Decentralized file mapping in a striped network file system in a distributed computing environment

• US 6,029,168 A
• Filed: 01/23/1998
• Issued: 02/22/2000
• Est. Priority Date: 01/23/1998
• Status: Expired due to Term

• Current Assignee: PMC-Sierra Incorporated (Microchip Technology Incorporated)
• Original Assignee: Tricord Systems, Inc. (Steel Partners Holdings LP)
• Inventors: Frey, Alexander H.
• Primary Examiner(s): Amsbury, Wayne
• Assistant Examiner(s): HAVAN, THU THAO
• Application Number:
  US09/012,267
• Time in Patent Office: 760 Days
• Field of Search: 707/3, 707/8, 707/9, 707/10, 707/201, 707/202, 707/203, 707/205, 707/206, 711/150, 345/424, 345/31, 345/335, 340/825, 709/104, 709/219, 364/228, 364/222, 364/943, 714/16, 714/6
• US Class Current: 1/1
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99938 Concurrency, e.g. lock management in shared database
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

The invention discloses apparatus and process in which data files are distributed across a large scale data processing system to enable balance of work loads and storage loads at a plurality of nodes. Specifically, the invention provides significant advances in data base management by distributing meta-data in a plurality of file storage nodes to isolatively and distributively store file data in a distributed computing environment. This provides efficient allocation of storage space and work loads among nodes.

Claims

1. A method to manage the decentralized access of files in a distributed computing system having a first file access manager controlling access to and having physical address information for only a first subset of logical file blocks of a stored file and a second file access manager controlling access to and having physical address information for only a second, disjoint, subset of logical file blocks of the stored file, the method comprising:

receiving, by the first file access manager, a request to access a specified logical file block of the stored file;

determining, by the first file access manager, the specified logical file block is in the second subset of logical file blocks of the stored file;

routing, by the first file access manager, the request to the second file access manager; and

accessing, by the second file access manager, the specified logical file block.

2. The method of claim 1, wherein the act of receiving a request to access a specified logical file block of the stored file comprises receiving a file identifier associated with the stored file, the file identifier including a starter node attribute.

3. The method of claim 2, wherein the starter node attribute of the file identifier identifies the first file access manager.

4. The method of claim 3, wherein the act of identifying a second file access manager is based on logical-file-block-to-file-access-manager mapping information stored at the first file access manager.

5. The method of claim 1, wherein the first file access manager and the second file access manager execute on a common node of the distributed computer system.

6. The method of claim 1, wherein the first file access manager executes on a first node of the distributed computer system and the second file access manager executes on a different node of the distributed computer system.

7. The method of claim 1, wherein the requested access comprises a read operation.

8. The method of claim 1, wherein the requested access comprises a delete operation.

9. A memory for storing data for access by a file access manager program being executed on a distributed computer system node, comprising:

a data structure stored in said memory, said data structure including,

a first table for storing information indicating a physical storage location for one or more logical file blocks associated with a first stored file only if the first stored file has a zero record stored on a storage device controlled by the file access manager program; and

a second table for storing information indicating one or more nodes in the distributed computer system, each indicated node having stored therein one or more logical file blocks associated with a second stored file, the second stored file without a zero record stored on a storage device controlled by the file access manager program.

10. The memory of claim 9, wherein the second table is without physical address information for those logical storage blocks associated with the second stored file.

11. The memory of claim 9, wherein the second table comprises a hash list.

12. A memory for storing data for use by a file access manager program being executed on a node in a distributed computer system to access a specified logical file block of a specified stored file, the memory comprising:

a data structure stored in said memory, said data structure including,

a node attribute to indicate one of a plurality of nodes in a distributed computer system, the indicated node having stored therein a zero record of the specified stored file;

an index attribute to identify a location of logical-file-block-to-file-access-manager information at that node indicated by the node attribute; and

a logical block attribute to indicate the specified logical block.

13. The memory of claim 12, wherein the logical block attribute comprises a data offset portion to indicate the position of the start of the specified logical file block relative to the start of the specified stored file.

14. The memory of claim 12, wherein the data structure further comprises a version attribute to indicate the version of the specified stored file.

15. The memory of claim 12, wherein the logical-file-block-to-file-access-manager information comprises physical address information only if the specified logical file block is stored in a media controlled by the file access manager program.

16. The memory of claim 12, wherein the logical-file-block-to-file-access-manager information does not comprise physical address information if the specified logical file block is stored in a media controlled by a second file access manager program.