公开(公告)号：CA2242833A1

公开(公告)日：1999-01-11

申请号：CA2242833

申请日：1998-07-09

Applicant: IBM

Inventor： SCHMUCK FRANK B ,  ZLOTEK ANTHONY J ,  KANDADAI RADHA R ,  WYLLIE JAMES C ,  CURRAN ROBERT J ,  KISH WILLIAM A ,  SHMUELI BOAZ

IPC: G06F17/30 ,  G06F15/16 ,  G06F15/167

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 sh aring access to files residing on network attached shared disks. Methods are provided for use as a par allel file system in a shared disk environment by use of a scalable directory service for the system with a stable cursor, a segmented allocation map. Dynamic prefetch and cached balance pools for multip le accesses improve the system. Extended file attributes and attribute value sharing are use d for implementation of access control lists in a parallel file system. A metadata node manages file metadata for parallel read and write actions. For our system, tokens are used for metadata node select ion and identification. Enhanced token modes for controlling file size, as well as smart caching of byte range tokens using file access patterns and a byte range lock algorithm using a byte r ange token interface are provided. 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 dis ks becomes unavailable or failed. Synchronous and asynchronous takeover of a metadata node occurs for correction of metadata which was under modification. Locks are not constantly re quired to allocate new blocks on behalf of a user. Recoverable local shares for quota management ar e provided. A quota server and a quota client are provided for each file system, and a quota c heck utility allows parallel environments with minimal interference to other applications which requ ire the use of quotas.

公开(公告)号：DE69840904D1

公开(公告)日：2009-07-30

申请号：DE69840904

申请日：1998-06-17

Applicant: IBM

Inventor： SCHMUCK FRANK B ,  CURRAN ROBERT J ,  WYLLIE JAMES C ,  SHMUELI BOAZ

IPC: G06F17/30

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. Methods are provided for use as a parallel file system in a shared disk environment by use of a scalable directory service for the system with a stable cursor, a segmented allocation map. Dynamic prefetch and cached balance pools for multiple accesses improve the system. Extended file attributes are used for implementation of Access Control Lists in a parallel file system.

公开(公告)号：DE69840903D1

公开(公告)日：2009-07-30

申请号：DE69840903

申请日：1998-06-17

Applicant: IBM

Inventor： SCHMUCK FRANK B ,  CURRAN ROBERT J ,  SHMUELI BOAZ

IPC: G06F17/30

公开(公告)号：CA2242833C

公开(公告)日：2004-09-07

申请号：CA2242833

申请日：1998-07-09

Applicant: IBM

Inventor： KANDADAI RADHA R ,  KISH WILLIAM A ,  SCHMUCK FRANK B ,  SHMUELI BOAZ ,  WYLLIE JAMES C ,  ZLOTEK ANTHONY J ,  CURRAN ROBERT J

IPC: G06F17/30 ,  G06F15/16 ,  G06F15/167

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 dat a sharing access to files residing on network attached shared disks. Methods are provided for use as a parallel file system in a shared disk environment by use of a scalable directory service for the system with a stable cursor, a segmented allocation map. Dynamic prefetch and cached balance pools for multiple accesses improve the system. Extended file attributes and attribute value sharing are used for implementation of access control lists in a parallel file system. A metadata node manages file metadata for parallel read and write actions. For our system, tokens are used for metadata node selection and identification. Enhanced token modes for controlling file size, as well as smart caching of byte range tokens using file access patterns and a byte range lock algorithm using a by te range token interface are provided. Locking techniques reduce the overhead of a token manager whic h 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. Locks are not constantl y required to allocate new blocks on behalf of a user. Recoverable local shares for quota managemen t are provided. A quota server and a quota client are provided for each file system, and a quotacheck utility allows parallel environments with minimal interference to other applications which require the use of quotas.

公开(公告)号：AU6778601A

公开(公告)日：2002-01-08

申请号：AU6778601

申请日：2001-06-19

Applicant: IBM

Inventor： LOY IRIT ,  MARBERG JOHN ,  SHMUELI BOAZ ,  YEHUDAI ZVI ,  CURRAN ROBERT ,  HASKIN ROGER ,  SCHMUCK FRANK ,  WYLLIE JAMES

IPC: G06F9/46 ,  G06F17/30

Abstract: In a cluster of computing nodes having shared access to one or more volumes of data storage using a parallel file system, a method for managing the data storage includes initiating a data management application in the cluster using a data management application programming interface (DMAPI) of the parallel file system. When a request submitted to the parallel file system is received on one of the nodes to perform an operation on a file in one of the volumes of data storage, a data management access right is obtained from the DMAPI responsive to the request The operation is performed on the file using the access right.

公开(公告)号：CA2413434A1

公开(公告)日：2002-01-03

申请号：CA2413434

申请日：2001-06-19

Applicant: IBM

Inventor： MARBERG JOHN ,  LOY IRIT ,  YEHUDAI ZVI ,  SHMUELI BOAZ ,  SCHMUCK FRANK ,  WYLLIE JAMES ,  CURRAN ROBERT ,  HASKIN ROGER

IPC: G06F9/46 ,  G06F17/30

Abstract: In a cluster (20) of computing nodes (22) having shared access to one or mor e volumes of data storage (24) using a parallel file system (28), a method for managing the data storage includes initiating a session of a data management application (32) on a first one of the nodes (40), while running a user application (30) on a second one of the nodes (42). A request is submitted t o the parallel file system by the user application on the second node to perfo rm a file operation on a file in one of the volumes of data storage. A data management event message is sent from the second node to the first node responsive to the request, for processing by the data management application on the first node.