 22 which initiated the remote method invocation call in step 100 (step 104). After the stub class instance 30 for the referenced remote method has been loaded in the execution environment, the method can be invoked as will be described below in connection with FIG. 3. 15
  • control module 19 determines that the invocation from the class instance 22 did not include a resource locator to identify the server computer 12(m) or other resource which maintains the class for the method to be invoked, and further that it (that is, the control module 19) or the stub class loader 33 is not otherwise provided with such a resource locator, a "class not found" exception may be indicated, at which point the control module 19 may call an exception handler.
  • the exception handler may perform any of a number of recovery operations, including, for example, merely notifying the control module 19 that the remote method could not be located and allow it to determine subsequent operations.
  • control module 19 may attempt to obtain a resource locator from the nameserver computer 13 or other resource provided by the network 14 (generally represented in FIG. 1 by the nameserver computer 13), using a call, for example, a default stub class instance 30.
  • the call to the default stub class instance 30 will include an identification of the class and method to be invoked and the name of the nameserver computer 13(m).
  • the control module 19 will enable the computer l l(n) to initiate communications with nameserver computer 13 to obtain an identifier for a server computer
  • the communications from the default stub class instance 30 will essentially correspond to a remote method invocation, with the method enabling the nameserver computer to provide the identification for the server computer 12(m), if one exists associated with the class and method to be remotely invoked, or alternatively to provide an indication that no server computer 12(m) is identified as being associated with the class and method.
  • the default stub class interface 30 will provide, as a parameter value, the identification of class and method to be invoked.
  • the nameserver computer 13 In response to the communications from the default stub class instance 30, the nameserver computer 13 will process the request as a remote method (step 111), with the result information comprising the identification for the server computer 12(m), if one exists that is associated with the class and method to be remotely invoked, or alternatively an indication that no server computer 12(m) is identified as being associated with the class and method. After finishing the method, the nameserver computer 13 will initiate communications with the default 16 stub class instance 30 to provide the result information to the default stub class instance 30 (step 112).
  • the stub class loader 33 determines whether the result information from the nameserver computer comprises the identification for the server computer 12(m) or an indication that no server computer 12(m) is identified as being associated with the class (step 114). If the stub class loader 33 determines that the result information comprises the identification for the server computer 12(m), it (that is, the stub class loader 33) will retum to step 101 to initiate communication with the identified server computer 12(m) to obtain stub class instance for the class and method that may be invoked.
  • the stub class loader 33 determines in step 114 that the nameserver computer 13 had provided an indication that no server computer 12(m) is identified as being associated with the class and method that may be invoked, the "class not found" exception may be indicated (step 115) and an exception handler called as described above.
  • the stub class instance 30 retrieved and loaded as described above in connection with FIG. 2 may be used in remote invocation of the method. Operations performed by the client computer l l(n) in connection with remote invocation of the method will be described in connection with the flow chart in FIG. 3.
  • the control module 19 may initially verify that a stub class instance 30 is present in the execution environment for remote method to be invoked (step 120). If a positive determination is made in step 120, the stub class instance 30 will be used for the remote invocation, and in the remote invocation will provide parameter values which are to be used in processing the remote method (step 121).
  • the stub class instance 30 for the remote method that may be invoked will be used to initiate communications with the server computer 12(m) which maintains the class for the remote method (step 122), in the process, the passing parameter values which are to be used in processing the remote method will be passed. It will be appreciated that, if the server computer 12(m) which is to process the method is the same physical computer as the client computer 12(n) which is invoking the method, the communications can be among execution environments which are being processed within the 17 physical computer.
  • the server computer 12(m) which is to process the method is a different physical computer from that of the client computer 12(n) which is invoking the method, the communications will be through the client computer's and server computer's respective network interfaces 15(n) and 16(m) and over the network 14.
  • the server computer 12(m) if necessary establishes an execution environment 24 for the class which maintains the method that may be invoked, and the uses the information provided by the skeleton 32 to create a class instance 26 for that class (step 123).
  • the server computer 12(m), under control of the control module 28, will process the method in connection with parameter values that were provided by stub class instance 30 (step 124).
  • the server computer 12(m), also under control of the control module 28, will initiate communications with the client computer's stub class instance 30 to provide result information to the stub class instance (step 125).
  • the communications can be among execution environments 24 and 20 which are being processed within the physical computer.
  • the server computer 12(m) which processed the method is a different physical computer from that of the client computer 12(n) which is invoking the method
  • the communications will be through the server computer's and client computer's respective network interfaces 16(m) and 15(n) and over the network 14.
  • the stub class instance 30 may provide result information to the class instance 22 which initiated the remote method invocation (step
  • control module 19 determines in that step that it does not have a stub class instance 30 that is appropriate for the remote method that may be invoked, it may at that point call an exception handler (step 127) to perform selected e ⁇ or recovery operations. 18
  • Methods and systems consistent with the present invention provide a number of advantages.
  • they provide a new system and method for facilitating dynamic loading of a stub which enables a program that is operating in one execution environment to remotely invoke processing of a method in another execution environment, so that the stub can be loaded by the program when it is ran and needed.
  • stubs are compiled with the program, and thus are statically determined when the program is compiled, they (the stubs) may implement subsets of the actual set of remote interfaces which are supported by the remote references that is received by the program, which can lead to errors and inefficiencies due to mismatches between the stub that is provided to a program and the requirements of the remote procedure that is called when the program is run.
  • the stub that is loaded can be obtained from the particular resource which provides the remote method, it (the stub) can define the exact set of interfaces to be provided to the invoking program at ran time, thereby obviating run-time incompatibilities which may result from mis-matches between the stub that is provided and the requirements of the remote method that is invoked.
  • stub class instances may instead be obtained and loaded when the remote methods are initially invoked.
  • Obtaining and loading of the stub class instance for a remote method when a reference thereto is received will have the advantages that (I) the stub class instance will be present in the execution environment when the remote method is actually invoked, and (ii) if the appropriate stub class instance can not be located, the program or an operator may be notified at an early time.
  • obtaining and loading of the stub class instance for a remote method when the method is to be invoked may result in a delay of the invocation until the correct stub class instance can be found, if the method is in fact not invoked even if a reference to it is received the stub class instance may not need to be located and loaded.
  • a system in accordance with the invention can be constructed in whole or in part from special purpose hardware or a general purpose computer system, or any 19 combination thereof, any portion of which may be controlled by a suitable program.
  • Any program may in whole or in part comprise part of or be stored on the system in a conventional manner, or it may in whole or in part be provided in to the system over a network or other mechanism for transferring information in a conventional manner.
  • the system may be operated and/or otherwise controlled by means of information provided by an operator using operator input elements (not shown) which may be connected directly to the system or which may transfer the information to the system over a network or other mechanism for transferring information in a conventional manner.
  • an altemative embodiment also consistent with the present invention loads and retrieves objects in a lookup service, where the objects contain code (stub information) for facilitating communication with a particular service or the objects contain code that performs the service.
  • the altemative embodiment is described below as downloading objects from the lookup service that represent stubs, the techniques described below are equally applicable to downloading objects that actually perform the services.
  • a lookup service defines a network's directory of services and stores references to these services. A user desiring use of a service on the network accesses the lookup service, which returns the stub information that facilitates the user's access of the service.
  • the lookup service may contain a subset of all services available in the network, referred to as a "Djinn" as described in copending U.S. Patent Application Serial No. , entitled
  • Djinn refers to a logical grouping of one or more of the services or resources that are provided by a network.
  • Devices connected to the network may either dynamically add themselves to the Djinn or dynamically remove themselves from the Djinn.
  • a device When added, a device provides zero or more of its services to the Djinn and may utilize all of the services crarently provided by the Djinn.
  • the services provided by the Djinn are defined by the lookup service, which provides a common way to both find and utilize the services for the Djinn. 20
  • the lookup service is a fundamental part of the infrastructure for a Djinn or other computer network offering a range of services. It is the primary means for programs to find services available within the Djinn and is the foundation for providing stubs through which users and administrators can discover and interact with services in the Djinn. Reference will now be made to FIG. 4, which depicts lookup service 400 in greater detail.
  • Server computer 12(m) also includes a lookup service 400, further described below.
  • the lookup service 400 located on a server 12(m) as shown in FIG. 1, maintains a collection of "service items" 410-418.
  • Each service item 410-418 represents an instance of a service available within the Djinn, and each service item 410 contains a service ID 402 that uniquely identifies the service item, a stub 404 providing code that programs use to access the service, and a collection of attributes 406 that describe the service.
  • the lookup service Upon registering a new service with the lookup service 400, the lookup service gives the new service item 410 a unique service ID 402, typically a number. This service ID 402 can later be used to access the specific service, avoiding unnecessary searching or locating several matching service items upon a query.
  • the service When a new service is created (e.g., when a new device is added to the Djinn), the service registers itself with the lookup service 400, providing a stub 404 to be used by a client to access the service and an initial collection of attributes 406 associated with the service.
  • a printer might include attributes indicating speed (in pages per minute), resolution (in dots per inch), color, and whether duplex printing is supported.
  • the lookup service administrator (not shown) might also add new attributes, such as the physical location of the service and common names for it.
  • the service can add an attribute that indicates what the problem is.
  • attributes are stored as multi-entries, and the addition, modification and deletion of attributes can be made using multi-templates and the techniques explained in co-pending U.S. Patent
  • An individual set of attributes is represented as an instance of a class, each attribute being a field of that class.
  • An example of an attribute set for a printer is: 21 public class Printer ⁇
  • Boolean duplex // supports two-sided printing Boolean color; // color or black-only
  • the class provides strong typing of both the set and the individual attributes.
  • the attributes 406 of service items 410 can also be represented as a set of sets of attributes. Attributes 406 of a service item 410 can contain multiple instances of the same class with different attribute values, as well as multiple instances of different classes. For example, the attributes 406 of a service item 410 might have multiple instances of a Name class, each giving the common name of the service in a different language, plus an instance of a Location class, a Type class, or various other service-specific classes. .An example of some added attributes to describe the printer may be the Name, Type, or Location: public class Name implements Entry ⁇
  • the attributes 406 for this service item 410 would be a set of attributes containing the Printer, Name, Type, and Location class instances, each class containing their own individual attributes.
  • the scheme used for attributes is not constrained by these examples. 22
  • Programs that need a particular type of service can use the lookup service 400 to find a stub that can be used to access the service.
  • a match can be made based on the type of service as well as the specific attributes attached to the service. For example, a client could search for a printer by requesting a stub type corresponding to the service desired or by requesting certain attributes such as a specific location or printing speed.
  • attributes are stored as multi-entries, and a match on attributes can be made using multi-templates, as explained in co-pending U.S. Patent Application No. , entitled "Method and System For Multi-Entry and Multi- Template Matching In A Database", previously incorporated herein.
  • the stub 404 corresponding to a service is registered in the lookup service 400 and is used by the client computer l l(n) to access the service methods remotely.
  • This stub 404 may also be a "smart proxy.”
  • a smart proxy code within which a stub is embedded, helps the client more efficiently implement the stub and the method to be remotely invoked.
  • a smart proxy often performs some local computation for efficiency before or after it actually calls the stub.
  • a smart proxy may contain code to cache information, so if a client requested it again, instead of going back to the server to get the information, it may have cached the answer and be able to return it quickly. If the situation called for it, a smart proxy might also transform the parameters received from the client into other types and then send the transformed types. The smart proxy concept is further explained in co-pending U.S.
  • FIG. 5 depicts a flowchart illustrating the steps used by systems consistent with the present invention for adding a service stub to the lookup service.
  • a device joins the network, it typically registers a service with the lookup service (step 500).
  • the device supplies a stub 404 to the lookup service, and it may also give its associated attributes 406 to the lookup service (step 502).
  • the lookup service 400 assigns a unique service ID 402, typically a number as previously stated, to the 23 service registered with the lookup service (step 504).
  • the device has supplied the lookup service 400 with a stub 404 and attributes 406, and the lookup service has assigned a unique service ID 402, the device has completed registration of the service with the lockup service (step 506). After services are registered with the lookup service 400, clients can use the lookup service to obtain stub information needed to access the registered services.
  • FIG. 6 depicts a flowchart illustrating steps used by systems and methods consistent with the present invention to download a service item from the lookup service.
  • the client computer 1 l(n) sends a request for a service to the server 12(m) with the lookup service 400 (step 600).
  • the request originates from the remote method invocation of a class instance 22 on client computer 11 (n), and the requested service may reside on a remote server as the exemplary service 38 resides on server 12(1).
  • the client computer 1 l(n) may request one or more services from the lookup service 400.
  • control 19 directs the stub class loader 33 to locate the co ⁇ esponding stub 404 from the server 12(m). To do so, the control 19 enables the stub class loader 33 to initiate communication with the server 12(m) to obtain a stub 404 for the service to be obtained.
  • control 28 in the server 12(m) Upon receipt of the request from the client computer 1 l(n), the control 28 in the server 12(m) searches the lookup service 400 for the stub 404 corresponding to the requested service (step 604). If there are no matches found, the control 28 returns a null value (steps 606 and 608).
  • the server 12(m) returns the stub to the stub class loader 33 on the client computer (step 612). If more than one stub was located matching the client' s request (step 610), in one embodiment consistent with the present invention, any one of the stubs is returned (step 616). In another implementation where the client requests more than one service, the server
  • the stub class loader 33 When the stub 404 is received by the stub class loader 33, the stub class loader loads it into the execution environment 20. After it is loaded, the service 38 can be remotely invoked. The use of the stub information to invoke remote processing of the service 38 is performed in the same manner as previously discussed in connection with FIG. 3. 24
  • the class instance 22 can use the stub 404 to access the service 38 on the server 12(1).
  • control 19 verifies that the stub 404 is present in the execution environment 20. If so, the class instance 22 can then use the stub 404 to initiate communications with the server 12(1) that maintains the service 38, and parameters will be passed to the service 38 for implementation.
  • This lookup service implementation is one application of the dynamic loading and retrieval of stub information to enable a program operating in one address space to invoke processing of a procedure in another address space.
  • This implementation of using the dynamic stub loading on the lookup service allows a client to receive stub information to facilitate use of that service directly.
  • the lookup service consistent with the present invention returns the code needed to access the service directly.
  • Using the dynamic loading of stub information in this way allows the client to receive all the code necessary to facilitate use of the service on a remote server.

Abstract

Methods and systems are provided that facilitate access to a service via a lookup service. A lookup service defines a network's directory of services and stores references to these services. A client desiring use of a service on the network accesses the lookup service, which returns the stub information that facilitates the user's access of the service. The client uses the stub information to access the service.

Description

1
METHOD AND SYSTEM FOR FACILITATING ACCESS TO A LOOKUP SERVICE
RELATED APPLICATIONS
This application is a Continuation-In-Part of U.S. Patent Application No. 08/636,706 filed on April 23, 1996, which is incorporated herein by reference. The following applications are relied upon and are hereby incorporated by reference in this application.
Provisional U.S. Patent Application No. 60/076,048, entitled "Distributed Computing System," filed on February 26, 1998.
U.S. Patent Application No. 09/044,923, entitled "Method and System for Leasing Storage," bearing attomey docket no. 06502.0011-01000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,838, entitled "Method, Apparatus, and Product for Leasing of Delegation Certificates in a Distributed System," bearing attomey docket no. 06502.0011-02000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,834, entitled "Method, Apparatus and Product for Leasing of Group Membership in a Distributed System," bearing attorney docket no.
06502.0011-03000, and filed on the same date herewith.
U.S. Patent Application No.09/044,916, entitled "Leasing for Failure Detection ' bearing attomey docket no. 06502.0011-04000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,933, entitled "Method for Transporting Behavior in Event Based System," bearing attomey docket no.06502.0054-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,919, entitled "Deferred Reconstruction of Objects and Remote Loading for Event Notification in a Distributed System," bearing attomey docket no. 06502.0062-01000, and filed on the same date herewith. U.S. Patent Application No. 09/044,938, entitled "Methods and Apparatus for Remote
Method Invocation," bearing attomey docket no. 06502.0102-00000, and filed on the same date herewith. 2
U.S. Patent Application No.09/045,652, entitled "Method and System for Deterministic Hashes to Identify Remote Methods," bearing attomey docket no. 06502.0103-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,790, entitled "Method and Apparatus for Determining Status of Remote Objects in a Distributed System," bearing attomey docket no.
06502.0104-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,930, entitled "Downloadable Smart Proxies for Performing Processing Associated with a Remote Procedure Call in a Distributed System," bearing attomey docket no. 06502.0105-00000, and filed on the same date herewith. U.S. Patent Application No. 09/044,917, entitled "Suspension and Continuation of
Remote Methods," bearing attomey docket no. 06502.0106-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,835, entitled "Method and System for Multi-Entry and Multi-Template Matching in a Database," bearing attomey docket no. 06502.0107-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,839, entitled "Method and System for In-Place Modifications in a Database," bearing attomey docket no. 06502.0108, and filed on the same date herewith.
U.S. Patent Application No. 09/044,945, entitled "Method and System for Typesafe Attribute Matching in a Database," bearing attomey docket no. 06502.0109-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,931, entitled "Dynamic Lookup Service in a Distributed System," bearing attomey docket no. 06502.0110-00000, and filed on the same date herewith. U.S. Patent Application No. 09/044,939, entitled "Apparatus and Method for Providing
Downloadable Code for Use in Communicating with a Device in a Distributed System," bearing attomey docket no. 06502.0112-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,932, entitled "Apparatus and Method for Dynamically Verifying Information in a Distributed System," bearing attomey docket no. 06502.0114-00000, and filed on the same date herewith. 3
U.S. Patent Application No. 09/030,840, entitled "Method and Apparatus for Dynamic Distributed Computing Over a Network," and filed on February 26, 1998.
U.S. Patent Application No. 09/044,936, entitled "An Interactive Design Tool for Persistent Shared Memory Spaces/' bearing attomey docket no. 06502.0116-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,934, entitled "Polymorphic Token-Based Control," bearing attomey docket no. 06502.0117-00000, and filed on the same date herewith.
U.S. Patent Application No.09/044,915, entitled "Stack-Based Access Control," bearing attomey docket no. 06502.0118-00000, and filed on the same date herewith. U.S. Patent Application No.09/044,944, entitled "Stack-Based Security Requirements," bearing attomey docket no. 06502.0119-00000, and filed on the same date herewith.
U.S. Patent Application No. 09/044,837, entitled "Per-Method Designation of Security Requirements," bearing attomey docket no. 06502.0120-00000, and filed on the same date herewith.
FIELD OF THE INVENTION The present invention relates generally to data processing systems and, more particularly, to a method and system for facilitating access to a lookup service.
BACKGROUND OF THE INVENTION In modem "enterprise" computing, a number of personal computers, workstations, and other devices such as mass storage subsystems, network printers and interfaces to the public telephony system, are typically interconnected in one or more computer networks. The personal computers and workstations are used by individual users to perform processing in connection with data and programs that may be stored in the network mass storage subsystems. In such an arrangement, the personal computers/workstations, operating as clients, typically download the data and programs from the network mass storage subsystems for processing. In addition, the personal computers or workstations will enable processed data to be uploaded to the network mass storage subsystems for storage, to a network printer for printing, to the telephony interface for transmission over the public telephony system, or the like. In such an arrangement, the network mass storage subsystems, network printers and telephony interface operate as servers, since they are available to service requests from all of the clients in the network. By organizing the network in such a manner, the servers are readily available for use by all of the personal computers/workstations in the network. Such a network may be spread over a fairly wide area, with the personal computers/workstations being interconnected by communication links such as electrical wires or optic fibers.
In addition to downloading information from servers for processing, a client, while processing a program, can remotely initiate processing by a server computer of particular routines and procedures (generally "procedures"), in connection with certain "parameter" information provided by the client. After the server has processed the procedure, it will provide results of its processing to the client, which the client may thereafter use in its processing operations. Typically in such "remote procedure calls" the program will make use of a local "stub" which, when called, transfers the request to the server which implements the particular procedure, receives the results and provides them to the program. Conventionally, the stub must be compiled with the program, in which case the information needed to call the remote procedure must be determined at compile time, rather than at the time the program is run. Since the stub available to the client's programs is static, it may be at best the closest that can be determined should be provided for the program when it (the program) is compiled. Accordingly, errors and inefficiencies can develop due to mismatches between the stub that is provided to a program and the requirements of the remote procedure that is called when the program is run. SUMMARY OF THE INVENTION
A new and improved system and method for facilitating the obtaining and dynamic loading of a stub is provided to enable a program operating in one address space to remotely invoke processing of a method or procedure in another address space, so that the stub can be loaded by the program when it is ran and needed, rather than being statically determined when the program is compiled. Indeed, the stub that is loaded can be obtained from the resource providing the remote method or procedure, and so it (the stub) can exactly define the invocation requirements of the remote method or procedure. Since the stub can be located and dynamically loaded while the program is being run, rather that being statically determined when the program is compiled, run-time errors and inefficiencies which may result from mis-matches between the stub that is provided and the requirements of the remote method or procedure that is invoked can 5 be minimized. In an altemative embodiment of the present invention, the stub is obtained from a lookup service to provide access to a service defined in the lookup service.
In brief summary, methods and systems consistent with an altemative embodiment of the present invention facilitate access to a service via a lookup service. A lookup service defines a network's directory of services and stores references to these services. A client desiring use of a service on the network accesses the lookup service, which returns the stub information that facilitates the user's access of the service. The client uses the stub information to access the service.
BRIEF DESCRIPTION OF THE DRAWINGS
This invention is pointed out with particularity in the appended claims. The above and further advantages of this invention may be better understood by referring to the following description taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a function block diagram of a computer network including an arrangement constructed in accordance with the present invention for facilitating the obtaining, dynamic loading and use of "stub" information to enable a program operating in one address space to invoke processing of a remote method or procedure in another address space;
FIGs. 2 and 3 are flow charts depicting the operations performed by the arrangement depicted in FIG. 1 , with FIG.2 depicting operations performed in connection with obtaining and dynamic loading of the stub information and FIG. 3 depicting operations performed in connection with use of the stub information to invoke processing of the remote method or procedure.
FIG. 4 is a diagram illustrating a lookup service consistent with the present invention; FIG. 5 is a flowchart illustrating a method of adding a stub to the lookup service consistent with the present invention.
FIG. 6 is a flowchart illustrating a method for retrieving a stub from a lookup service by systems consistent with the present invention.
DETAILED DESCRIPTION OF AN ILLUSTRATIVE EMBODIMENT FIG. 1 is a schematic diagram of a computer network 10 including an arrangement for facilitating dynamic loading of "stub" information to enable a program operating in one address space to remotely invoke processing of a method or procedure in another address space, where this method or procedure represents a network service. With reference to FIG. 1, computer network 10 includes a plurality of client computers 11(1) through 11 (N) (generally identified by reference numeral ll(n)), a plurality of server computers 12(1) through 12(M) (generally identified by reference numeral 12(m)), all of which are interconnected by a network represented by a communication link 14. In addition, the network 10 may include at least one nameserver computer 13, which may also be connected to communication link 14, whose purpose will be described below. As is conventional, at least some of the client computers 11 (n) are in the form of personal computers or computer workstations, each of which typically includes a system unit, a video display unit and operator input devices such as a keyboard and mouse (all of which are not separately shown). The server computers 12(m) and nameserver computer 13 also typically include a system unit (also not separately shown), and may also include a video display unit and operator input devices.
The client computers 11 (n), server computers 12(m) and nameserver computer 13 are all of the conventional stored-program computer architecture. A system unit generally includes processing, memory, mass storage devices such as disk and/or tape storage elements and other elements (not separately shown), including network interface devices 15(n), 16(m) for interfacing the respective computer to the communication link 14. The video display unit permits the computer to display processed data and processing status to the operator, and an operator input device enables the operator to input data and control processing by the computer. The computers 1 l(n) and 12(m) and 13 transfer information, in the form of messages, through their respective network interface devices 15(n), 16(m) among each other over the communication link 14.
In one embodiment, the network 10 is organized in a "client-server" configuration, in which one or more computers, shown in FIG. 1 as computers 12(m), operate as servers, and the other computers, shown in FIG. 1 as computers 1 l(n) operate as clients. In one aspect, one or more of the server computers 12(m) may, as "file servers," include large-capacity mass storage devices which can store copies of programs and data which are available for retrieval by the 7 client computers over the communication link 13 for use in their processing operations. From time to time, a client computer 1 l(n) may also store data on the server computer 12, which may be later retrieved by it (the client computer that stored the data) or other client computers for use in their processing operations. In addition, one or more of the server computers 12(m) may, as "compute servers," perform certain processing operations in response to a remote request therefor from a client computer 1 l(n), and retum the results of the processing to the requesting client computer 1 l(n) for use by them (that is, the requesting client computers 1 l(n)) in their subsequent processing. In either case, the server computers may be generally similar to the client computers 1 l(n), including a system unit, video display unit and operator input devices and may be usable by an operator for data processing operations in a manner similar to a client computer. Alternatively, at least some of the server computers may include only processing, memory, mass storage and network interface elements for receiving and processing retrieval, storage or remote processing requests from the client computers, and generating responses thereto. It will be appreciated a client computer 1 l(n) may also perform operations described herein as being performed by a server computer 12(m), and similarly a server computer 12(m) may also perform operations described herein as being performed by a client computer 1 l(n).
The network represented by communication link 14 may comprise any of a number of types of networks over which client computers 1 l(n), server computers 12(m) and nameserver computers 13 may communicate, including, for example, local area networks (LANs) and wide area networks (WANs) which are typically maintained within individual enterprises, the public telephony system, the Internet, and other networks, which may transfer digital data among the various computers. The network may be implemented using any of a number of communication media, including, for example, wires, optical fibers, radio links, and/or other media for carrying signals representing information among the various computers depicted in FIG. 1. As noted above, each of the computers typically includes a network interface which connects the respective computer to the communications link 14 and allows it to transmit and receive information thereover.
Systems consistent with the present invention facilitate the obtaining and dynamic loading of "stub" information to enable a program operating in one address space to invoke processing of a remote method or procedure in another address space, which may be located on 8 the same computer as the invoking program or on a different computer. Reference will be made to programs provided in the Java™ programming language, as described in James Gosling, Bill Joy, Guy Steele, "The Java™ Language Specification", Addison-Wesley, 1996, (hereinafter referred to as the "Java language specification"), incorporated herein by reference, which are processed in connection with an execution environment which is provided by a Java virtual machine. The Java virtual machine, in turn, is specified in the Lindholm andYellin, "The Java Virtual Machine Specification", Addison-Wesley, 1996, incorporated herein by reference. As described in the Java language specification, programs in the Java programming language define "classes" and "interfaces." Classes are used to define one or more methods or procedures, each of which may be invoked by reference to an interface. A class may be associated with and extend a "super-class," and in that regard will incorporate all of the interfaces and methods of the super-class, and may also include additional interfaces and/or methods. A class may also have one or more sub-classes (and thus will comprise a super-class of each of its sub-classes), with each sub-class incorporating and possibly extending their respective super-classes. An interface provides a mechanism by which a set of methods may be declared. In that connection, an interface identifies each method that is declared by the interface by, for example, a name, identities the data type(s) of argument(s) that are to be provided for the method, the data type(s) of retum values that are to be returned by the method, and identifiers for exceptions which can be thrown during processing of the method. A class may indicate that it implements a particular interface, and in that connection will include the program code which will be used in processing all of the methods which are declared in the interface. In addition, different classes may indicate that they implement the same interface, and each will have program code which will be used in processing all of the methods which are declared in the interface, but the program code provided in each class to for use in processing the methods may differ from the program code provided in the other classes which is used in processing the same methods; thus, an interface provides a mechanism by which a set of methods can be declared without providing an indication of the procedure which will be used in processing any of the methods. An interface may be declared independently of the particular class which implements the method or methods which can be invoked using the interface. In that regard, a class that invokes the method and a class that actually implements the method will not need to share a common super-class. During processing of a Java program, as described in the Java virtual machine specification, a client computer 1 l(n) provides an execution environment 20 for interpreting the Java program. The Java virtual machine includes a class loader 21 that, under control of a control module 19, can dynamically link instances of classes, generally identified in FIG. 1 by reference numeral 22, into the running program's execution environment while the program is being executed. In that operation, the control module 19 effectively enables the class loader to retrieve uninstantiated classes, which generally identified by reference numeral 23, instantiate them and link them as class instances 22 into the execution environment's address space at the Java program's run time as the methods which the respective classes 23 implement are called. In addition, the class loader 21 can discard ones of the class instances 22 when they are not needed or to conserve memory. It will be appreciated that, if a class instance 22 has been discarded, it may be reloaded by the class loader 21 at a later point if it is then needed.
Systems consistent with the present invention provide an arrangement which facilitates the remote invocation, by a program executing in an execution environment 20 by a client computer 11 (n), of methods implemented by classes on a server computer 12(m). In executing a method, the server computer 12(m) will also provide an execution environment 24 for processing, under control of a control module 28, the Java method. In that operation, the Java virtual machine which provides the execution environment 21 includes a class loader 25 (which may be similar to the class loader 21) that, under control of the control module 28, can dynamically link an instance of the class 26, to enable the method to be processed in the execution environment 24, and instances of other classes (also generally represented by reference numeral 26) which may be needed to process the remotely-invoked method. In that operation, the control module 28 effectively enables the class loader 25 to retrieve an uninstantiated class for the method to be invoked, from a plurality of uninstantiated classes which are generally identified by reference numeral 27, instantiate it (that is, the uninstantiated class which provides the method to be invoked) and link it as a class instance 26 into the execution environment. In addition, the class loader 25 can discard the class instances 26 when processing of the method has terminated. It will be appreciated that, if class instances 26 has been discarded, it may be reloaded by the class loader 25 at a later point if it is then needed. 10
The structure of nameserver computer 13, if provided, is generally similar to that of the server computer 12(m), and will not be separately described.
To facilitate remote invocation of a method, the control module 19 of the client computer's execution environment 21 makes use of one or more stub class instances generally identified by reference numeral 30 which are provided as part of the execution environment 21 in which the various class instances 22, including the class instance which is invoking the remote method, are being processed. Each stub class instance 30 is an instance of an uninstantiated stub class 31, which the ser er computer 12(m) may maintain for the various class instances 26 and uninstantiated classes 27 which the server computer 12(m) has "exported," that is, which the server computer 12(m) makes available to client computers 1 l(n) for use in remote invocation of methods provided thereby. An uninstantiated stub class 31 includes declarations for the complete set of interfaces for the particular remote uninstantiated class 27 which implements the remote method to be invoked, and also provides or invokes methods which facilitate accessing of the remote method(s) which are implemented by the remote class. The uninstantiated stub class 31, when it is instantiated and provided to the execution environment 20 of the client computer 1 l(n) as a stub class instance 30, effectively provides the information which is needed by the control module 19 of the execution environment 20 of the invoking Java program, so that, when a remote method that is implemented by its associated class is invoked by a Java program running in a particular execution environment, the remote method will be processed and the retum value(s) provided to the invoking Java program. In one embodiment, the aπangement by which the stub class instance may be provided to the execution environment 20 is similar to that described in the aforementioned Waldo, et al., patent application.
In addition, the server computer 12(m) provides a skeleton 32, which identifies the particular classes and methods which have been exported by the server computer 12(m) and information as to how it (that is, the server computer 12(m)) may load the respective classes and initiate processing of the particular methods provided thereby. Additionally, the server computer 12(m) contains a lookup service 400 for registering services on a network. The lookup service 400 will be discussed below.
When a class instance invokes a remote method maintained by a server computer 12(m), it will provide values for various parameters to the stub class instance 30 for the remote method, 11 which values the remote method will use in its processing. If the remote method is implemented on the same computer as the invoking Java program, when the invoking Java program invokes a remote method, the computer may establish an execution environment, similar to the execution environment 20, enable the execution environment's class loader to load and instantiate the class which implements the method as a class instance similar to class instances 22, and process the remote method using values of parameters which are provided by the invoking class instance in the remote invocation. After processing of the method has been completed, the execution environment in which the remote method has been processed will provide the results to the stub class instance 30 for the remote method that was invoked, which, in turn, will provide to the particular class instance 22 which invoked the remote method.
Similar operations will be performed if client computer 11 (n) and server computer 12(m) are implemented on different physical computers. In that case, in response to a remote invocation, the client computer l l(n) that is processing the invoking class instance 22, under control of the control module 19 for the execution environment 20 for the invoking class instance 22, will use the appropriate stub class instance 30 to communicate over the network represented by the communication link 14 with the server computer 12(m) which implements the remote method to enable it (that is, the server computer 12(m)) to establish an execution environment 24 for the class which implements the remote method, and to use the class loader 25 to load an instance of the class as a class instance 26. In addition, the client computer 1 l(n), also using the appropriate stub class instance 30, will provide any required parameter values to the server computer 12(m) over the network 14. Thereafter, the server computer 12(m) will process the remote method using parameter values so provided, to generate result value(s) which are transferred over the network to the client computer 11 (n), in particular to the appropriate stub class instance 30. The client computer 1 l(n) will, after it receives the result value(s) from the network, provide them to the invoking class instance 22 for its processing.
In any case, when the control module 19 of the client computer's execution environment 20 determines that a reference to the remote object has been received, if it determines that the stub class instance 30 is not present when it receives the reference, it will attempt to obtain the stub class instance 30 from, for example, the server computer 12(m) which implements the remote method, and enable the stub class instance 30 to be dynamically loaded in the execution 12 environment 20 for the invoking class instance 22. A reference to the remote object may be received, for example, either as a retum value of another remote method invocation or as a parameter that is received during another remote method invocation. The stub class instance may be dynamically loaded into the execution environment in a manner similar to that used to load class instances 22 in the execution environment 22. The execution environment 20 is provided with a stub class loader 33 which, under control of the control module 19, will attempt to find and load the stub class instances 30 as required by the class instances 22 processed in the execution environment. The location of a particular server computer 12(m) that maintains the class that implements a method to be invoked remotely may be included in the call from the invoking class instance or may be made known to the stub class loader 33 through another mechanism (not shown) maintained by the client computer 1 l(n).
However, if the stub class loader 33 is not otherwise notified of which server computer 12(m) maintains the class which implements a method which may be invoked remotely, it may use the nameserver computer 13 to provide that identification. The identification may comprise any identifier which may be used to identify a server computer 12(m) or other resource which is available on the network 14 and to which the server computer 12(m) can respond. Illustrative identifiers include, for example, a network address which identifies the server computer and/or resource, or, if the network 14 is or includes the Internet, an identifier to, for example, a World Wide Web resource which may provide the identification or a "uniform resource locator" ("URL") which provides a uniform mechanism for identifying resources that are available over the Internet. The server computer 12(m) which implements the remote method, in response to a request from the client computer 1 l(n) will provide stub class instance 30 which the client computer l l(n) may load into the execution environment 21 to thereafter enable the remote invocation to be initiated. As noted above, if the stub class loader 33 does not know which server computer 12(m) implements the remote method which may be invoked (and thus does not know which computer is to provide the stub class code for the remote invocation), it may, under control of the control module 19, obtain the identification from the nameserver computer 13. In that operation, the stub class loader 33 may use a previously-provided default stub class which is provided for use in such cases. The default class stub, when used by the invoking Java program, enables the 13 computer that is processing the invoking Java program to communicate with the nameserver computer 13 to obtain information which can be used in invoking the remote method. This operation is essentially the same as the invocation of a remote method to be processed by the nameserver computer 13 , with the remote method including a parameter identifying the class and method to be remotely invoked, and enabling the nameserver computer 13 to provide the identification of a server computer 12(m) which can process the method to the requesting client computer 1 l(n) and other information which may be helpful in communicating with the server computer 12(m) and invoking the particular method. It will be appreciated that the nameserver computer 13 will maintain a table (not separately shown) of "exported" resources, that is, resources, such as classes and methods, that are available to client computers 1 l(n) connected to the network 14, and information, such as the identifications of the particular server computers 12(m) which provide those resources, which will be useful to the client computers l l(n) in making use of the exported resources.
It will be appreciated that the nameserver computer 13 may create and maintain the exported resource table in a number of ways that are known in the art. For example, the nameserver computer 13 may periodically broadcast requests for exported resource information over the network 14, to which the various server computers 12(m) which maintain exported resources may respond; in that case, the nameserver computer 13 may establish its exported resource table based on the responses from the server computers 12(m). Alternatively, each of the various server computers 12(m) which maintains an exported resource may periodically broadcast information as to the exported resources which it maintains, and the nameserver computer 13 can update its exported resource table based on the broadcasts from the server computer. In addition, the nameserver computer's exported resource table may be established by a system operator and may be fixed until he or she updates it. In any case, the information provided by the nameserver computer 13 in response to a request initiated by the default stub would include such information as, for example, the identification of a computer 12(m) which can provide a class which implements the remote method to be invoked, particular information which the computer (that is, the computer which implements the remote method) will require to provide the required stub class code, and the like. After receiving the information from the nameserver computer 13, the computer 1 l(n) that is 14 processing the invoking Java program may, under control of the control module 19, use the infoπnation communicate with the computer (that is, the computer which implements the remote method) to obtain the stub class, and may thereafter invoke the method as described above.
With this background, the operations performed by client computer l l(n), server computer 12(m) and, if necessary, nameserver 13 in connection with obtaining and dynamic loading of a stub class instance when a reference to a remote method is received will be described in connection with the flow chart depicted in FIG. 2. In addition, operations performed by the client computer l l(n) and server computer in connection with remote invocation of a method using the stub class instance will be described in connection with the flow chart depicted in FIG. 3. With reference initially to FIG. 2, the execution environment control module 19 will, when it receives a reference to a remote method, will initially determine whether an appropriate stub class instance is present in the execution environment 20 to facilitate invocation of the remote method (step 100). If the control module 19 determines that such a stub class instance 30 for the remote method is present in the execution environment, it may continue other operations (step 101). However, if the control module 19 determines in step 101 that such a stub class instance is not present in the execution environment 20 for the remote method, the control module 19 will use the stub class loader 33 to attempt to locate and load a stub class instance 30 for the class to process the remote method. In that case, the control module 19 will initially determine whether the invocation from the class instance 22 included a resource locator to identify the server computer 12(m) or other resource which maintains the class for the method to be invoked, or whether it (that is, the control module 19) or the stub class loader 33 otherwise are provided with such a resource locator (step 102). If the control module 19 makes a positive determination in that step, it will sequence to step 103 to enable the stub class loader 33 to initiate communications with identified server computer 12(m) to obtain stub class instance for the class and method to be invoked (step 103). When the stub class loader 33 receives the stub class instance 30 from the server computer 12(m), it will load the stub class instance 30 into execution environment 20 for the class instance 22 which initiated the remote method invocation call in step 100 (step 104). After the stub class instance 30 for the referenced remote method has been loaded in the execution environment, the method can be invoked as will be described below in connection with FIG. 3. 15
Returning to step 102, if the control module 19 determines that the invocation from the class instance 22 did not include a resource locator to identify the server computer 12(m) or other resource which maintains the class for the method to be invoked, and further that it (that is, the control module 19) or the stub class loader 33 is not otherwise provided with such a resource locator, a "class not found" exception may be indicated, at which point the control module 19 may call an exception handler. The exception handler may perform any of a number of recovery operations, including, for example, merely notifying the control module 19 that the remote method could not be located and allow it to determine subsequent operations.
Alternatively, the control module 19 may attempt to obtain a resource locator from the nameserver computer 13 or other resource provided by the network 14 (generally represented in FIG. 1 by the nameserver computer 13), using a call, for example, a default stub class instance 30. The call to the default stub class instance 30 will include an identification of the class and method to be invoked and the name of the nameserver computer 13(m). Using the default stub class instance 30, the control module 19 will enable the computer l l(n) to initiate communications with nameserver computer 13 to obtain an identifier for a server computer
12(m) which maintains the class and method to be invoked (step 110). The communications from the default stub class instance 30 will essentially correspond to a remote method invocation, with the method enabling the nameserver computer to provide the identification for the server computer 12(m), if one exists associated with the class and method to be remotely invoked, or alternatively to provide an indication that no server computer 12(m) is identified as being associated with the class and method. During the communications in step 110, the default stub class interface 30 will provide, as a parameter value, the identification of class and method to be invoked.
In response to the communications from the default stub class instance 30, the nameserver computer 13 will process the request as a remote method (step 111), with the result information comprising the identification for the server computer 12(m), if one exists that is associated with the class and method to be remotely invoked, or alternatively an indication that no server computer 12(m) is identified as being associated with the class and method. After finishing the method, the nameserver computer 13 will initiate communications with the default 16 stub class instance 30 to provide the result information to the default stub class instance 30 (step 112).
After receipt of the result information from the nameserver computer 13 , the default stub class instance, under control of the control module 19, will pass result information to the stub class loader 33 (step 113). Thereafter, the stub class loader 33 determines whether the result information from the nameserver computer comprises the identification for the server computer 12(m) or an indication that no server computer 12(m) is identified as being associated with the class (step 114). If the stub class loader 33 determines that the result information comprises the identification for the server computer 12(m), it (that is, the stub class loader 33) will retum to step 101 to initiate communication with the identified server computer 12(m) to obtain stub class instance for the class and method that may be invoked. On the other hand, if the stub class loader 33 determines in step 114 that the nameserver computer 13 had provided an indication that no server computer 12(m) is identified as being associated with the class and method that may be invoked, the "class not found" exception may be indicated (step 115) and an exception handler called as described above.
As noted above, the stub class instance 30 retrieved and loaded as described above in connection with FIG. 2 may be used in remote invocation of the method. Operations performed by the client computer l l(n) in connection with remote invocation of the method will be described in connection with the flow chart in FIG. 3. As depicted in FIG. 3, when a class instance 22 invokes a method, the control module 19 may initially verify that a stub class instance 30 is present in the execution environment for remote method to be invoked (step 120). If a positive determination is made in step 120, the stub class instance 30 will be used for the remote invocation, and in the remote invocation will provide parameter values which are to be used in processing the remote method (step 121). Thereafter, the stub class instance 30 for the remote method that may be invoked will be used to initiate communications with the server computer 12(m) which maintains the class for the remote method (step 122), in the process, the passing parameter values which are to be used in processing the remote method will be passed. It will be appreciated that, if the server computer 12(m) which is to process the method is the same physical computer as the client computer 12(n) which is invoking the method, the communications can be among execution environments which are being processed within the 17 physical computer. On the other hand, if the server computer 12(m) which is to process the method is a different physical computer from that of the client computer 12(n) which is invoking the method, the communications will be through the client computer's and server computer's respective network interfaces 15(n) and 16(m) and over the network 14. In response to the communications from the stub class instance in step 122, the server computer 12(m), if necessary establishes an execution environment 24 for the class which maintains the method that may be invoked, and the uses the information provided by the skeleton 32 to create a class instance 26 for that class (step 123). Thereafter, the server computer 12(m), under control of the control module 28, will process the method in connection with parameter values that were provided by stub class instance 30 (step 124). After completing processing of the method, the server computer 12(m), also under control of the control module 28, will initiate communications with the client computer's stub class instance 30 to provide result information to the stub class instance (step 125). In a manner similar to that described above in connection with step 102, if the server computer 12(m) which processed the method is the same physical computer as the client computer 12(n) which invoked the method, the communications can be among execution environments 24 and 20 which are being processed within the physical computer. On the other hand, if the server computer 12(m) which processed the method is a different physical computer from that of the client computer 12(n) which is invoking the method, the communications will be through the server computer's and client computer's respective network interfaces 16(m) and 15(n) and over the network 14. After the stub class instance 30 receives the result information from the server computer, it may provide result information to the class instance 22 which initiated the remote method invocation (step
126), and that class instance 22 can continue processing under control of the control module 19.
Returning to step 120, if the control module 19 determines in that step that it does not have a stub class instance 30 that is appropriate for the remote method that may be invoked, it may at that point call an exception handler (step 127) to perform selected eπor recovery operations. 18
Methods and systems consistent with the present invention provide a number of advantages. In particular, they provide a new system and method for facilitating dynamic loading of a stub which enables a program that is operating in one execution environment to remotely invoke processing of a method in another execution environment, so that the stub can be loaded by the program when it is ran and needed. In systems in which stubs are compiled with the program, and thus are statically determined when the program is compiled, they (the stubs) may implement subsets of the actual set of remote interfaces which are supported by the remote references that is received by the program, which can lead to errors and inefficiencies due to mismatches between the stub that is provided to a program and the requirements of the remote procedure that is called when the program is run. However, since, in the dynamic stub loading system and method, the stub that is loaded can be obtained from the particular resource which provides the remote method, it (the stub) can define the exact set of interfaces to be provided to the invoking program at ran time, thereby obviating run-time incompatibilities which may result from mis-matches between the stub that is provided and the requirements of the remote method that is invoked.
It will be appreciated that a number of modifications may be made to the arrangement as described above. For example, although the execution environment 20 has been described as obtaining and loading stub class instances to facilitate invocation of remote methods when references to the remote methods are received, it will be appreciated that stub class instances may instead be obtained and loaded when the remote methods are initially invoked. Obtaining and loading of the stub class instance for a remote method when a reference thereto is received will have the advantages that (I) the stub class instance will be present in the execution environment when the remote method is actually invoked, and (ii) if the appropriate stub class instance can not be located, the program or an operator may be notified at an early time. On the other hand, obtaining and loading of the stub class instance for a remote method when the method is to be invoked may result in a delay of the invocation until the correct stub class instance can be found, if the method is in fact not invoked even if a reference to it is received the stub class instance may not need to be located and loaded.
It will be appreciated that a system in accordance with the invention can be constructed in whole or in part from special purpose hardware or a general purpose computer system, or any 19 combination thereof, any portion of which may be controlled by a suitable program. Any program may in whole or in part comprise part of or be stored on the system in a conventional manner, or it may in whole or in part be provided in to the system over a network or other mechanism for transferring information in a conventional manner. In addition, it will be appreciated that the system may be operated and/or otherwise controlled by means of information provided by an operator using operator input elements (not shown) which may be connected directly to the system or which may transfer the information to the system over a network or other mechanism for transferring information in a conventional manner.
Altemative Embodiment of the Present Invention
Although an embodiment consistent with the present invention has been previously described that dynamically retrieves and loads stubs, an altemative embodiment also consistent with the present invention loads and retrieves objects in a lookup service, where the objects contain code (stub information) for facilitating communication with a particular service or the objects contain code that performs the service. Although the altemative embodiment is described below as downloading objects from the lookup service that represent stubs, the techniques described below are equally applicable to downloading objects that actually perform the services. A lookup service defines a network's directory of services and stores references to these services. A user desiring use of a service on the network accesses the lookup service, which returns the stub information that facilitates the user's access of the service.
The lookup service may contain a subset of all services available in the network, referred to as a "Djinn" as described in copending U.S. Patent Application Serial No. , entitled
"Dynamical Lookup Service in a Distributed System," assigned to a common assignee, filed on even date herewith, which has been previously incorporated by reference. A "Djinn" refers to a logical grouping of one or more of the services or resources that are provided by a network.
Devices connected to the network may either dynamically add themselves to the Djinn or dynamically remove themselves from the Djinn. When added, a device provides zero or more of its services to the Djinn and may utilize all of the services crarently provided by the Djinn. The services provided by the Djinn are defined by the lookup service, which provides a common way to both find and utilize the services for the Djinn. 20
The lookup service is a fundamental part of the infrastructure for a Djinn or other computer network offering a range of services. It is the primary means for programs to find services available within the Djinn and is the foundation for providing stubs through which users and administrators can discover and interact with services in the Djinn. Reference will now be made to FIG. 4, which depicts lookup service 400 in greater detail. Server computer 12(m), also includes a lookup service 400, further described below. The lookup service 400, located on a server 12(m) as shown in FIG. 1, maintains a collection of "service items" 410-418. Each service item 410-418 represents an instance of a service available within the Djinn, and each service item 410 contains a service ID 402 that uniquely identifies the service item, a stub 404 providing code that programs use to access the service, and a collection of attributes 406 that describe the service.
Upon registering a new service with the lookup service 400, the lookup service gives the new service item 410 a unique service ID 402, typically a number. This service ID 402 can later be used to access the specific service, avoiding unnecessary searching or locating several matching service items upon a query.
When a new service is created (e.g., when a new device is added to the Djinn), the service registers itself with the lookup service 400, providing a stub 404 to be used by a client to access the service and an initial collection of attributes 406 associated with the service. For example, a printer might include attributes indicating speed (in pages per minute), resolution (in dots per inch), color, and whether duplex printing is supported. The lookup service administrator (not shown) might also add new attributes, such as the physical location of the service and common names for it. Additionally, if a service encounters some problem that needs administrative attention, such as a printer running out of toner, the service can add an attribute that indicates what the problem is. In one implementation consistent with the present invention, attributes are stored as multi-entries, and the addition, modification and deletion of attributes can be made using multi-templates and the techniques explained in co-pending U.S. Patent
Application No. , entitled "Method and System for In-Place Modifications In A
Database", previously incorporated herein.
An individual set of attributes is represented as an instance of a class, each attribute being a field of that class. An example of an attribute set for a printer is: 21 public class Printer {
Integer ppm; // pages per minute
Integer dpi; // resolution in dots per inch
Boolean duplex; // supports two-sided printing Boolean color; // color or black-only
}
The class provides strong typing of both the set and the individual attributes.
The attributes 406 of service items 410 can also be represented as a set of sets of attributes. Attributes 406 of a service item 410 can contain multiple instances of the same class with different attribute values, as well as multiple instances of different classes. For example, the attributes 406 of a service item 410 might have multiple instances of a Name class, each giving the common name of the service in a different language, plus an instance of a Location class, a Type class, or various other service-specific classes. .An example of some added attributes to describe the printer may be the Name, Type, or Location: public class Name implements Entry {
String name; // the user-friendly name of the service String description; // free-form description of the service String language; // language (e.g., English, French) used in the above } public class Type implements Entry {
String type; // the general type of service String vendor; // vendor of product that implements the service String model; // model number/name of product
String version; // version number of product
} public class Location implements Entry {
Integer floor; // what floor the service is on String building;// what building it's in
String room; // what room it's in }
In this example, the attributes 406 for this service item 410 would be a set of attributes containing the Printer, Name, Type, and Location class instances, each class containing their own individual attributes. However, it should be noted that the scheme used for attributes is not constrained by these examples. 22
Programs (including other services) that need a particular type of service can use the lookup service 400 to find a stub that can be used to access the service. A match can be made based on the type of service as well as the specific attributes attached to the service. For example, a client could search for a printer by requesting a stub type corresponding to the service desired or by requesting certain attributes such as a specific location or printing speed. In one implementation consistent with the present invention, attributes are stored as multi-entries, and a match on attributes can be made using multi-templates, as explained in co-pending U.S. Patent Application No. , entitled "Method and System For Multi-Entry and Multi- Template Matching In A Database", previously incorporated herein.
Accessing a Lookup Service Employing Dynamic Stub Loading and Retrieval
Referring back to FIG. 4, the stub 404 corresponding to a service is registered in the lookup service 400 and is used by the client computer l l(n) to access the service methods remotely. This stub 404 may also be a "smart proxy." A smart proxy, code within which a stub is embedded, helps the client more efficiently implement the stub and the method to be remotely invoked. A smart proxy often performs some local computation for efficiency before or after it actually calls the stub. For example, a smart proxy may contain code to cache information, so if a client requested it again, instead of going back to the server to get the information, it may have cached the answer and be able to return it quickly. If the situation called for it, a smart proxy might also transform the parameters received from the client into other types and then send the transformed types. The smart proxy concept is further explained in co-pending U.S.
Patent Application No. , entitled "Downloadable Smart Proxies for Performing
Processing Associated with a Remote Procedure Call in a Distributed System," assigned to a common assignee, filed on even date herewith, which is hereby incorporated by reference. FIG. 5 depicts a flowchart illustrating the steps used by systems consistent with the present invention for adding a service stub to the lookup service. When a device joins the network, it typically registers a service with the lookup service (step 500). Upon registration with the lookup service 400, the device supplies a stub 404 to the lookup service, and it may also give its associated attributes 406 to the lookup service (step 502). In response, the lookup service 400 assigns a unique service ID 402, typically a number as previously stated, to the 23 service registered with the lookup service (step 504). Once the device has supplied the lookup service 400 with a stub 404 and attributes 406, and the lookup service has assigned a unique service ID 402, the device has completed registration of the service with the lockup service (step 506). After services are registered with the lookup service 400, clients can use the lookup service to obtain stub information needed to access the registered services.
FIG. 6 depicts a flowchart illustrating steps used by systems and methods consistent with the present invention to download a service item from the lookup service. In one implementation, the client computer 1 l(n) sends a request for a service to the server 12(m) with the lookup service 400 (step 600). The request originates from the remote method invocation of a class instance 22 on client computer 11 (n), and the requested service may reside on a remote server as the exemplary service 38 resides on server 12(1). In one implementation consistent with the present invention, the client computer 1 l(n) may request one or more services from the lookup service 400. The client's request comes in the form of a specific service ID 402, a type of stub 404, or a set of attributes 406, or any combination thereof (step 602). In response to the request, control 19 directs the stub class loader 33 to locate the coπesponding stub 404 from the server 12(m). To do so, the control 19 enables the stub class loader 33 to initiate communication with the server 12(m) to obtain a stub 404 for the service to be obtained.
Upon receipt of the request from the client computer 1 l(n), the control 28 in the server 12(m) searches the lookup service 400 for the stub 404 corresponding to the requested service (step 604). If there are no matches found, the control 28 returns a null value (steps 606 and 608).
Otherwise, if it locates the stub 404 corresponding to the service that the client computer 1 l(n) is attempting to access, the server 12(m) returns the stub to the stub class loader 33 on the client computer (step 612). If more than one stub was located matching the client' s request (step 610), in one embodiment consistent with the present invention, any one of the stubs is returned (step 616). In another implementation where the client requests more than one service, the server
12(m) returns the requested number of the stubs with their attributes (steps 614 and 618).
When the stub 404 is received by the stub class loader 33, the stub class loader loads it into the execution environment 20. After it is loaded, the service 38 can be remotely invoked. The use of the stub information to invoke remote processing of the service 38 is performed in the same manner as previously discussed in connection with FIG. 3. 24
Generally, the class instance 22 can use the stub 404 to access the service 38 on the server 12(1). When the class instance 22 requires use of the service 38 corresponding to the retumed stub 404, control 19 verifies that the stub 404 is present in the execution environment 20. If so, the class instance 22 can then use the stub 404 to initiate communications with the server 12(1) that maintains the service 38, and parameters will be passed to the service 38 for implementation.
This lookup service implementation is one application of the dynamic loading and retrieval of stub information to enable a program operating in one address space to invoke processing of a procedure in another address space. This implementation of using the dynamic stub loading on the lookup service allows a client to receive stub information to facilitate use of that service directly. Unlike previous lookup services, the lookup service consistent with the present invention returns the code needed to access the service directly. Using the dynamic loading of stub information in this way allows the client to receive all the code necessary to facilitate use of the service on a remote server. The foregoing description has been limited to a specific embodiment of the present invention. It will be apparent, however, that various variations and modifications may be made. It is the object of the appended claims to cover these and such other variations and modifications as come within the true spirit and scope of the invention.

Claims

25What is claimed is:
1. A method in a data processing system for accessing network services associated with a lookup service, comprising the steps of: receiving a request by the lookup service for access to one of the network services; and returning code to facilitate the access of the one network service.
2. The method of claim 1, further including the step of: accessing the network service using the retumed code.
3. The method of claim 1 , wherein the step of returning the code includes the step of: returning a stub.
4. The method of claim 1 , wherein the step of returning the code includes the step of: returning a smart proxy.
5. The method of claim 1, wherein the receiving step includes the step of: receiving the request to access the one network service such that the one network service is identified by a type of service.
6. The method of claim 1, wherein the receiving step includes the step of: receiving the request to access the one network service such that the one network service is identified by a service identification number.
7. The method of claim 1, wherein the receiving step includes the step of: receiving the request to access the one network service such that the one network service is identified by an attribute of the one network service.
8. The method of claim 1, wherein the step of returning code includes the step of: 26 retuming a plurality of stubs associated with a plurality of network services that match the request.
9. The method of claim 8 , wherein the step of retuming a plurality of stubs includes the step of: retuming attributes associated with each stub.
27
10. A method in a data processing system having a lookup service with a plurality of services, the method comprising the steps of: sending to the lookup service a request to access one of the services; and receiving code, responsive to the request, that facilitates use of the one service.
11. The method of claim 10, further including the step of: accessing the one service using the received code.
12. The method of claim 10, wherein the step of receiving code includes the step of: receiving a stub that facilitates access of the one service.
13. The method of claim 10, wherein the step of receiving code includes the step of: receiving a smart proxy that facilitates access of the one service.
28
14. A method in a data processing system having a first computer with a client and a second computer with a lookup service containing service stubs used for accessing associated services, the method comprising the steps of: sending a request by the client to the lookup service identifying one of the associated services to be accessed; receiving the request by the lookup service; searching the lookup service for the identified service; retuming the service stub associated with the identified service to the client; receiving the service stub by the client; loading the service stub into an address space of the client to render the service stub available for use to invoke the identified service; and using the stub by the client to access the identified service.
15. The method of claim 14, wherein the sending step includes the step of: sending a request for more than one service to be retumed, and wherein the retuming step further includes the step of: retuming more than one stub to the client in response to the request.
16. The method of claim 14, wherein the searching step includes the step of: retuming a null value if no service is found matching the request.
29
17. A distributed system with a plurality of network services, comprising: a server computer with a lookup service having stubs for facilitating access to the network services; and a client computer with a program that sends a request to the lookup service for one of the stubs corresponding to one of the network services, that receives the stub, and that accesses the one network service using the received stub.
18. The distributed system of claim 17, wherein the client computer utilizes a remote procedure call mechanism to receive the stub.
30
19. A system having a first computer with a client and a second computer with a lookup service containing service stubs used for accessing associated services, the system comprising: means for sending a request by the client to the lookup service identifying one of the associated services to be accessed; means for receiving the request by the lookup service; means for searching the lookup service for the identified service; means for retuming the service stub associated with the identified service to the client; means for receiving the service stub by the client; means for loading the service stub into an address space of the client to render the service stub available for use to invoke the identified service; and means for using the stub by the client to access the identified service.
31
20. A computer-readable medium containing instructions for controlling a data processing system to perform a method for accessing network services contained in a lookup service, the method comprising the steps of: receiving a request by the lookup service for access to one of the network services; and retuming code to facilitate the access of the one network service.
21. The computer-readable medium of claim 20, wherein the method further includes the step of: accessing the network service using the retumed code.
22. The computer-readable medium of claim 20, wherein the returning step includes the step of: retuming a stub.
23. The computer-readable medium of claim 20, wherein the retuming step includes the step of: returning a smart proxy.
32
24. A computer-readable medium containing instructions for controlling a data processing system to perform a method for accessing network services contained in a lookup service, the method comprising the steps of: sending to the lookup service a request to access one of the services; and receiving code, responsive to the request, that facilitates use of the one service.
25. The computer-readable medium of claim 24, wherein the method further includes the step of: accessing the one service using the received code.
26. The computer-readable medium of claim 24, wherein the step of receiving code includes the step of: receiving a stub that facilitates access of the one service.
27. The computer-readable medium of claim 24, wherein the step of receiving code includes the step of: receiving a smart proxy that facilitates access of the one service.
PCT/US1999/004066 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service WO1999044134A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020007009230A KR20010034515A (en) 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service
JP2000533818A JP2002505474A (en) 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service
EP99908444A EP1058884A1 (en) 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service
AU27878/99A AU2787899A (en) 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US7604898P 1998-02-26 1998-02-26
US60/076,048 1998-02-26
US09/044,826 1998-03-20
US09/044,826 US6832223B1 (en) 1996-04-23 1998-03-20 Method and system for facilitating access to a lookup service

Publications (1)

Publication Number Publication Date
WO1999044134A1 true WO1999044134A1 (en) 1999-09-02

Family

ID=26722037

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/004066 WO1999044134A1 (en) 1998-02-26 1999-02-25 Method and system for facilitating access to a lookup service

Country Status (5)

Country Link
EP (1) EP1058884A1 (en)
JP (1) JP2002505474A (en)
CN (1) CN1298506A (en)
AU (1) AU2787899A (en)
WO (1) WO1999044134A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001015392A1 (en) * 1999-08-24 2001-03-01 Telefonaktiebolaget Lm Ericsson (Publ) System and method of creating subscriber services in an ip-based telecommunications network
EP1202174A2 (en) * 2000-10-23 2002-05-02 Nec Corporation Stub search loading system and method, server apparatus, client apparatus, and computer-readable recording medium
WO2003005181A1 (en) * 2001-07-05 2003-01-16 Anoto Ab A computer readable medium storing instructiuons for managing communication in a system
WO2003023614A2 (en) * 2001-09-12 2003-03-20 Hewlett-Packard Company System and method to automatically obtain a service
US6757729B1 (en) * 1996-10-07 2004-06-29 International Business Machines Corporation Virtual environment manager for network computers
US7002559B2 (en) 2000-11-13 2006-02-21 Anoto Ab Method, system and product for information management
EP1875373A1 (en) * 2005-04-18 2008-01-09 Research In Motion Limited System and method of data source detection

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7080089B2 (en) * 2003-03-12 2006-07-18 Microsoft Corporation Customization of process logic in a software system
US7644376B2 (en) * 2003-10-23 2010-01-05 Microsoft Corporation Flexible architecture for notifying applications of state changes
JP2008287357A (en) * 2007-05-15 2008-11-27 Toshiba Corp Service providing system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5557798A (en) * 1989-07-27 1996-09-17 Tibco, Inc. Apparatus and method for providing decoupling of data exchange details for providing high performance communication between software processes
EP0803811A2 (en) * 1996-04-23 1997-10-29 Sun Microsystems, Inc. System and method for stub retrieval and loading
EP0817025A2 (en) * 1996-06-26 1998-01-07 Sun Microsystems, Inc. Data structures and method for implementing subcontracts in a distributed object oriented system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5557798A (en) * 1989-07-27 1996-09-17 Tibco, Inc. Apparatus and method for providing decoupling of data exchange details for providing high performance communication between software processes
EP0803811A2 (en) * 1996-04-23 1997-10-29 Sun Microsystems, Inc. System and method for stub retrieval and loading
EP0817025A2 (en) * 1996-06-26 1998-01-07 Sun Microsystems, Inc. Data structures and method for implementing subcontracts in a distributed object oriented system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
DAVE A ET AL: "PROXIES, APPLICATION INTERFACE, AND DISTRIBUTED SYSTEMS", PROCEEDINGS INTERNATIONAL WORKSHOP ON OBJECT ORIENTATION IN OPERATING SYSTEMS, 24 September 1992 (1992-09-24), pages 212 - 220, XP002009478 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6757729B1 (en) * 1996-10-07 2004-06-29 International Business Machines Corporation Virtual environment manager for network computers
WO2001015392A1 (en) * 1999-08-24 2001-03-01 Telefonaktiebolaget Lm Ericsson (Publ) System and method of creating subscriber services in an ip-based telecommunications network
US6839897B2 (en) 2000-10-23 2005-01-04 Nec Corporation Stub search loading system and method, server apparatus, client apparatus, and computer-readable recording medium
EP1202174A3 (en) * 2000-10-23 2004-07-21 Nec Corporation Stub search loading system and method, server apparatus, client apparatus, and computer-readable recording medium
EP1202174A2 (en) * 2000-10-23 2002-05-02 Nec Corporation Stub search loading system and method, server apparatus, client apparatus, and computer-readable recording medium
US7002559B2 (en) 2000-11-13 2006-02-21 Anoto Ab Method, system and product for information management
WO2003005181A1 (en) * 2001-07-05 2003-01-16 Anoto Ab A computer readable medium storing instructiuons for managing communication in a system
US7089308B2 (en) 2001-07-05 2006-08-08 Anoto Ab C/O Technologies Ab Communication protocol
WO2003023614A2 (en) * 2001-09-12 2003-03-20 Hewlett-Packard Company System and method to automatically obtain a service
WO2003023614A3 (en) * 2001-09-12 2003-11-27 Hewlett Packard Co System and method to automatically obtain a service
GB2395397A (en) * 2001-09-12 2004-05-19 Hewlett Packard Co System and method of automatically obtain a service
GB2395397B (en) * 2001-09-12 2004-11-10 Hewlett Packard Co System and method to automatically obtain a service
US6907444B2 (en) 2001-09-12 2005-06-14 Hewlett-Packard Development Company, L.P. System and method to automatically obtain a service
EP1875373A1 (en) * 2005-04-18 2008-01-09 Research In Motion Limited System and method of data source detection
EP1875373A4 (en) * 2005-04-18 2008-10-22 Research In Motion Ltd System and method of data source detection

Also Published As

Publication number Publication date
EP1058884A1 (en) 2000-12-13
CN1298506A (en) 2001-06-06
JP2002505474A (en) 2002-02-19
AU2787899A (en) 1999-09-15

Similar Documents

Publication Publication Date Title
US6708171B1 (en) Network proxy
US6654793B1 (en) System and method for facilitating dynamic loading of stub information to enable a program operating in one address space to invoke processing of a remote method or procedure in another address space
US6832223B1 (en) Method and system for facilitating access to a lookup service
US9183066B2 (en) Downloadable smart proxies for performing processing associated with a remote procedure call in a distributed system
US7734747B2 (en) Dynamic lookup service in a distributed system
US6134603A (en) Method and system for deterministic hashes to identify remote methods
US6983285B2 (en) Apparatus and method for dynamically verifying information in a distributed system
US6604127B2 (en) Dynamic lookup service in distributed system
US20020046228A1 (en) Method and system for facilitating access to a lookup service
US20010002473A1 (en) Dynamic lookup service in a distributed system
JP2002505464A (en) Apparatus and method for providing downloadable code for use in communicating with devices in a distributed system
US6629154B1 (en) Method and system for deterministic hashes to identify remote methods
US6438614B2 (en) Polymorphic token based control
EP1058884A1 (en) Method and system for facilitating access to a lookup service
US7089263B2 (en) Apparatus and method for dynamically verifying information in a distributed system
WO1999044123A1 (en) Downloadable smart proxies for performing processing associated with a remote procedure call in a distributed system
KR20010034515A (en) Method and system for facilitating access to a lookup service
WO2000077619A2 (en) Dynamic lookup service in a distributed system

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 99805369.4

Country of ref document: CN

AK Designated states

Kind code of ref document: A1

Designated state(s): AL AM AT AU AZ BA BB BG BR BY CA CH CN CU CZ DE DK EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT UA UG UZ VN YU ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW SD SL SZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
WWE Wipo information: entry into national phase

Ref document number: 1020007009230

Country of ref document: KR

ENP Entry into the national phase

Ref document number: 2000 533818

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 1999908444

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1999908444

Country of ref document: EP

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

WWP Wipo information: published in national office

Ref document number: 1020007009230

Country of ref document: KR

WWW Wipo information: withdrawn in national office

Ref document number: 1020007009230

Country of ref document: KR

WWW Wipo information: withdrawn in national office

Ref document number: 1999908444

Country of ref document: EP