170 results about "Logical disk" patented technology

Methods and systems for mapping logical disk addresses to physical locations so as to achieve consistent sustained performance for a striped disk array I / O subsystem. Stripes (regions) are defined by a region mapping table. Zones are defined by the disk manufacturer as groups of cylinders having identical number of sectors per track. Outer zones store more data and therefore provide a higher level of sustained performance as compared to inner zones. Substantially half the disks in the array are mapped such that logical sequential blocks are allocated from outer most, higher performance, zones to inner, lower performance, zones. The other half of the drives in the array are mapped from inner zones to outer zones. Each region (stripe) therefore includes a mix of higher performance zones and lower performance zones. Each region therefore provides more consistent sustained performance as compared to prior techniques. Improved consistency in performance enables better utilization of storage system performance and storage capacity as compared to prior techniques. Alternate embodiments of the invention provide for striping such regions using multiple actuators within a single drive spindle and for providing such mapping functions within low level drive control electronics.

A data storage system and a method for managing a data storage system are provided. A storage controller is programmed with a disk configuration for each of one or more logical disk arrays and a protection level k. The available storage space from one or more disk drives in the data storage system is merged into a single virtual address space and the merged storage space is divided into storage segments. Next, the storage segments are allocated among the logical disk arrays and a configuration table is generated indicating the number of storage segments in each logical disk array and the physical location of each storage segment on a disk drive. The configuration table is stored in the storage controller and k copies of data may then be stored on the logical disk arrays.

Provided is a database management system obtains storage mapping information which associates addresses in a plurality of physical disk drives within the storage system and addresses in logical disk drives including these physical disk drives, to create queues individually for each of the plurality of physical disk drives. The database management system receives a plurality of read requests which request to read data out of the physical disk drives via the logical disk drives provided by the storage system, sorts the read requests by their destination, and accumulates the read requests in the respective queues associated with the request destination physical disk drives. The database management system reallocates the accumulated read requests into an order that shortens the data read time in each physical disk drive, and then issues the read requests to the storage system.

A storage system connected to a terminal, the computer system includes: a plurality of drive devices that respectively drive a plurality of physical disks each having a physical storage area; a RAID configuration unit that configures a plurality of RAID groups by grouping two or more of the plurality of physical disks; a logical disk creation unit that creates, for the terminal through the RAID group, a logical disk having a logical storage area associated with the physical storage area; a memory for storing a RAID group control table showing, for each the RAID group, (i) a free capacity that is the amount of physical storage area remaining in the RAID group to be able to be associated with the logical disk and (ii) a power status of the RAID group; a receiver that receives a request for creating a new logical disk; and an area allocation unit that allocates to the new logical disk the physical storage area remaining in the RAID group selected by giving priority to a RAID group in a powered state over a RAID group in a non-powered state with reference to the RAID group control table.

The embodiment of the invention discloses a method, a device and a system for managing a simply-configured memory array. The method for managing the simply-configured memory array comprises the following steps: receiving a request message for establishing a virtual logical disk, wherein the request message comprises the memory capacity of the request; establishing a first kind of virtual logical disks according to the request message; feeding back the size of the first kind of the virtual logical disks according to the memory capacity of the request; and allocating logical units for the first kind of the virtual logical disks in a memory resource pool. Through the scheme, the embodiment of the invention realizes the conversion from the memory array in traditional arrangement into the memory array supporting the simple configuration under the condition of not changing the framework of the original memory array and not influencing the original memory data and service, thereby reducing the cost for realizing the simple configuration.

The invention discloses an automatic thin-provisioning method for optimizing space management. The method comprises the following steps: forming a storage pool through a plurality of logical disks; forming a logic volume space by crossing the plurality of logical disks; dividing the storage pool space into two stages, namely, dividing the logical disk space into a plurality of data sections; dividing each data section into a plurality of data blocks; distributing the data sections to a logical volume through the storage pool; maintaining the distributed data section by each logical volume; distributing the space for storing the written data from the maintained data section through the logical volume while the upper application writes data to the logical volume; applying to the storage pool for distribution of a new data section through the logical volume in case that the data section maintained by the logical volume does not have enough space for distributing. Compared with the prior art, the automatic thin-provisioning method for optimizing the space management, the random distribution degree of the data in a physical disk can be effectively reduced; meanwhile, the storage space utilization rate is high; in addition, the load balance of a storing device is ensured while distributing the storage space.

The invention discloses a memory virtualization system and a method to balance IO load of physical disks. In the system, a striping module processes the physical disks by means of striping data to build a logic pond; a disk-building module builds a logic disk or a logic disk array on a logic memory pond; and a volume-building module builds a logic volume on the logic disk or the logic disk array. With the three layers of virtualization process, a read and write request of a customer is performed by being dispersed to the plurality of physical disks, and compared with the prior art that the read and write request of the customer is performed by being concentrated on one physical disk, the invention balances the IO load of the physical disks, effectively prevents some date of the physical disks from over heating, and improves the security and the performance of the memory system.

The invention discloses a storage space configuration and management method applied to cloud computing, which is characterized by including the following steps that a) a storage controller enables a heterogeneous storage physical disk to be integrated with a virtual volume pool and virtualizes storage capacity in the virtual volume pool into a dynamic configuration pool; b) the storage controller divides the dynamic configuration pool into one or more than one small block storage units which form a logic disk; c) the storage controller distributes logic unit number for each logic disk which is divided into a preposition type and a special service type, and the logic unit number of the special service type logic disks correspond to all services and identification of the services; and d) the storage controller utilizes the preposition type logic disk to distribute unit storage units for task writing in each time and transfers and stores metadata in the unit storage units to the corresponding special service type logic disk.

Data volumes on local hosts are discovered and managed by federated Java beans that run on each host. The Java beans form part of a three-tiered data services management. The lowest tier comprises management facade software running on each machine that converts platform-dependent interface written with the low-level kernel routines to platform-independent method calls. The middle tier is a set of federated Java beans that communicate with the management facades and with the upper tier of the system. The upper tier of the inventive system comprises presentation programs that can be directly manipulated by management personnel to view and control the system. The federated beans can configure and control data volumes with either a SCSI terminal emulation interface or a storage volume interface and use a logical disk aggregator to present all volumes available on a local host as a single “logical volume” in which all information regarding the various volumes is presented in a uniform manner.

A computer database system has one or more application buffers to use in performing input / output (I / O) operations. A file system receives contents of the application buffers. Contents of the file system are written into a nonvolatile memory. A backup command directed at the file system is received. A data contents of the one or more application buffers is moved to the file system in response to receiving the backup command, and the data contents are written to the nonvolatile memory. An operating system blocks I / O operations directed to the file system after the data contents of the one or more application buffers are moved to the file system. A snapshot of the nonvolatile memory is generated while the I / O operations directed to the file system are blocked.

The invention provides a terminal identification method and a terminal identification device. The terminal identification method comprises the following steps of: judging whether a unique identifier of a terminal exists in a system disk or a registry of the terminal, wherein the unique identifier is obtained by computation according to identifier computation parameters which comprise at least one of a terminal name of the terminal, a globally unique identifier (GUID) and the remaining disk size of the system disk; if the unique identifier does not exist in the system disk or the registry of the terminal, acquiring the unique identifier from other logical disks of the terminal, wherein the unique identifier is pre-stored in the other logical disks; and identifying the terminal by using the unique identifier which is obtained from the other logical disks. By the invention, an effect of accurately identifying the terminal is achieved.