Abstract: Systems and methods for determining retention periods or policies for backups are disclosed. A rule book stores relationships between rules and recommended retention periods. Data related to a backup is collected and organized. A query is generated from the organized data and used to identify a rule from the rule book. The retention period corresponding to the identified rule in the rule book is then applied to the corresponding backup.

Abstract: A method for managing backups includes obtaining a plurality of parameters for a data item, filtering the plurality of parameters to obtain a plurality of filtered parameters, evaluating each filtered parameter of the plurality of filtered parameters to obtain a plurality of evaluated parameters, updating a promotion parameter register based on the plurality of evaluated parameters, and processing a backup request using the promotion parameter register, wherein the backup request specifies the data item.

Abstract: A method for performing backup operations includes obtaining, by a stream allocator operating in a backup storage system, a plurality of media management entries, wherein a media management entry of the plurality of media management entries specifies a data object, a relative priority, and a maximum number of streams, selecting the media management entry based on the relative priority, allocating a number of backup streams to the data object based on the maximum number of streams, and initiating a transfer of a backup of the data object from the production host to the backup storage system.

Abstract: A method for managing data includes obtaining, by a backup agent, a migration request, wherein the migration request specifies a plurality of users, and in response to the migration request: identifying a set of source-format data sets in a backup storage system, mounting the set of source-format data sets from the backup storage system, performing data mining on the set of source-format data sets in response to the mounting to obtain a hierarchical structure, generating a cloud-based service user account for each user of the plurality of users, wherein the cloud-based service user accounts are associated with a cloud-based service, obtaining source-format data sets associated with each user of the plurality of users using the hierarchical structure, and initiating a migration of the source-format data sets to the cloud-based service.

Abstract: A pass-through snapshot engine records lifecycle data and chronological status of files or objects, in one or more pass-through snapshots during a backup. If there is an abort, the pass-through snapshot engine freezes the pass-through snapshot(s) and terminates the backup. To restart after the abort, the pass-through snapshot engine sends one or more pass-through snapshots to the backup agent. The backup agent restarts the backup, and enforces states of lifecycle and chronology to files or objects, and discards files or objects that have already been backed up to the target device, in accordance with the one or more pass-through snapshots.

Abstract: For restoring data, a system with uninterrupted block-based restore has a hybrid container. The hybrid container has an operational buffer and a virtual container unit. The system receives blocks of data from a target device into the hybrid container and sends the blocks of data from the hybrid container to a destination device for a restore session. The system writes information about the blocks of data into one or more virtual containers. If there is an abort of the restore session, the system freezes state of the hybrid container and the virtual container(s). For a next restore session that resumes from where the aborted restore session stopped, the system resumes receiving blocks of data from the target device into the hybrid container and sends only leftover blocks of data to the destination device in accordance with the information in the virtual container(s).

Abstract: A method and system for prioritizing critical data object recovery during restore operations. Specifically, the method and system disclosed herein entail reordering data objects, awaiting being read from backup storage and thus queued in one or more data object queues, in accordance with a nearest-critical based sequential order. The nearest-critical based sequential order may be derived through modified weight-based Euclidean distances calculated between adjacent data object pairs queued in any given data object queue. Further, the calculated modified weight-based Euclidean distances incorporate data criticality factors associated with the adjacent data object pairs. By reordering data objects in a nearest-critical based sequential order, critical data objects may be recovered first, thereby avoiding possible critical data loss should a disaster occur during restore operations.

Abstract: A method for performing a backup operation includes obtaining, by a backup server, a backup request, wherein the backup request specifies a virtual machine to be backed up, wherein the virtual machine is hosted by a production host, and in response to the backup request: obtaining classification data from the backup agent, initiating a backup classification on an unprocessed backup associated with the virtual machine based on the classification data to obtain a sensitivity tag, and initiating a data processing on the unprocessed backup based on the sensitivity tag.

Abstract: A conditional construction container (CCC) provides seamless restore operation during abort event. During normal restore, the CCC assumes a hold state, wherein data blocks are sent from the backup source to the restore host directly. When interrupt occurs, the CCC assumes compression mode, wherein the CCC compresses the data blocks and stores the compressed blocks in memory allocated to the CCC. When the restore operation can resume, the backup agent instructs the CCC to begin data reconstruction operation and the CCC will assume busy mode no longer accepting new data blocks. During the busy state the CCC decompresses the compressed data blocks from the container and reconstruct the original data onto the restore host. Since the container constructed in the memory allocated to the CCC contain data that was obtained upon the interrupt, the restore operation resumes by reconstructing the data from the point of the interrupt seamlessly.

Abstract: Embodiments are described for performing an uninterrupted restore in a storage system in view of one or more abort events. A restore agent receives writes one or more data blocks to a conditional construction container. A parent interrupt service routine (ISR) polls for abort events. In response to an abort event, an intermediate interrupt is generated that spawns a child processes for each process of the restore. The intermediate ISR logs each child ISR, the process it is responsible for, and the intermediate interrupt, for later restoration of the restore state. After a recovery of the above event, then each child ISR can be called to restore its state. After restoring the state, the restore agent resumes the restore from where the abort event was detected. The child ISRs are re-entrant. If another abort event is detected, the restore state can again be saved and later resumed from that state.

Abstract: A method and system for prioritizing critical data object recovery during restore operations. Specifically, the method and system disclosed herein entail reordering data objects, awaiting being read from backup storage and thus queued in one or more data object queues, in accordance with a nearest-critical based sequential order. The nearest-critical based sequential order may be derived through modified weight-based Euclidean distances calculated between adjacent data object pairs queued in any given data object queue. Further, the calculated modified weight-based Euclidean distances incorporate data criticality factors associated with the adjacent data object pairs. By reordering data objects in a nearest-critical based sequential order, critical data objects may be recovered first, thereby avoiding possible critical data loss should a disaster occur during restore operations.

Abstract: A method and system for prioritizing critical data object storage during backup operations. Specifically, the method and system disclosed herein entail reordering data objects, awaiting being written to storage and thus queued in one or more data object queues, in accordance with a nearest-critical based sequential order. The nearest-critical based sequential order may be derived through modified weight-based Euclidean distances calculated between adjacent data object pairs queued in any given data object queue. Further, the calculated modified weight-based Euclidean distances incorporate data criticality factors associated with the adjacent data object pairs. By reordering data objects in a nearest-critical based sequential order, critical data objects may be written into storage first, thereby avoiding possible critical data loss should a disaster occur during backup operations.

Abstract: A method for managing backups includes obtaining a plurality of parameters for a data item, filtering the plurality of parameters to obtain a plurality of filtered parameters, evaluating each filtered parameter of the plurality of filtered parameters to obtain a plurality of evaluated parameters, updating a promotion parameter register based on the plurality of evaluated parameters, and processing a backup request using the promotion parameter register, wherein the backup request specifies the data item.

Abstract: A method, apparatus, and system for regulating resource usage of a backup application is disclosed. The operations comprise: continuously monitoring and obtaining system resource usage, process states, and backup policy parameters; generating a normalized band value based on the system resource usage, the process states, and the backup policy parameters; generating a straight line plot based on the band value, a total CPU value, and a deviation value; calculating a slope of the straight line plot, wherein the slope is utilized to generate a deviation value based further on a first value and a second value received from a user, and wherein the deviation value is fed back to the generation of the straight line plot to normalize the slope; generating control values based on the slope; and adjusting performance of the backup application based on the control values to regulate system resource usage of the backup application.

Abstract: Systems and methods for determining retention periods or policies for backups are disclosed. A rule book stores relationships between rules and recommended retention periods. Data related to a backup is collected and organized. A query is generated from the organized data and used to identify a rule from the rule book. The retention period corresponding to the identified rule in the rule book is then applied to the corresponding backup.

Abstract: A method, apparatus, and system for regulating resource usage of a backup application is disclosed. The operations comprise: continuously monitoring and obtaining system resource usage, process states, and backup policy parameters; generating a normalized band value based on the system resource usage, the process states, and the backup policy parameters; generating a straight line plot based on the band value, a total CPU value, and a deviation value; calculating a slope of the straight line plot, wherein the slope is utilized to generate a deviation value based further on a first value and a second value received from a user, and wherein the deviation value is fed back to the generation of the straight line plot to normalize the slope; generating control values based on the slope; and adjusting performance of the backup application based on the control values to regulate system resource usage of the backup application.

Abstract: A decoupled backup solution for distributed databases across a failover cluster. Specifically, a method and system disclosed herein improve upon a limitation of existing backup mechanisms involving distributed databases across a failover cluster. The limitation entails restraining backup agents, responsible for executing database backup processes across the failover cluster, from immediately initiating these aforementioned processes upon receipt of instructions. Rather, due to this limitation, these backup agents must wait until all backup agents, across the failover cluster, receive their respective instructions before being permitted to initiate the creation of backup copies of their relative distributed database. Subsequently, the limitation imposes an initiation delay on the backup processes, which the disclosed method and system omit, thereby granting any particular backup agent the capability to immediately (i.e., without delay) initiate those backup processes.