System and method for locating resources in a distributed network

• US 6,058,423 A
• Filed: 12/23/1996
• Issued: 05/02/2000
• Est. Priority Date: 12/23/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Factor, Michael Edward
• Primary Examiner(s): Lee, Thomas C.
• Assistant Examiner(s): NGUYEN, DZUNG C
• Application Number:
  US08/772,705
• Time in Patent Office: 1,226 Days
• Field of Search: 395/800.28, 395/800.32, 395/200.56, 395/200.53, 395/200.54, 707/10, 707/3, 712/28, 712/31, 709/104, 709/106, 709/201, 709/223, 709/226, 709/228, 709/303
• US Class Current: 709/226
• CPC Class Codes: G06F 2209/462 Lookup
• CPC Class Codes: G06F 9/465 Distributed object oriented systems remote method invocation [RMI] G06F9/548
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]

Abstract

Disclosed is a system and method for implementing a distributed network system, such as the World Wide Web, in which distributed location services are utilized and include a collection of server processes that map from resource identifiers to a resource'"'"'s controlling server. In addition, distributed location services provide an architecture for assigning resource identifiers and set of protocols for accessing server processes. The server processes are logically defined to provide a more flexible system. Each of these logical server processes are an abstraction which provides the external view of the server. The logical server processes may be implemented by different numbers of physical processes running on different machines at different points in time.

Claims

1. A distributed network that includes a system for locating resources, said network comprising:

a plurality of logical server processes;

a client computer, said client computer having one of said logical server process associated therewith;

a plurality of resource computers wherein each of said resource computers has one of said logical server processes associated therewith;

a plurality of resources wherein each resource is controlled by at most one of said resource computers; and

a plurality of resource identifiers wherein each resource identifier identifies one of said resources in a one-to-one fashion, and each of said resource identifiers includes a logical address of the logical server process associated with the resource computer on which the resource was born.

2. The distributed network of claim 1 wherein each of said logical server processes includes a database of all resources that are identified as being born on a resource computer associated with that particular logical server process.

3. The distributed network of claim 2 wherein for each resource in said database, a current logical server process address is provided.

4. The distributed network of claim 2 wherein for each resource in said database, an owning resource may be provided.

5. The distributed network of claim 2 wherein for each resource in said database, an owned resource may be provided.

6. The distributed network of claim 1 wherein each of said logical server processes includes a database of all resource computers and their controlled resources that are associated with that particular logical server process.

7. The logical server processes of claim 6 wherein for each of said controlled resources, said database further includes an address for the logical server process associated with the resource computer on which the resource was born.

8. The logical server processes of claim 6 wherein for each of said controlled resources, said database further includes information regarding owning and owned resources.

9. A program product, said program product comprising:

a recordable media; and

a resource management system recorded on said recordable media, said resource management system for use on a distributed network having at least one client, a plurality of resource computers, and a plurality of resources controlled by said resource computers, said resource management system including a mechanism for creating resource identifiers that uniquely identifying each of said plurality of resources wherein each of said resource identifiers comprises;

an LLSP (logical location server process) field, said LLSP field for identifying a birth LLSP associated with said resource;

a birth computer field, said birth computer field for identifying a birth computer associated with said resource; and

an ID (identification) field, said ID field for identifying said resource among a set of all resources among all of said resources born on said birth computer.

10. The program product of claim 9 wherein each of said resource identifiers further includes a type field, said type field for interpreting said LLSP field, said birth computer field, and said ID field.

11. The program product of claim 9 wherein said LLSP field and said birth computer field together create a logical identifier for said birth computer.

12. The program product of claim 9 wherein said birth computer field and said ID field together create an identifier for uniquely identifying each resource that shares a common birth LLSP.

13. The program product claim 9 wherein each of said resource computers are assigned a logical identifier that includes an identity of said resource computer'"'"'s birth LLSP and a value that uniquely identifies said resource computer among the set of all resource computers with the same birth LLSP.

14. A distributed network comprising:

at least one client;

a plurality of resource computers;

a plurality of resources controlled by said resource computers;

a plurality of resource identifiers wherein each resource identifier uniquely identifies one of said resources; and

a distributed location service system, said distributed location service system including;

(a) a plurality of distributed server processes wherein said distributed server processes include a distributed directory of all resources and resource computers;

(b) a resource identifier creation mechanism, said resource identifier creation mechanism for creating and assigning resource identifiers each time a new resource is created;

(c) an update mechanism for updating relevant logical location server processes when one of said resources migrates; and

(d) a location mechanism that examines said distributed directory to locate resource computers from resource identifiers.

15. The distributed network of claim 14 wherein each of said plurality of distributed server processes includes a set of logical and set of physical addresses, and wherein said distributed location service system maintains the mapping between the two address types.

16. The distributed network of claim 14 wherein each of said distributed server processes includes a database of all resources born under that distributed server process and a database of all resources currently controlled within that distributed process.

17. The distributed network of claim 16 wherein said update mechanism includes:

a first mechanism for amending the database in the distributed server process associated with the birth computer of the migrating resource to indicate a new controlling computer;

a second mechanism for amending the database in the distributed server process that was formerly responsible for the migrating resource to indicate that the migrating resource has moved; and

a third mechanism for amending the database in the distributed server process that is presently responsible for the migrating resource.

18. The distributed network of claim 16 where said location mechanism includes:

a first mechanism to detect if the resource'"'"'s birth resource computer presently controls the resource;

a second mechanism to detect if the distributed server process associated with the client knows which resource computer presently controls the resource; and

a third mechanism to request from the distributed server process associated with the resource'"'"'s birth resource computer what resource computer presently controls the resource.

19. The distributed network of claim 18 where said location mechanism further includes a fourth mechanism for caching results returned from said third mechanism.

20. An internet having a uniform resource identification system, said network comprising:

a plurality of logical location service processes;

a plurality of clients wherein each client has one of said logical location service process associated therewith;

a plurality of resource computers wherein each of said resource computers has one of said logical location service processes associated therewith;

a plurality of resources wherein each resource is controlled by one of said resource computers; and

a plurality of resource identifiers wherein each resource identifier uniquely identifies one of said resources, and each of said resource identifiers includes an identity of the resource computer on which the resource was born and the logical location service process associated with the resource computer on which the resource was born.

21. The internet of claim 20 wherein each of said resource identifiers are included as part of a uniform resource name.

22. The internet of claim 20 wherein each of said logical location service processes includes a database that lists all resources that identify this logical location service process as a birth logical location service process.

23. The internet of claim 20 wherein each of said logical location service processes includes a database that lists all resource computers that identify this logical location service process as a current logical location service process.

24. The internet of claim 20 where said internet is used as a world wide web.

25. A method of finding a resource on a world wide web network from a client wherein said resource has a unique resource identifier associated therewith and the client knows the resource identifier, and wherein each resource, computer, and client associated with the network has a unique current logical location service process associated therewith, and wherein said resource identifier includes information regarding a birth computer and birth logical location service process of said resource, said method comprising the steps of:

sending a message to the birth computer of said resource to determine if said resource is controlled by said birth computer;

checking a response from said birth computer;

for the case where said resource is not controlled by said birth computer, sending a message to said logical location service process associated with the client to determine if it knows who controls said resource;

checking a response from said logical location service process associated with the client;

for the case where said logical location service process associated with the client does know the identity of the computer controlling said resource, sending a message to said birth logical location service process of said resource to determine who controls said resource.

26. The method of claim 25 further comprising the step of caching said response from said logical location service process associated with the client when said logical location service process associated with the client knows the location of said resource.