摘要:
In a system management apparatus, a failure detection unit detects a readout failure in one of blocks constituting distributed data stored in a first RAID disk array. A request unit requests a computer to supplement one of the blocks of the distributed data stored in the first RAID disk array in which a readout failure irrecoverable by use of only the data stored in the first RAID disk array occurs, where the computer has a second RAID disk array storing a duplicate of the distributed data stored in the first RAID disk array. And an overwriting unit receives data corresponding to the one of the blocks from the computer, and overwrites the one of the blocks with the received data.
摘要:
In a system management apparatus, a failure detection unit detects a readout failure in one of blocks constituting distributed data stored in a first RAID disk array. A request unit requests a computer to supplement one of the blocks of the distributed data stored in the first RAID disk array in which a readout failure irrecoverable by use of only the data stored in the first RAID disk array occurs, where the computer has a second RAID disk array storing a duplicate of the distributed data stored in the first RAID disk array. And an overwriting unit receives data corresponding to the one of the blocks from the computer, and overwrites the one of the blocks with the received data.
摘要:
A storage management device includes a management unit for managing a storage device assigned thereto and a temporary storage unit assigned thereto. The management unit comprises a backup unit for, when data is written into the storage device, storing the data in the previously assigned temporary storage unit before the writing of the data into the storage device is completed, and a take-over unit for, when data which is already stored in the temporary storage unit, but which is not yet written into the storage device exists when the storage device and the temporary storage unit are assigned, writing the not-yet-written data into the storage device.
摘要:
In a cluster system: a heartbeat transmission unit in a service provision node transmits to a monitoring node a heartbeat packet which contains interval information indicating an interval greater than normal intervals at which normal heartbeat packets are transmitted, when an execution unit in the service provision node receives a stop command. Thereafter, when software realizing the execution unit is updated and the execution unit is restarted, the heartbeat transmission unit periodically transmits the normal heartbeat packets to the monitoring node at the normal intervals, where the normal heartbeat packets contain interval information indicating the normal intervals. A heartbeat monitoring unit in the monitoring node determines that the execution unit is in operation, while the monitoring node receives heartbeat packets at intervals indicated by the interval information stored in the monitoring node and updated with received interval information.
摘要:
A storage management apparatus manages a plurality of storage apparatuses connected to each other over a network in a storage system that distributes data among the storage apparatuses and stores the data therein. The storage management apparatus has a patrol process executing unit configured to execute a patrol process to confirm whether a storage area of each storage apparatus operates normally and a patrol flow controlling unit configured to control a patrol flow indicating the speed of the patrol process executed by the patrol process executing unit.
摘要:
A simultaneous failure occurrence degree calculation section calculates a simultaneous failure occurrence degree a value of which becomes smaller with an increase in a difference between numeric values indicative of manufacture date information on a computer which performs a process on the basis of a duplication combination management program. A combination pattern generation section then combines two of a plurality of storage apparatus and generates a plurality of combination patterns. A simultaneous failure occurrence degree adding section then calculates a total of simultaneous failure occurrence degrees corresponding to combinations of storage apparatus for each of the plurality of combination patterns. A duplication combination pattern determination section then determines that a combination pattern for which the total of simultaneous failure occurrence degrees is the smallest is a duplication combination pattern.
摘要:
A data allocation control program manages data allocation when data is distributively stored in a plurality of disk nodes that are shifted to a power saving mode unless access is performed for a certain time. The program produces a plurality of allocation pattern candidates each indicating the disk nodes in which the respective data are to be stored. The program calculates a no-access period expectation that represents an expected value of occurrence of a no-access period during which access is not performed to some of the disk nodes. The program selects as an allocation pattern for data reallocation, one of the plurality of produced allocation pattern candidates with the largest calculated no-access period expectation. The program instructs the disk nodes to reallocate the respective data in accordance with the selected allocation pattern.
摘要:
A data allocation control program manages data allocation when data is distributively stored in a plurality of disk nodes that are shifted to a power saving mode unless access is performed for a certain time. The program produces a plurality of allocation pattern candidates each indicating the disk nodes in which the respective data are to be stored. The program calculates a no-access period expectation that represents an expected value of occurrence of a no-access period during which access is not performed to some of the disk nodes. The program selects as an allocation pattern for data reallocation, one of the plurality of produced allocation pattern candidates with the largest calculated no-access period expectation. The program instructs the disk nodes to reallocate the respective data in accordance with the selected allocation pattern.
摘要:
In a computer which executes an access control program, an authentication information storage unit stores authentication information. A logical volume acquiring unit acquires a logical volume associating data with storage nodes storing the data, from a predetermined database. In response to an access request to access data, a data access unit identifies a storage node to be accessed, based on the logical volume, and sends the authentication information and a command corresponding to the access request to the identified storage node.
摘要:
In a distributed storage system, client terminals make access to virtual storage areas provided as logical segments of a storage volume. Those logical segments are associated with physical segments that serve as real data storage areas. A management data storage unit stores management data describing the association between such logical segments and physical segments. Upon receipt of access requests directed to a specific access range, a segment identification unit consults the management data to identify logical segments in the specified access range and their associated physical segments. A remapping unit subdivides the identified logical segments and physical segments into logical sub-segments and physical sub-segments, respectively, and remaps the logical sub-segments to the physical sub-segments according to a predetermined remapping algorithm. A data access unit executes the access requests based on the remapped logical sub-segments and physical sub-segments.