摘要:
A storage system has a plurality of disk storage devices and a controller for controlling the plurality of disk storage devices, and diagnosis of a write miss to a disk of the disk storage device. The controller for controlling the plurality of disk storage devices attaches an ID to write data, and after write access to the disk drive, the controller reads the write data from the disk drive, and compares the ID of the read data and the attached ID. The controller updates the ID for each write command. Therefore the old write data and the new write data can be identified, accurate diagnosis is possible, and the failure of an individual disk drive can be detected immediately.
摘要:
A device adaptor that controls a disk, wherein a storage area is divided into a plurality of blocks, detects a block that includes a read error. Next, the device adaptor detects a write omission in blocks that are within a predetermined range from the block that includes the read error. To be specific, the device adaptor reads data from the blocks that are in the vicinity of the block that includes the unrecovered read error in a suspect disk. Next, the device adaptor reads from a redundant disk, data of a block that corresponds to the block that includes the unrecovered read error (data of block “B” in an example shown in FIG. 1) and compares the read data to the data read from the suspect disk to detect the write omission.
摘要:
A storage system has a control module for controlling a plurality of disk storage devices via a transmission path so as to discern the abnormalities of the plurality of disk devices and those of the transmission paths. When a control module for controlling the plurality of disk storage devices detects an error when the disk storage devices are accessed, the control module dummy-accesses the plurality of the disk storage devices on the transmission path, and specifies the suspected failure location based on the result. Therefore it can be discerned whether the suspected failure location is in the transmission path or the disk drive.
摘要:
A magnetic disk apparatus capable of effectively achieving speedup of the response to a read request, a control method and control program therefore are provided. In a control of a magnetic disk apparatus having a RAID configuration in which the same data is written on two disks, a given algorism is selected from a plurality of different algorisms with respect to the data stored in the disks, read commands are issued to the data stored in the two disks, respectively, according to the selected algorithm and the data that has been read first by the read commands that has been issued to the two disks in a read command issuance step is adopted as read data.
摘要:
A storage unit includes a cache memory, a cache controller which accesses the cache memory, one or more disk units, a data receiving unit, a merge interpolation determination unit, a data readout unit, a write data generation unit and a data write unit. The data receiving unit receives, from the cache controller, unit readout data that includes update records updated by the cache controller and is unit of data read from the cache memory. The merge interpolation determination unit determines whether the received unit readout data is merge interpolated. The data readout unit reads, from the disk unit, data corresponding to the unit readout data when the unit readout data is determined to be merge interpolated. The write data generation unit generates data to be written to the disk unit by merge interpolating the unit readout data. The data write unit writes, to the disk unit, the generated data.
摘要:
According to an aspect of an embodiment, a method for controlling a controller connected to a plurality of storage units storing data, the controller including a cache and a buffer, the method comprising the steps of: storing data in the cache; generating parity data corresponding to the data and storing the parity data in the buffer; writing the data and the parity data into the plurality of the storage units; comparing the parity data stored in the buffer with the parity data written into and read out from at least one of the storage unit; deleting, when the parity data stored in the buffer is different from the parity data read out from the storage unit, the parity data from the buffer; and regenerating parity data from data stored in the cache and rewriting the regenerated parity data into at least one of the storage unit.
摘要:
A control method includes the steps of storing data in a cache; generating parity data corresponding to the data stored in the cache and storing the parity data in a buffer; writing the data stored in the cache and the parity data stored in the buffer into a plurality of the storage units; reading the data and the parity data from the plurality of storage units; checking error over the read out data by using the read out parity data; storing, if an error is detected in the read out data and the read out parity data, data for reading a plurality of data constituting a stripe in each of the plurality of storage units in the buffer, regenerating parity data by using the plurality of data readout from the storage units; and overwriting the plurality of data and the parity data into the plurality of the storage units.
摘要:
According to an aspect of an embodiment, a method for controlling a controller connected to a plurality of storage units storing data, the controller including a cache and a buffer, the method comprising the steps of: storing data in the cache; generating parity data corresponding to the data and storing the parity data in the buffer; writing the data and the parity data into the plurality of the storage units; comparing the parity data stored in the buffer with the parity data written into and read out from at least one of the storage unit; deleting, when the parity data stored in the buffer is different from the parity data read out from the storage unit, the parity data from the buffer; and regenerating parity data from data stored in the cache and rewriting the regenerated parity data into at least one of the storage unit.
摘要:
According to an aspect of an embodiment, a method for controlling a controller connected to a plurality of storage units which are arranged in a redundant configuration, the controller reading data stored in the plurality of storage units in accordance with requests received from a host computer, the method comprising the steps of: receiving requests to read data successively from one of the storage units from the host computer; reading a part of requested target data from said one of the storage units; reading associated data and parity data stored in other storage units corresponding to other part of requested target data; generating other part of requested target data on the basis of the associated data and the parity data read out from the other storage units; and transmitting the part of the target data and the other part of the target data to the host computer.
摘要:
To provide a staging method capable of detecting an error in data read from a disk device during staging, a disk array control apparatus 100 includes a data read unit 101 for reading data, a first reference data generation unit 102 for generating first reference data from the read data, a second reference data generation unit 103 for similarly generating second reference data, a true-false determination unit 104 for determining whether or not the data read by the data read unit 101 is correct, and a data write unit 105 for writing data to cache memory.