Abstract: A method for accelerating a background replication process on storage volumes during application I/O (input/output) requests includes reading requested data from a first storage volume. The method stores the requested data in an embedded memory device, and providing the requested data to the application. The method receives a read request from the background replication process. The method responds to the read request from the background replication process by providing data from the embedded memory device to the requesting background replication process concurrently with providing data to the requesting application. The method stores, by the background replication process, the data provided from the embedded memory device onto a second storage volume.

Abstract: A mechanism is provided in a data processing system for data replication. The mechanism divides a set of data to be replicated into a plurality of slices. Each slice of the plurality of slices is to be copied from a primary site to a master secondary site within a plurality of secondary sites. The plurality of secondary sites comprises the master secondary site and at least one subsidiary peer site. The mechanism copies the plurality of slices from the primary site to the plurality of secondary sites such that the master secondary site stores a first slice of the set of data and each of the at least one subsidiary peer site stores a corresponding slice of the set of data. Responsive to completion of copying the plurality of slices to the plurality of secondary sites, for each of the at least one subsidiary peer site, the mechanism initiates copying of the corresponding slice of the set of data to the master secondary site.

Abstract: Embodiments of the present invention provide, systems, methods, and computer program products for optimizing disk images. Embodiments of the present invention generate sets of data that more efficiently describe unallocated regions of a virtual disk. Embodiments of the present invention can afford users to read a virtual disk and write data on the virtual disk as a disk image on local or remote storage components. Embodiments of the present invention can reduce network bandwidth required to create a disk image by reducing memory capacity needed to describe unallocated regions of the virtual disk.

Abstract: Embodiments of the present invention provide, systems, methods, and computer program products for optimizing disk images. Embodiments of the present invention generate sets of data that more efficiently describe unallocated regions of a virtual disk. Embodiments of the present invention can afford users to read a virtual disk and write data on the virtual disk as a disk image on local or remote storage components. Embodiments of the present invention can reduce network bandwidth required to create a disk image by reducing memory capacity needed to describe unallocated regions of the virtual disk.

Abstract: In a method for replicating data, a first controller receives a request to write data from a device. The first controller communicates with a second controller to obtain information necessary for the second controller to receive the data. The first controller determines settings that allow for the first controller and the second controller to each receive the data using the information necessary for the second controller to receive the data. The first controller sends the settings that allow for the first controller and the second controller to each receive the data to the device. The first controller receives the data sent from the device to the first controller and the second controller based on the settings that allow for the first controller and the second to each receive the data.

Abstract: A method for accelerating a background replication process on storage volumes during application I/O (input/output) requests includes reading requested data from a first storage volume. The method stores the requested data in an embedded memory device, and providing the requested data to the application. The method receives a read request from the background replication process. The method responds to the read request from the background replication process by providing data from the embedded memory device to the requesting background replication process concurrently with providing data to the requesting application. The method stores, by the background replication process, the data provided from the embedded memory device onto a second storage volume.

Abstract: A mechanism is provided in a data processing system for data replication. The mechanism divides a set of data to be replicated into a plurality of slices. Each slice of the plurality of slices is to be copied from a primary site to a master secondary site within a plurality of secondary sites. The plurality of secondary sites comprises the master secondary site and at least one subsidiary peer site. The mechanism copies the plurality of slices from the primary site to the plurality of secondary sites such that the master secondary site stores a first slice of the set of data and each of the at least one subsidiary peer site stores a corresponding slice of the set of data. Responsive to completion of copying the plurality of slices to the plurality of secondary sites, for each of the at least one subsidiary peer site, the mechanism initiates copying of the corresponding slice of the set of data to the master secondary site.

Abstract: A method for replicating data from a first volume to a second volume includes receiving a first data request comprising a request for a first portion of data, wherein the first portion is part of a first volume. The first portion of data is read, and so is at least a second portion of data in addition to the first portion of data requested in the first data request. In response to determining that the second portion of data should be replicated to the second volume, the second portion of data is written to the second volume.

Abstract: A mechanism is provided in a data processing system for data replication. The mechanism divides a set of data to be replicated into a plurality of slices. Each slice of the plurality of slices is to be copied from a primary site to a master secondary site within a plurality of secondary sites. The plurality of secondary sites comprises the master secondary site and at least one subsidiary peer site. The mechanism copies the plurality of slices from the primary site to the plurality of secondary sites such that the master secondary site stores a first slice of the set of data and each of the at least one subsidiary peer site stores a corresponding slice of the set of data. Responsive to completion of copying the plurality of slices to the plurality of secondary sites, for each of the at least one subsidiary peer site, the mechanism initiates copying of the corresponding slice of the set of data to the master secondary site.

Abstract: In a method for replicating data, a first controller receives a request to write data from a device. The first controller communicates with a second controller to obtain information necessary for the second controller to receive the data. The first controller determines settings that allow for the first controller and the second controller to each receive the data using the information necessary for the second controller to receive the data. The first controller sends the settings that allow for the first controller and the second controller to each receive the data to the device. The first controller receives the data sent from the device to the first controller and the second controller based on the settings that allow for the first controller and the second to each receive the data.

Abstract: Embodiments of the present invention disclose a method, computer program product, and system for data deduplication. Receiving a protocol data unit (PDU) that includes data to be stored on a system and a hash value that corresponds to the data. Determining whether the hash value of the received PDU matches a stored hash value that corresponds to data that is stored in the system. Responsive to determining that the hash value of the received PDU does not match a stored hash value, storing the data included in the received PDU in the system. In another embodiment, the system is an iSCSI attached storage system, and the PDU is an iSCSI PDU.

Abstract: Embodiments of the present invention disclose a method, computer program product, and system for data deduplication. Receiving a protocol data unit (PDU) that includes data to be stored on a system and a hash value that corresponds to the data. Determining whether the hash value of the received PDU matches a stored hash value that corresponds to data that is stored in the system. Responsive to determining that the hash value of the received PDU does not match a stored hash value, storing the data included in the received PDU in the system. In another embodiment, the system is an iSCSI attached storage system, and the PDU is an iSCSI PDU.

Abstract: A method that includes identifying a failure indication for a first data storage device that is a member of a first RAID group within a storage array. The method further can include, via a processor external to the storage array, identifying a virtual drive that is defined to include at least one logical storage volume defined in a second RAID group. The virtual drive can be provisioned to serve as a virtual hot spare within the first RAID group to replace the first data storage device.

Abstract: A method that includes identifying a failure indication for a first data storage device that is a member of a first RAID group within a storage array. The method further can include, via a processor external to the storage array, identifying a virtual drive that is defined to include at least one logical storage volume defined in a second RAID group. The virtual drive can be provisioned to serve as a virtual hot spare within the first RAID group to replace the first data storage device.

Abstract: A method for replicating data from a first volume to a second volume includes receiving a first data request comprising a request for a first portion of data, wherein the first portion is part of a first volume. The first portion of data is read, and so is at least a second portion of data in addition to the first portion of data requested in the first data request. In response to determining that the second portion of data should be replicated to the second volume, the second portion of data is written to the second volume.

Abstract: Embodiments of the present invention utilize a hard disk drive buffer to minimize seek delays during background replication of a first volume and fulfillment of application input/output requests. According to one aspect of the present invention, a requested first sector of data is read along with one or more additional sectors of data. Upon determining that one or more of the read sectors of data should be replicated, the sectors are read from the disk buffer and written to a second volume.

Abstract: A mechanism is provided in a data processing system for data replication. The mechanism divides a set of data to be replicated into a plurality of slices. Each slice of the plurality of slices is to be copied from a primary site to a master secondary site within a plurality of secondary sites. The plurality of secondary sites comprises the master secondary site and at least one subsidiary peer site. The mechanism copies the plurality of slices from the primary site to the plurality of secondary sites such that the master secondary site stores a first slice of the set of data and each of the at least one subsidiary peer site stores a corresponding slice of the set of data. Responsive to completion of copying the plurality of slices to the plurality of secondary sites, for each of the at least one subsidiary peer site, the mechanism initiates copying of the corresponding slice of the set of data to the master secondary site.

Abstract: Embodiments of the present invention utilize a hard disk drive buffer to minimize seek delays during background replication of a first volume and fulfillment of application input/output requests. According to one aspect of the present invention, a requested first sector of data is read along with one or more additional sectors of data. Upon determining that one or more of the read sectors of data should be replicated, the sectors are read from the disk buffer and written to a second volume.

Abstract: A mechanism is provided for provisioning and allocating logically extended virtual disks. Responsive to an identification of a negative operational issue with a storage device in a plurality of storage devices in a storage subsystem, a determination is made as to whether a hot spare disk is available to replace the storage device. Responsive to the hot spare disk being unavailable, a logically extended virtual disk is allocated as a replacement for the storage device. Data stored on the storage device is then rebuilt on the logically extended virtual disk.

Abstract: A method that includes identifying a failure indication for a first data storage device that is a member of a first RAID group within a storage array. The method further can include, via a processor external to the storage array, identifying a virtual drive that is defined to include at least one logical storage volume defined in a second RAID group. The virtual drive can be provisioned to serve as a virtual hot spare within the first RAID group to replace the first data storage device.