摘要:
A method and apparatus for verifying the organization of a magnetic disk storage system in which individual storage logical volumes are grouped in sequence as components of a meta device. Data concerning the organization of the meta device is stored in each storage logical volume in a dedicated storage area and in a global memory accessible to the data processing system. Whenever the integrity of the meta device is tested, the data in the configuration file and the data in the dedicated storage area are analyzed and compared. If identity exists, integrity is assumed. If identity does not exist various errors are generated to prevent use of the storage logical volume until the problem is resolved.
摘要:
A method, system, and program product for data replication comprising, building a virtual image of a point in time in a Data Protection Appliance (DPA), taking a storage array based snapshot of a current image, creating a hot pull session and exposing a LUN containing the point in time image at the DPA.
摘要:
Migrating data from a source storage device to a target storage device includes creating new paths to the target storage device, setting the target storage device to a state where I/O operations are initially accepted, where accepted I/O operations are rejected some time after acceptance, setting the source storage device to a state where at least some I/O operations are rejected, transferring metadata corresponding to the source storage device to the target storage device, where state information is transferred from the source storage device to the target storage device and setting the target storage device to a state where I/O operations are accepted and performed. Migrating data from a source storage device to a target storage device may also include creating new volumes on the target storage device and transferring data from the source storage device to the target storage device.
摘要:
Described are techniques for migrating data from a source device of a source data storage system to a target device of a target data storage system. The target data storage system is configured to operate in accordance with a spoofing mode and a pass-through mode. A first set of commands from the host are processed for the target device when in the pass-through mode and the spoofing mode with respect to the target device. Data is migrated from the source device to the target device. The target data storage system is configured to operate in an echo mode for the target device and processes a second set of commands from the host when in the echo mode for the target device using information stored in the cache from processing the first set of commands.
摘要:
In one aspect, a method includes providing a first storage volume to receive I/Os from a host and having a device ID, providing a second storage volume to receive the I/Os and having a device ID and performing a recovery that includes rebooting the host and recognizing, at the host, the second storage volume as the first storage volume using the device ID of the second storage volume being identical to the device ID of the first storage volume. The first storage volume is in an active mode and the second storage volume is in a passive mode.
摘要:
Described are techniques for storing first data to a logical data in accordance with a static policy to service reads using data stored in a first storage area. The logical device has its data stored on a first physical device that is mirrored on a second physical device. Each of the first and second physical devices include a first storage area and a second storage area. The first storage areas provide better performance than the second storage areas. The first data is stored at a first location in one of the first storage areas of one of the physical devices and at a second location in one of the second storage areas at the other of the first and the second physical devices. An operation to read the first data is serviced by accessing the first data stored in the first area of the one physical device.
摘要:
Described are techniques for performing path selection. A data operation is received for a metadevice comprising a plurality of metadevice members. A first of the plurality of metadevice members is determined. The first member is the member to which the data operation is directed. In accordance with a load balancing technique for the first metadevice member, path selection is performed to select a first of a plurality of paths. The metadevice is accessible over each of the plurality of paths. The data operation is sent over the first path.
摘要:
A system creates a storage area network (SAN) centric storage virtualization layer in front of storage devices. The system allows data mobility and migration without disruption to the one or more host servers attached to the SAN. Specifically, the host servers are not disrupted when switching I/Os between physical and virtual storage, for example, by taking advantage of WWPN spoofing and Fibre Channel VSAN technology. The use of VSANs effectively allow multiple virtual directors and/or switches to be created within a physical director and/or switches, each with their own separate name server, thereby providing complete isolation from one another. The host-storage pathing information is unchanged as the original physical storage port's WWPNs are spoofed by the virtual storage port. The result is two identical WWPNs within the SAN which is normally disallowed; however, by separating the physical port WWPN into one VSAN and the virtual port WWPN into another, the restriction may be circumvented.
摘要:
Described are techniques for providing a plurality of logical connections over a physical connection. A login frame is transmitted for each of the plurality of logical connections from an initiator port to a target port. The login frame includes a payload comprising a first identifier of a virtual port associated with the initiator port. Upon receiving the login frame, a login acceptance frame is transmitted from the target port to the initiator port and includes a second identifier of a virtual port associated with said target port in the payload. The initiator and target ports are included in hardware components in compliance with a version of a protocol that does not support multiple logical connections over a same physical connection. Each logical connection is established as a result of processing including transmitting a login frame and a login acceptance frame.
摘要:
Described is a distributed copying technique that may be used in migrating large amounts of data from one or more source devices to one or more destination devices. The data source is divided into partitions. As Fibre Channel adapters (FAs) become available, each of the FAs may copy a partition of the data. In connection with specifying paths used for the distributed copying technique, a preferred path selection (source port-target port mapping) may be made by executing code in a controlling data storage system to perform discovery processing. The preferred path selection is used for the duration of the distributed copying unless the preferred path is unable to transmit data. A target port of the preferred path may be randomly selected from all accessible target ports, and/or in accordance with a specified portion of the target ports. Preferred paths may also be specified using an API (application programming interface).