Methods and apparatus for storage and retrieval of name space information in a distributed computing system

• US 5,724,512 A
• Filed: 04/17/1995
• Issued: 03/03/1998
• Est. Priority Date: 04/17/1995
• Status: Expired due to Term

• Current Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Winterbottom, Philip Steven
• Primary Examiner(s): Kulik, Paul V.
• Application Number:
  US08/424,137
• Time in Patent Office: 1,051 Days
• Field of Search: 395/200.12, 395/200.14, 395/610, 395/612, 395/616, 395/617, 395/680, 395/684, 395/200.09, 395/200.05
• US Class Current: 709/226
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: G06F 2209/463 Naming
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A distributed computing environment is disclosed which allows a user at one location to access resources at other locations. Each resource in the distributed computing environment is represented as a hierarchical file system. A user or process has a name space comprised of at least one hierarchical file system provided by a connected resource. The distributed computing environment allows a first processor to invoke execution of a processing task by a remote processor. The first processor transmits a representation of its current name space to the remote processor. The remote processor will execute the processing task on a name space modified in accordance with the name space representation received from the first processor. The transmitted representation of the name space associated with the first processor includes at least one dynamic name space modification command, such as those executed by a user after logging into the distributed computing system. A plurality of data structures are provided for storing path information which allows the pathname of a given channel, which represents a file, to be determined. The stored path information allows the hierarchical file tree of a connected file system to be generated.

Claims

1. A method performed by a first processor in a distributed computing system for invoking execution of a processing task on a remote processor, said processing task operating on a first name space associated with said first processor, said first name space being formed by a plurality of name space modification commands, said method comprising the steps of:

establishing a communication link between said first processor and said remote processor;

identifying the plurality of name space modification commands that form said first name space, said plurality of identified name space modification commands including at least one dynamic name space modification command; and

transmitting said plurality of name space modification commands to said remote processor for execution of said processing task by said remote processor on a name space modified in accordance with said transmitted plurality of name space modification commands.

2. The method according to claim 1 wherein a transmitted name space modification command includes an identification of the communication link to be utilized by said remote processor to access a mounted resource if said name space modification command includes the mounting of a file system of a resource in an indicated portion of said first name space.

3. The method according to claim 2, wherein said identified communication link is an interprocess communication pipe utilized by said first processor to communicate with said resource, if said mounted file system is provided by a resource on the same network node as said first processor.

4. The method according to claim 2, wherein said identified communication link indicates that said remote processor should establish a network connection to said resource, if said mounted file system is provided by a resource on a different network node than said first processor.

5. The method according to claim 1, wherein each of said name space modification commands are stored in an entry of a mount table, each of said entries in said mount table including a value that indicates the order in which said entry was created, and wherein said step of identifying the plurality of name space modification commands that form said first name space retrieves said name space modification commands from said mount table entries in the order in which they were executed.

6. The method according to claim 1, further including the step of cooperating with said remote processor to process file operation requests received by said first processor from said remote processor for files resident on the same network node as said first processor.

7. A method performed by a remote processor in a distributed computing system for executing a processing task received from a first processor, said processing task operating on a first name space associated with said first processor, said first name space being formed by a plurality of name space modification commands, said remote processor having a second name space, said method comprising the steps of:

processing a connection request from said first processor;

allocating a process on said remote processor to execute said invoked processing task;

receiving from said first processor the plurality of name space modification commands that form said first name space, said plurality of received name space modification commands including at least one dynamic name space modification command;

establishing a communication link to said first processor;

mounting said communication link to said first processor in a predefined place in said second name space;

executing said plurality of received name space modification commands to modify said second name space; and

executing said invoked processing task on said modified second name space.

8. The method according to claim 7, wherein a received name space modification command includes an identification of the communication link to be utilized by said remote processor to access a particular mounted resource if the name space modification command includes the mounting of a file system in an indicated portion of said name space.

9. The method according to claim 8, wherein said identified communication link is an interprocess communication pipe utilized by said first processor to communicate with said resource, if said mounted file system is provided by a resource on the same network node as said first processor.

10. The method according to claim 8, wherein said identified communication link indicates that said remote processor should establish a network connection to said mounted resource, if said mounted file system is provided by a resource on a different network node than said first processor.

11. The method according to claim 7, further including the step transmitting file operation requests to said first processor for files resident on the same network node as said first processor.

12. An apparatus for storing name space information in a node of a distributed computing system comprised of a plurality of resources, each of said resources being represented as a hierarchical file system, said name space comprised of at least one of said resource hierarchical file systems joined to said name space by one or more name space modification commands, said name space being modifiable on a per process basis, said apparatus comprising:

a mount table for storing an indication of each of said resource hierarchical file systems joined to an indicated portion of said name space by said name space modification commands; and

a server tree path table associated with each of said hierarchical file systems joined to said name space, said server tree path table providing path information for each of said files in said associated hierarchical file system, said path information allowing said file system hierarchy to be recreated, said path information for each of said files including the name of said associated file and an indication of the parent file in said hierarchy of said associated file.

13. The apparatus according to claim 12, wherein said mount table includes an entry for each of said name space modification commands, each entry in said mount table including a value that indicates the order in which said entry was created.

14. The apparatus according to claim 12, wherein said mount table includes an entry for each of said name space modification commands, each entry in said mount table indicating the interpretation of a first file in said name space should be translated into a second file.

15. The apparatus according to claim 12, wherein each of said server tree path tables includes at least one pointer for accessing a plurality of path data structures, each of said path data structures being associated with a file in said associated hierarchical file system and storing said file name and said parent information.

16. A method for identifying a pathname associated with a given channel in a distributed computing system, said distributed computing system including a plurality of distributed resources, each of said resources being represented as a hierarchical file system comprised of one or more files, a collection of one or more hierarchical file systems forming a name space, each of said hierarchical file systems having a root directory, each of said channels representing a given file in said name space, each of said files having a filename comprised of an alphanumeric label, said pathname identifying said file by specifying a path through said hierarchical file system from said root directory to said file, said method comprising:

associating a path data structure with each of said channels, said path data structure storing the filename of an associated file and a pointer to said path data structure associated with a parent file in said name space hierarchy;

receiving an indication of the channel for which the pathname is to be identified;

identifying the path data structure associated with said indicated channel;

retrieving said filename and said pointer to said path data structure associated with the parent file from said identified path data structure; and

following said retrieved pointer to said path data structure associated with the parent file and repeating said previous step until said root directory of said hierarchical file system is reached, each of said filenames retrieved from said path data structures appended to the front of said generated pathname associated data structures being appended to the front of said generated pathname associated with said indicated channel.