US6021508A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Parallel file system and method for independent metadata loggin
Assignee: International Business Machines Corp
•
Filing: 1997-07-11
•
Granted: 2000-02-01
•
Expires: 2017-07-11
Abstract
Abstract A computer system having a shared disk file system running on multiple computers each having their own instance of an operating system and being coupled for parallel data sharing access to files residing on network attached shared disks. Locking techniques reduce the overhead of a token manager which is also used in the file system recovery if a computer participating in the management of shared disks becomes unavailable or failed. Synchronous and asynchronous takeover of a metadata node occurs for correction of metadata which was under modification and a new computer node to be a metadata node for that file. Locks are not constantly required to allocate new blocks on behalf of a user.
Inventors
- 👤 Frank B. Schmuck
- 👤 Radha R. Kandadai
- 👤 Anthony J. Zlotek
- 👤 Robert J. Curran
- 👤 William A. Kish
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-03 01:14
CPC Classifications
G; G06; G06F; G06F9/00; G06F9/06; G06F9/46; G06F9/52; G06F16/00; G06F16/10; G06F16/18; G06F16/1858
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
158 words
In a computer system having a shared disk file system running on multiple computer nodes each having their own instance of an operating system and being coupled for parallel data sharing access to files residing on network attached shared disks, a method for managing metadata and for facilitating recovery from a computer node failure and/or a failure of communication among computer nodes, including the steps of: providing independent access to attached shared disks, while maintaining consistency for each file through the use of a distributed token manager and locking mechanism, and transferring metadata for a file to and from said shared disk via a single metadata computer node, while logging metadata updates that could result in inconsistencies in event of a failure independently by each computer node to a shared disk file log accessible to all computer nodes in case of failure in order to avoid the need for a file system scan upon failure of a node.
Claim 2
Dependent Claim
158 words
In a computer system having a shared disk file system running on multiple computer nodes each having their own instance of an operating system and being coupled for parallel data sharing access to files residing on network attached shared disks, a method for managing metadata and for facilitating recovery from a computer node failure and/or a failure of communication among computer nodes, including the steps of: providing independent access to attached shared disks, while maintaining consistency for each file through the use of a distributed token manager and locking mechanism, and transferring metadata for a file to and from said shared disk via a single metadata computer node, while logging metadata updates that could result in inconsistencies in event of a failure independently by each computer node to a shared disk file log accessible to all computer nodes in case of failure in order to avoid the need for a file system scan upon failure of a node.
Claim 3
Dependent Claim
86 words
In a system employing a file system according to claim 1, wherein said metadata includes token state, and metadata logs, including the further step of monitoring a process group of nodes and processes on all of said process group of nodes for detecting failures at a node in said group and in communication between nodes, in said group and informing members of said process group of nodes when a processor node fails or when a new processor node attempts to join the process group of nodes.
Claim 4
Dependent Claim
67 words
In a system employing a file system according to claim 2, including the further step of coordinating disk access among group members through distributed locking, wherein a member must obtain a lock before reading or changing a particular piece of data or metadata on a shared disk, and designating one group member as a designated lock coordinator for determining which locks can be held on which node.
Claim 5
Dependent Claim
42 words
In a system employing a file system according to claim 3, including the further step of halting and denying further file system operations when less than a majority of processors capable of accessing a disk are a member of a process group.
Claim 6
Dependent Claim
30 words
In a system employing a file system according to claim 1, including the further step of blocking access from a given node to a given disk upon a failure event.
Claim 7
Dependent Claim
40 words
In a system employing a file system according to claim 5, wherein after the initiation of a recovery after a failure event, recovery steps are executed on all nodes before a next step is enabled anywhere in the file system.
Claim 8
Dependent Claim
57 words
In a system employing a file system according to claim 5, including the further step of terminating communication to a filed process, fencing the failed processor whereby the shared disks system ceases to honor disk request from a failed node denying access to a shared disk even though the failed node has not detected the communication failure.
Claim 9
Dependent Claim
47 words
In a system employing a file system according to claim 7, including recovery of a lock state, if necessary, by informing a token manager and locking mechanism's lock coordinator which then suspends granting of locks that are held by a failed node at the time of failure.
Claim 10
Dependent Claim
26 words
In a system employing a file system according to claim 8, including replaying of a metadata log of the failed node by the file system coordinator.
Claim 11
Dependent Claim
39 words
In a system employing a file system according to claim 1, including recovering from a failure of a node for each file system of said computer system, by repeating the recovery for each file system of the computer system.
References Cited (Backward Citations)
US4274139A
Cited
US5202971A
Cited
US5043876A
Cited
US5226159A
Cited
US5339427A
Cited
US5463736A
Cited
US5454108A
Cited
US5490270A
Cited
US5566297A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-01-31 | 2003-08-02 | 2004-02-01 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-01-31 | 2007-08-02 | 2008-02-01 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-01-31 | 2011-08-02 | 2012-02-01 | $7700.00 | Window Closed / Lapsed |