摘要:
A system for injecting I/O faults into a closed system, for example, the injection of link level I/O faults, involves the use of a simulated computing environment. In an embodiment, the system provides for fault injection using an emulated IBM System z environment and including the use of FICON and/or other suitable communication channel protocols. The emulated System z environment may include a simulated z/OS and/or emulated System z hardware and software components.
摘要:
Transferring storage device functionality includes providing a device coupled to the storage device, where the device is separate from the storage device, having the device handle I/O requests between an application and the storage device, and, in response to the application issuing an I/O request, having the device determine if the request corresponds to functionality being transferred and, if so, having the device provide the functionality. The device may be a host computer. The application may run on the host computer. A driver on the host computer may provide the functionality. The functionality may include read only functionality, data compression, data encryption, mirroring, and/or status reporting.
摘要:
Described are techniques for providing access to storage devices. An I/O request directed to a storage device is sent over a network connection from a first computer system to a second computer system. The second computer system includes a simulated computing environment simulating aspects of the first computer system. The I/O request is received at the second computer system. The data storage system is connected to the second computer system over a first connection operating in accordance with a first protocol thereby providing the second computer system access to the storage device of the data storage system. The simulated computing environment of the second computer system may, for example, include emulation capabilities in accordance with the first protocol thereby allowing the second computer system to issue the first I/O request to the first storage device on behalf of the first computer system.
摘要:
A method and apparatus for migrating one or more data sets each having one or more extents from one or more source logical devices to one or more target logical devices concurrently with interaction between the application and the data being migrated. A background operation copies each extent from the source logical device to the target logical device in a copy state. When a certain level of data has been copied, the extent is locked to assure synchronization of the data in the target logical device to the corresponding data in the source logical device. The status is changed to a mirrored state. When the extents for a data set in a source logical device or in a group of data sets have been mirrored, all the extents are changed to a diverted state. I/O requests to the diverted extents thereafter are intercepted and processed according to whether they access an extent that is in the copy, mirrored, or diverted state.
摘要:
Determining when to migrate a data set includes providing performance criteria for the data set, measuring performance of the data set to provide a measured performance of the data set, and, if the measured performance indicates that the data set is not performing according to the performance criteria, choosing a new location for migrating the data set and migrating the data set to the new location. Migrating the data set may include moving the data set to a new location irrespective of whether other applications are accessing the data set. Choosing a new location may include analyzing each of a plurality of locations to determine if a projected performance thereof is according to the performance criteria. Determining when to migrate a data set may include adjusting the performance criteria after the data set has been made operational.
摘要:
Managing data on a storage device includes intercepting unencrypted data to be stored on the storage device, where intercepting unencrypted data is transparent to an application storing the data on the storage device and encrypting the data prior to storage on the storage device. The storage device may include a tape drive and/or a disk drive. Managing data on a storage device may also include migrating data from a first storage location to a second storage location. The first storage location may be the same as the second storage location or the first storage location may be different from the second storage location. The unencrypted data may be intercepted during migration. Managing data on a storage device may also include decrypting data read from the storage device.
摘要:
A dynamic address switching system for use in a data processing system with redundant data storage facilities. A validation routine or module assures that dynamic switching can occur between logical devices. A swapping routine exchanges information in unit control blocks for logical devices in the redundant system thereby to enable I/O requests from a host to be redirected over another channel to the other data storage facility.
摘要:
Described are techniques used in defining and maintaining group data used by multiple host systems. Group data is stored in a distributed fashion on one or more data storage systems. If a device of a data storage system belongs to a group, group data for that group is stored on that data storage system. Group data changes made by one host are communicated to the other hosts by accessing a common data area on each data storage system using a daemon executing on each host. Remotely mirrored groups may be defined on a remote data storage system. A remote mirror of a group includes group data modified in accordance with a point of view of the remote data storage system and a remote host.
摘要:
A data-mirroring method in which, while an I/O process is processing a first I/O request for writing data to a first device, control is obtained from that I/O process executing on a host computer. A second I/O request, this one for writing the data to the second device, is then created. Then, control is returned to the I/O process by causing the first I/O request to be made available to a first data storage system managing the first device.
摘要:
A reclaim system provides for marking each track that is intended to be reclaimed in one or more storage devices and detecting and reclaiming such tracks. The reclaim system may include a feature and/or utility that may be used to mark, detect and return tracks to a free track pool of a virtual provisioning storage system in connection with the use of one or more thin Count Key Data (CKD) devices. Reclaim processing may include identifying tracks to be reclaimed, marking mark the tracks to be reclaimed, detecting tracks that have been marked to be reclaimed, and performing a reclaim task to return the marked and detected tracks to the free track pool, thereby reclaiming the tracks.