摘要:
According to this invention, a computer system which provides a client node with a file system mounted into a virtual namespace, comprising: at least one file server which provides the file system; a root node which mounts the file system provided by the file server into the virtual namespace; and a network which couples the file server and the root node. The root node has an interface coupled to the network, a processor coupled to the interface, and a memory coupled to the processor. The root node is configured to: obtain access information set to a directory above a point in the virtual namespace at which the file system is to be mounted when mounting the file system into the virtual namespace; judge whether the obtained access information is to be inherited; and set, to the file system to be mounted, the obtained access information that is to be inherited.
摘要:
A storage system configured to be applied to a broad range of access forms to the file and improve performance of file access control processing. The storage system may store a part of the access control information of a variable size in the basic attribute area of the file, and store the storage format information in the basic attribute area of the file or in the file system metadata area. At the time of processing the access to the file, with reference to the storage format information, the storage system may decode a part of the access control information into an access mask bit string as needed, and perform the preceding access control processing.
摘要:
When receiving a file access from the client, the file access program refers to the mapping table, and processes an access to files of on volumes of RAID groups. The file server analyzes the file access states, and groups the files depending on the access time period, defines the file migration pattern based on the grouping, migrates the files, and then carries out a power management operation like a spin-down/-up operation on the RAID groups based on the migration pattern.
摘要:
An integrated search process on files can be executed by multiple search devices even if setting granularities of access control to the files are different among the search devices. Upon execution of file migration or access control information batch update, information needed for a search server to execute security trimming is extracted from ACL information in a specific format on a document management server. Then the extracted information is transferred to the search server, converted to ACL information in a general format, and registered in a search index in the search server. With this process, a storage system using different access control methods is configured to perform the security trimming based on the ACL information with which access control having the finest setting granularity can be set.
摘要:
There is provided a file server which is capable of suppressing power consumption of a storage apparatus. When a file sharing program receives a file access from a client, the program makes reference to a mapping table. Then, the program addresses the access to the target file in the volume of a RAID group in which the target file is stored. A coupling-request reception program memorizes a coupling time for each user into a coupling history table. A grouping program applies a grouping to users whose coupling time-zones are similar to each other. A data transfer program transfers, into the same RAID group, data of the files associated with the grouped users, thereby collecting the data of the files into the same RAID group. On account of this, the time-zone during which no access is made to the RAID group (i.e., non-coupling time-zone) can be made longer. Accordingly, a spin-up/down request program makes a spin-down request to the RAID group in the non-coupling time-zone.
摘要:
A storage network system that prevents waste of a core's resources and is thereby operated efficiently, and a method for controlling such a storage network system are provided.Policy differences between a core and a plurality of edges are buffered by enabling hierarchical control of data storage on the side of the plurality of edges in cooperation with hierarchical control of data storage on the core side, and the buffered policy is applied to the hierarchical control of the data storage on the core side.
摘要:
A global namespace (GNS) is constructed across heterogeneous network-attached storage (NAS) devices by organizing metadata servers into a cluster layer having cluster nodes and a consistent hashing overlay layer having overlay nodes. The global namespace is distributed based on workload at the cluster and overlay nodes, and a portion of the GNS namespace may be migrated from one layer to the other. Cluster nodes may be added to the cluster layer, overlay nodes may be added to the consistent hashing overlay layer. Further, a cluster node may become an overlay node or an overlay node may become a cluster node. To access the global namespace, a request for namespace metadata managed by a cluster node is serviced by the cluster node, but if the requested namespace metadata have been migrated to the consistent hashing overlay layer, the cluster node will forward the request to the responsible overlay node for servicing.
摘要:
Time required for a file server to determine files to be migrated from a first storage region to a second storage region is reduced. A file service system coupled to the storage system for providing a first storage region and a second storage region stores identifiers of files that have been accessed by a client, in accessed file identifier information and determines the files to be migrated from the first storage region to the second storage region based on metadata of the files stored in the first storage region and the accessed file identifier information.
摘要:
In the HSM function, when supporting the file system snapshot function, a differential data volume is required for both a high-speed volume and a low-speed volume. Accordingly, when supporting the snapshot function by using a file system provided by the HSM function, a large amount of differential data volume is required. When the manager gives an instruction to acquire a snapshot, a snapshot management program executes snapshot acquisition processing to acquire a snapshot. In this processing, a snapshot reference relation building processing part copies a file storage location management table, and relates the file storage location management table to the acquired snapshot. Then, the snapshot reference relation building processing part checks a file that is migrated to a file system which does not have a differential data volume. Finally, the snapshot reference relation building processing part creates a link of the migrated file to the snapshot which is acquired previous time.
摘要:
The system of the present invention curbs an unnecessary inter-tier file migration and provides an inter-tier file migration that conforms to the utilization state. The hierarchical storage management system respectively migrates a frequently accessed file to the upper Tier, and an infrequently accessed file to the lower Tier. This system carries out inter-tier file migration in conformance with the utilization state. This system provides functions for anchoring a Tier for storage for a specified file, and for not permitting an inter-tier file migration.