Abstract: A computer-implemented method for performing a storage reclaim on a storage device includes generating a second tape image on a first storage device. The computer-implemented method further includes copying metadata associated with a first data extent of a valid file from a first index partition of a first tape image to a second index partition of the second tape image. The computer-implemented method further includes copying file data associated with the first data extent from a first data partition of the first tape image to a second data partition of the second tape image. The computer-implemented method further includes adjusting metadata associated with the file data of the first data extent copied to the second data partition of the second tape image. The computer-implemented method further includes releasing a storage area corresponding to the first data extent of the valid file included in the first tape image.

Abstract: A request to open a file from a plurality of files in a storage is received from an application. The storage is made up of an index partition, containing recordings of file system indexes, and a data partition, containing recordings of the indexes and the file system data body. A file descriptor is created with an update flag that references the file. A determination is made that the file is being updated by the application, and the update flag is set to a value representing that the file is being updated. A request to write an index of the file system is received. A determination is made whether a specific file from the plurality of files is being updated. The index is written to the storage with an extended attribute for the specific file indicating that the specific file was being updated at the time the index was written.

Abstract: A linear tape file system (LTFS) maintains a series of indexes, and an incremental change dataset, that records incremental changes that have taken place in each new index over the previous one, as each new index of the series is created. Upon receiving a command to roll back the file system from a current generation to a target previous generation, a generation control module applies information in the incremental change dataset to the current index. In this way, the current index is restored, at least functionally, to the state it was in when the target generation was created, thereby effectuating the rollback of the file system to the target generation.

Abstract: A method for indexing a tape for unmounting from a tape drive is discussed. The method can determine that an update has been made to the tape since it was last mounted. The method can identify a tape head position with respect to the tape, the tape including a data partition and an index partition. The method can identify first position and a second position on the tape. The method can determine first distance, a length of tape between first position and tape head position. The method can determine second distance, a length of tape between first position and second position. The method can determine that first distance is less than second distance. The method can reposition the tape to a primary section and write the metadata index. The method can reposition tape to the second position write a data index beginning at the second position of the data partition.

Abstract: Provided is a method for writing and deleting files on a tape medium and a cache storage device. The method includes receiving a command to write one or more files of a directory to a tape medium. The method further includes identifying a cache limit associated with the tape medium. The method further comprises determining whether the amount of data of the directory that is already on the cache storage device exceeds the cache limit. In response to the amount of data not exceeding the cache limit, the method includes writing data of the one or more files to the cache storage device and the tape medium in parallel.

Abstract: A computer-implemented method for indexing a tape for unmounting from a tape drive. The computer-implemented method includes: receiving a request to unmount the tape from the tape drive, wherein the tape includes an index partition and a data partition, determining that an update to the tape is not reflected in any data index written in the data partition, appending a new data index at the end of the last written data file in the data partition, wherein the new data index includes information associated with the update to the tape, unthreading the tape from the tape drive, wherein unthreading the tape from the tape drive is performed at low tension writing a metadata index to an ancillary storage device, wherein the metadata index includes information associated with the update to the tape.

Abstract: Deep learning applications access modified files stored in high speed storage. The modified files are custom formatted for use by deep learning applications to minimize file size. Parent files from which the modified files are made are stored in low speed storage as files having the original characteristics of the parent files. Requesters identified as deep learning applications are provided access to the modified files in high speed storage. Requesting applications not identified as deep learning applications receive the parent files via conventional processing from low speed storage.

Abstract: A method includes setting a maximum total size of data saved in a snapshot area, creating a snapshot of a file system, including reference to a plurality of logical volumes, saving at least a part of the plurality of logical volumes in the snapshot area on condition that it does not exceed the maximum total size of data saved in the snapshot area, and executing copy-on-write processing in response to a writing request for data in the file system, the data not being saved in the snapshot area.

Abstract: A computer-implemented method, a computer program product, and a computer system for reordering a sequence of files based on compression rates in data transfer. A host determines predicted compression rates of first files that are to be saved onto a first tape, based on a relationship between compression rates and file attributes of second files that have been saved on a second tape. The host reorders a sequence of the first files stored in host cache storage and generates a new sequence of transferring the first files from the host cache storage to the tape drive hosting the first tape, based on the predicted compression rates of the first files. The host transfers the first files to the tape drive hosting the first tape and writes the first files to the first tape, according to the new sequence.

Abstract: A computer-implemented method, according to one embodiment, includes: receiving, from a tape drive, a first error location on a magnetic tape where a first error occurred, in addition to determining one or more areas on the magnetic tape to be examined based on the first error location. Independent of a read and/or write operation, the tape drive is instructed to induce 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. Moreover, 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: A computer program product, according to one embodiment, includes a computer readable storage medium having program instructions embodied therewith. The program instructions are readable and/or executable by a processor to cause the processor to: determine, by the processor, a first error location on a magnetic tape where a first error occurred. One or more areas on the magnetic tape to be examined are also determined, by the processor, based on the first error location. Independent of a read and/or write operation, relative motion between a tape head and the magnetic tape is induced by the processor, such that the tape head is positioned adjacent to each of the one or more areas in turn. Moreover, each of the one or more areas having a respective number of measured servo errors which exceeds a threshold value are identified, by the processor, as a damaged area of the magnetic tape.

Abstract: A computer-implemented method, a computer program product, and a computer system for optimizing timing for migrating data. The computer system sets a migration required flag for each of first tapes, in response to determining an estimated time period required for data migration from the first tapes to second tapes being smaller than an expiration time period of the first tapes. The computer system sets the migration required flag for a respective one of the first tapes, in response to determining the estimated time period being not smaller than the expiration time period and a predetermined time period has passed for the respective one of the first tapes. Data on tapes that migration required flags are set is to be migrated to the second tapes.

Abstract: A plurality of files are stored on magnetic tape having at least one wrap for storing data sets. Each data set includes one or more records. A number of data sets that will remain in a buffer memory after performing a read ahead operation is determined. A number of records that are likely to remain in the buffer memory after performing a read ahead operation is determined. For each record likely to remain in the buffer memory after performing a read ahead operation, positions of the records on the magnetic tape are determined. Each file of the plurality of files are classified into one group of a plurality of groups based on a position of a record corresponding with each respective file and the number of records that are likely to remain in the buffer memory after performing a read ahead operation.

Abstract: A computer-implemented method for faster unmount of tape drives includes generating a batch file to be migrated from a primary storage to a tape media acting as a secondary storage. The size of the batch file is determined based on an amount of available storage space on the tape media between a position in a longitudinal direction of storage (LPOS) and longitudinal position three (LP3) on the tape media. The computer-implemented method further includes migrating the batch file from the primary storage to the tape media mounted on a tape drive. Upon completion of migrating the batch file, the LPOS is within a predetermined threshold distance of LP3 on the tape media.

Abstract: A computer-implemented method, a computer program product, and a computer system for optimizing timing for migrating data. The computer system sets a migration required flag for each of first tapes, in response to determining an estimated time period required for data migration from the first tapes to second tapes being smaller than an expiration time period of the first tapes. The computer system sets the migration required flag for a respective one of the first tapes, in response to determining the estimated time period being not smaller than the expiration time period and a predetermined time period has passed for the respective one of the first tapes. Data on tapes that migration required flags are set is to be migrated to the second tapes.

Abstract: A computer-implemented method for faster unmount of tape drives includes generating a batch file to be migrated from a primary storage to a tape media acting as a secondary storage. The size of the batch file is determined based on an amount of available storage space on the tape media between a position in a longitudinal direction of storage (LPOS) and longitudinal position three (LP3) on the tape media. The computer-implemented method further includes migrating the batch file from the primary storage to the tape media mounted on a tape drive. Upon completion of migrating the batch file, the LPOS is within a predetermined threshold distance of LP3 on the tape media.

Abstract: Communicating data with a medium is provided. A cache is provided for storing target data of a file identified by an access request from an application of a host. The cache is divided into a read cache, a write cache, and an index cache. Responsive to receiving the access request: the medium is loaded onto a drive using a file system; target data is stored to the write cache and to the read cache; and the index file stored in the index cache is updated to reflect position metadata about the target data stored in the write cache. Responsive to initiating unloading of the medium from the drive: the updated index file stored in the index cache is written to the index partition of the medium; and the target data stored in the write cache is written onto a data partition of the medium without using the file system.

Abstract: Aspects of the present disclosure relate to managing index writes within a tape. A determination can be made that a change was made to at least one file stored within a tape medium while the tape medium was mounted within a tape drive. In response to the determination, an index can be written to a data partition of the tape medium. A determination can be made whether the change included any changes to a file at least equal to a threshold hierarchy level. In response to determining that the change did not include any changes to a file at least equal to the threshold hierarchy level, a first instruction can be issued to the tape drive to reposition the tape medium to the end of the tape. A second instruction can be issued to the tape drive to perform an unthread operation at low tension to rewind the tape medium.

Abstract: An approach, for storing video data in a plurality of tape media via a ring buffer. A tape data splitter receives video data stored in a ring buffer. The tape data splitter writes the video data to a current tape media. The tape data splitter detects an end of tape media event. The tape data splitter responds to the end of tape media event, by switching the current tape media to a next tape media. The tape data splitter selects a resume point for the video data stored in the ring buffer and writes to the next tape media, the video data stored in the ring buffer that is subsequent to the resume point.

Abstract: A computer-implemented method, a computer program product, and a computer system for reordering a sequence of files based on compression rates in data transfer. A host determines predicted compression rates of first files that are to be saved onto a first tape, based on a relationship between compression rates and file attributes of second files that have been saved on a second tape. The host reorders a sequence of the first files stored in host cache storage and generates a new sequence of transferring the first files from the host cache storage to the tape drive hosting the first tape, based on the predicted compression rates of the first files. The host transfers the first files to the tape drive hosting the first tape and writes the first files to the first tape, according to the new sequence.