摘要:
A RAID apparatus that at least duplicates identical data to store thus duplicated data, which, when an instruction of writing data is given, writes data to a local cache and to a mirror cache, and writes data of the local cache to a primary disk and writes data of the mirror cache to a secondary disk. When an instruction of reading out data is given, and specified data is retained in the caches, the RAID apparatus outputs valid data of the local cache or the mirror cache, while when specified data is not retained in the caches, the RAID apparatus outputs valid data of the primary disk or the secondary disk.
摘要:
A cache processing unit manages the data, which is in a cache memory, in a page unit including plurality pieces of block data each of which serves as an access unit of a host, and processes input and output requests from a host to a storage device. In a case in which dirty data in a cache memory which is updated by a host and newer than stored data in a storage device is to be written back to the storage device, and when an incontinuous area is determined in a page of the dirty data, a write-back process unit reads out an incontinuous area from a storage device when a read command is issued, subjects it to a merge so as to generate a continuous area, and performs a write back by one time of write command issue.
摘要:
A storage control apparatus concatenates a plurality of logical units to construct a large capacity logical unit, wherein logical units extending over a plurality of controllers can be concatenated. The channel adapter sends an I/O request to one controller which charges one logical unit constituting the concatenation logical unit, out of a plurality of controllers when an I/O request is sent from a host to the concatenation logical unit LU linking a plurality of logical units, executes I/O processing in the one controller, then sends the I/O request to another controller which charges another logical unit constituting the concatenation logical unit, and continues I/O processing in the another controller.
摘要:
A cache processing unit manages the data, which is in a cache memory, in a page unit including plurality pieces of block data each of which serves as an access unit of a host, and processes input and output requests from a host to a storage device. In a case in which dirty data in a cache memory which is updated by a host and newer than stored data in a storage device is to be written back to the storage device, and when an incontinuous area is determined in a page of the dirty data, a write-back process unit reads out an incontinuous area from a storage device when a read command is issued, subjects it to a merge so as to generate a continuous area, and performs a write back by one time of write command issue.
摘要:
A system creates a configuration definition table to be used for accessing a physical disk according to a host I/O request, to decrease a memory size of the configuration definition table even if common firmware is used for models with different scales. For this, a model correspondence table to show the definition of each model is provided in a configuration setup processing module, the maximum values of this model are recorded in a memory based on this content, and a memory area for each item is secured. And the actual content of the configuration definition is developed on the secured memory. Therefore even for a diversity of models, the memory size of the configuration definition table can be decreased to a size matching the model, with common firmware regardless the model.
摘要:
A RAID apparatus that at least duplicates identical data to store thus duplicated data, which, when an instruction of writing data is given, writes data to a local cache and to a mirror cache, and writes data of the local cache to a primary disk and writes data of the mirror cache to a secondary disk. When an instruction of reading out data is given, and specified data is retained in the caches, the RAID apparatus outputs valid data of the local cache or the mirror cache, while when specified data is not retained in the caches, the RAID apparatus outputs valid data of the primary disk or the secondary disk.
摘要:
In order to provide a disk cache control apparatus allowing an upper level apparatus to carry out a high speed access even if the upper level apparatus accesses a discretionary area of a logical volume, the disk cache control apparatus comprises at least a data storage unit for storing data read out of a lower level apparatus temporarily or for a predefined time, a management information storage unit for storing management information which correlates an area of a logical volume with that of the data storage unit, a management information generation unit for generating the management information, and an access processing unit for accessing data of either the data storage unit or a lower level apparatus.
摘要:
In the present invention, for each set of blocks [#(0) to #(N−1)] storing update data, a history block [#(N)] storing an update state value, for example, a generation, time or check code, showing an update state is provided, constituting a set of management data as a check object. When writing update data on a disk, a new update state value is calculated for the same set of management data and stored in memory as update state confirmation value. The new update state value is also set as write data in the history block in the same set of management data, and the entire of the set of management data including update data and update state value are written onto a disk. The history block is read and disk write omissions are detected by comparing the update state value and the update state confirmation value stored in memory.
摘要:
A storage system prevents a time out error of the host I/O caused by a stagnation of command processing in the storage system for accessing a disk device according to the host I/O request and internal I/O request. In DTC, counters for managing the requests in-process for host I/Os (host read) and internal I/Os (pre-fetch, write back) individually are provided, and the number of processing requests to be issued to a virtual disk (RLU) is limited individually. By assigning priority to the host I/Os, the load balance of the host I/Os and internal I/Os can be controlled. For rebuild/copy back, a dedicated load control mechanism is disposed where the load adjustment between ordinary I/Os and rebuild/copy back is performed.
摘要:
In the present invention, for each set of blocks [#(0) to #(N-1)] storing update data, a history block [#(N)] storing an update state value, for example, a generation, time or check code, showing an update state is provided, constituting a set of management data as a check object. When writing update data on a disk, a new update state value is calculated for the same set of management data and stored in memory as update state confirmation value. The new update state value is also set as write data in the history block in the same set of management data, and the entire of the set of management data including update data and update state value are written onto a disk. The history block is read and disk write omissions are detected by comparing the update state value and the update state confirmation value stored in memory.