Abstract: A method, executed by at least one processor, includes generating a snapshot for a plurality of data files, receiving an update request for a selected file of the plurality of data files, determining if the selected file is subject to a backup policy, updating the selected file without preserving the snapshot of the selected file if the selected file is not subject to the backup policy, and updating the selected file while preserving the snapshot of the selected file if the selected file is subject to the backup policy. A corresponding computer program product and computer system are also disclosed herein.

Abstract: Embodiments provide a method, a system, and a computer program product for performing copy operations of one or more data units in a hierarchical storage management (HSM) system. The HSM system includes an upper layer and a lower layer. The upper layer includes multiple storage nodes having a grid configuration. The method comprises scheduling a copy operations of multiple data units each of which is stored in at least one of the multiple storage nodes such that loads on the copy operations are distributed among the multiple storage nodes in which the multiple data units are stored and copying the multiple data units to the lower layer in accordance with the scheduling.

Abstract: As disclosed herein, a computer implemented method for reading a record on a storage device includes reading, by a processor, the record from the storage device, and generating, by a processor, a header corresponding to the record. The method further includes adding, by a processor, header information corresponding to the metadata to the header, and providing, by a processor, the header and the record to a host. Also disclosed herein, a computer implemented method for writing a record on a storage device includes receiving, by a processor, a header and the record from a host to be written to the storage device, and determining, by a processor, metadata included in header information corresponding to the header. The method further includes determining, by a processor, a write behavior specified by the metadata, and writing, by a processor, the metadata and the record to the storage device according to the write behavior.

Abstract: In an approach to determining a storage to which to store a file in a storage system including at least one server and at least two storages which can communicate with the at least one server through a communication network, a computer obtains a file name of a file to be stored in a storage. The computer determines an effect of deduplication of data of the file based on the file name, and responsive to determining the effect of deduplication exists, selects a part of the file name. The computer then calculates a hash value using the part of the file name and determines a storage to which to store the file based on the hash value.

Abstract: A storage system receives one or more records from a host system. The records are compressed in a first compression format that is native to the host system. The storage system identifies an incompatibility between the first compression format and a first operation of the storage system. In response to the identified incompatibility, the storage system decompresses the received records. The decompression is based on the first compression format. The storage system compresses the decompressed records in a second compression format. The storage system stores the secondarily compressed records onto a storage medium.

Abstract: An approach for dynamically changing a copy policy associated with a cluster grid for tape storage virtualization which involves monitoring one or more statuses of a plurality of clusters associated with the cluster grid, wherein the one or more of the plurality of clusters are assigned one of a plurality of copy modes, detecting that at least one of the plurality of clusters, having a highest priority copy mode associated with the cluster grid, is unavailable, selecting an available second cluster of the plurality of clusters and changing the copy policy wherein the available second cluster is assigned the highest priority copy mode.

Abstract: According to one embodiment there is provided a method for managing a location of an object in a storage system having hierarchical storage with a plurality of storage layers. The method includes transmitting a get-list request for viewing contents of a designated directory on the hierarchical storage. The method also includes receiving a response to the get-list request, the response including a list of objects in the designated directory and listing each storage layer corresponding to the objects. The list is output, and a select request to select a different storage layer for one of the objects is received. The select request is transmitted to the storage system for invoking migration of the selected object to the different storage layer. A second response is received from the storage system indicating the selected object is stored in the different storage layer, and an updated list is output.

Abstract: A computer-implemented method for determining a damaged area of a magnetic tape loaded in a tape drive, according to one embodiment, includes: determining a first error location on a magnetic tape where a first error occurred in response to experiencing the first error. Moreover, the method includes determining one or more areas on the magnetic tape to be examined. The one or more areas are determined using a predetermined algorithm which incorporates the first error location. Independent of a read and/or write operation, relative motion is induced between a tape head and the magnetic tape such that the tape head is positioned adjacent to each of the respective one or more areas in turn. Furthermore, each of the one or more areas having a respective number of measured servo errors which exceeds a threshold value is identified as a damaged area of the magnetic tape.

Abstract: Restoring a data processing system. Files on one or more storage devices of a data processing system are backed up to magnetic tape recording media. Metadata associated with the backed up files is recorded, including magnetic tape identification and tape location for each backed up file. Files on the one or more storage devices are identified that meet a selection policy for files required for rapid restoration of the data processing system. An index list is generated of the identified files and their associated magnetic tape metadata. The index list is sorted by magnetic tape identification and magnetic tape location. In response to a request to restore the data processing system, the identified files on the sorted index list are restored in magnetic tape identification and magnetic tape location sort order. Operation of the restored data processing system is then begun in accordance with the restored files.

Abstract: There is disclosed a method performed by a computer system for managing a location of an object in a hierarchical storage that includes a plurality of storage layers. In the method, the computer system receives a selection of one storage layer for storing a selected object from among the plurality of the storage layers via a network connected to the computer system. The computer system also moves the selected object to the one storage layer from other storage layer currently that stores the selected object among the plurality of the storage layers. The computer system further excludes the selected object from targets of a predetermined space management that is being applied to objects in the hierarchical storage.

Abstract: A method, a computer program product, and a computer system for backing up data in a backup storage space. A computer receives a request for backing up a second data file, wherein the second data file is created from a first data file. The computer determines whether a third data file satisfies one or more of predetermined conditions for backup; wherein the third data file is generated when the second data file is created from the first data file and the third data file comprises information for creating the second data file from the first data file. In response to determining that the third data file satisfies the one or more of the predetermined conditions for backup, the computer backs up the third data file, without backing up the second data file.

Abstract: A computer-implemented method, according to one embodiment, includes: detecting a first error while accessing a magnetic tape, determining a first error location on the magnetic tape where the first error occurred, determining one or more areas on the magnetic tape to be examined, independent of a read and/or write operation, inducing relative motion between a tape head and the magnetic tape such that the tape head is positioned adjacent to each of the respective one or more areas in turn, using the tape head to measure a number of servo errors that occur in each of the respective one or more areas, and identifying each of the one or more areas having a respective number of measured servo errors which exceeds a threshold value as a damaged area of the magnetic tape. The one or more areas are determined using a predetermined algorithm which incorporates the first error location.

Abstract: Aspects of the present disclosure relate to a method for timing index write to a tape medium of a tape system. Index write timing data that specifies index write timing based on file metadata attributes is stored. Metadata attributes of a first file are analyzed to determine whether a first index should be written at a first time based on the index write timing data. In response to determining that an index should be written, the first index is written to the tape medium at the first time.

Abstract: A system accesses metadata on a file system of the magnetic tape, where the metadata comprising one or more fields enabling to determine a longitudinal position (LPOS) of one or more files located on the magnetic tape. The system determines the LPOS of the one or more files located on the magnetic tape. The system determines an optimal location of the head on the magnetic tape based on computing an average value to the determined LPOS of the one or more files located on the magnetic tape and moves the head on the magnetic tape to the optimal location.

Abstract: A memory that includes a data partition and an index partition is mounted. The data partition includes an original data and an original index written directly after the original data. The index partition includes the original index at the beginning of the index partition. The first data is written to a data partition of the memory. The first data is written directly after the at least one data on the data partition and the original index is deleted and invalidated. Responsive to a time interval elapsing, a first index is written to the data partition of the memory directly after the first data. The time interval starts at the writing of the first data. Responsive to the time interval elapsing, the first index is written to the index partition of the memory at the closest location, in terms of movement time, to the first index on the data partition.

Abstract: A method, executed by at least one processor, includes receiving a migration request for a plurality of files, assigning files of the plurality of files that are smaller than a selected size to one or more small-file secondary storage devices, assigning files of the plurality of files that are larger than the selected size to one or more large-file secondary storage devices, and initiating migration of the plurality of files from one or more primary storage devices to the secondary storage devices to which they are assigned. A corresponding computer program product and computer system are also disclosed herein.

Abstract: Restoring a data processing system. Files on one or more storage devices of a data processing system are backed up to magnetic tape recording media. Metadata associated with the backed up files is recorded, including magnetic tape identification and tape location for each backed up file. Files on the one or more storage devices are identified that meet a selection policy for files required for rapid restoration of the data processing system. An index list is generated of the identified files and their associated magnetic tape metadata. The index list is sorted by magnetic tape identification and magnetic tape location. In response to a request to restore the data processing system, the identified files on the sorted index list are restored in magnetic tape identification and magnetic tape location sort order. Operation of the restored data processing system is then begun in accordance with the restored files.

Abstract: A method for operating a tape drive may include receiving a command to read a record stored on a tape by a tape drive having a capability to compress and decompress data. The tape drive compresses a record prior to writing the record to tape when a command to write the record to tape specifies that the record is to be compressed, and when a compressed size of the record is smaller than an uncompressed size of the record. The method further include reading the record from the tape by the tape drive, where the tape includes a plurality of compressed records and a plurality of uncompressed records. The method further includes determining by the tape drive whether the record is not compressed. The method may additionally include generating a compressed record by compressing the record by the tape drive when it is determined that the record is not compressed.

Abstract: Linear Tape File System that can accommodate various storage orders, such as sequential and reverse sequential, on a tape medium. The storage order may be reversed (or otherwise changed) at the block level and also the bit order within the block level. The LFTS related data associated with the LFTS file system stores data sufficient to indicate storage order (for example, reverse), so that the motions of the tape medium and tape drive can be coordinated during reading or erasing operations to be performed in the correct order, with a view to the order the data was stored on the tape. Storing data in reverse order on an LFTS tape can lead to certain operational advantages.

Abstract: A mechanism is provided for providing information about fragmentation of a file on a sequential access medium by a computer system is disclosed. An actual time for reading the file recorded on the sequential access medium is estimated based on a physical position of the file. A total length of the file on the sequential access medium is calculated based on a physical length of each data piece constituting the file. An expected time for reading the file by assuming that the file is rewritten continuously is estimated based on the total length of the file. Information about the fragmentation of the file is then provided based on the actual time and the expected time.