Abstract: A write control method based on write behavior prediction, a memory storage device, and a memory control circuit unit are provided. The method includes: monitoring a first data write behavior of a host system during a first time range; according to the first data write behavior, predicting a second data write behavior of the host system during a second time range; obtaining a first measurement parameter and a first target parameter corresponding to the first data write behavior; according to the first measurement parameter, the first target parameter, and the second data write behavior, determining a write control parameter; and sending a write command sequence according to the write control parameter to instruct a rewritable non-volatile memory module to perform a data write based on multiple write modes during the second time range.

Abstract: A timer calibration method and an electronic device are disclosed. The method includes: performing a fitting operation according to a clock frequency of a clock device and an output of a timer to generate a fitting function; obtaining a first value output by the timer; and adjusting the first value to be a second value according to the fitting function to calibrate the timer.

Abstract: A risk assessment method based on data priority, a memory storage device, and a memory control circuit unit are provided. The method includes: receiving a query command from a host system; in response to the query command, performing a data health detection on a rewritable non-volatile memory module, wherein the rewritable non-volatile memory module stores data with multiple data priorities; generating risk assessment information according to a detection result, wherein the risk assessment information reflects a health degree of data with different data priorities in the rewritable non-volatile memory modules by different risk levels; and transmitting the risk assessment information to the host system.

Abstract: A timer calibration method and an electronic device are disclosed. The method includes: performing a fitting operation according to a clock frequency of a clock device and an output of a timer to generate a fitting function; obtaining a first value output by the timer; and adjusting the first value to be a second value according to the fitting function to calibrate the timer.

Abstract: A method is used in managing data accesses in storage systems. An access path to a storage object is evaluated. The storage object is associated with a first storage processor of a storage system for providing access to the storage object. The storage object is accessed using a second storage processor of the storage system. The storage system includes source and destination storage pools configured such that performance characteristics associated with the source storage pool is different from performance characteristics associated with the destination storage pool. The storage object resides in the source storage pool associated with the first storage processor. Based on the evaluation, the access path of the storage object is changed to the second storage processor based on a policy. The policy determines available storage space in the destination storage pool associated with the second storage processor for migrating the storage object to the destination storage pool.

Abstract: A method is used in managing data relocation in storage systems. Data access activity information is gathered for a set of slices of a storage tier in a data storage system for migrating the set of slices from the storage tier to another storage tier. The data storage system includes a first storage tier and a second storage tier configured such that performance characteristics associated with the first storage tier is superior to the second storage tier. Based on a pattern indicated by the data access activity information, a temperature for the set of slices is determined by applying a predictive analysis technique. Based on the determination, relocation of data is effected in the storage system.

Abstract: A technique for managing temporal data placement in data storage systems is disclosed. An application program interface (API) is provided, where a host application is configured to communicate sub-LUN data segment location, application configuration tiering, and timing parameter information to the API. The sub-LUN data segment location, application configuration tiering, and timing parameter information is transmitted to the data storage system via an out-of-band communications path. The sub-LUN data segment location, application configuration tiering, and timing parameter information is received at the data storage system and associated with the corresponding LUN. A timing estimate for implementing sub-LUN data segment relocation is generated based on system derived tiering information and recent relocation timing information.

Abstract: A method is used in managing data placement in storage systems. A portion of an allocated storage space is identified in a data storage system for managing data placement. The allocated storage space includes first and second set of slices. The first set of slices associated with the portion of the allocated storage space in the data storage system is identified. Data of the first set of slices is transferred to the second set of slices of the allocated storage space in the data storage system. The first set of slice is removed from the allocated storage space in the data storage system.

Abstract: A method is used in managing data relocation in storage systems. Metadata of a slice of a storage tier in a data storage system is evaluated for migrating the slice from the storage tier to another storage tier. The data storage system includes a first storage tier and a second storage tier configured such that performance characteristics associated with the first storage tier is superior to the second storage tier. Based on the evaluation, relocation of the slice of the storage tier is effected. The metadata of the slice indicates whether the slice includes user data.

Abstract: A method is used in managing data in a data storage system. A tiered storage pool is identified wherein the storage pool includes multiple tiers having multiple storage units. A wear indicator for each of the multiple storage units is monitored. A first storage unit having a wear indicator greater than a second storage unit is identified. High activity data on the first storage unit is also identified. The identified high activity data is migrated from the first storage unit to the second storage unit.

Abstract: Techniques are described for storing data. A command is issued from a client to a data storage system. The data storage system includes a plurality of storage tiers comprising a first storage tier of physical storage devices and a second storage tier of physical storage devices, wherein data stored on any physical storage device of the first storage tier is stored in an encrypted form and data stored on any physical storage device of the second storage tier is not stored in an encrypted form. The command includes a hint indicating whether data stored at a first logical address range of a first logical device is stored in an encrypted form. The command is received at the data storage system. First data written to the first logical device at the first logical address range is stored on one or more physical storage devices of any of said first storage tier and said second storage tier in accordance with the hint.

Abstract: A method is used in balancing I/O load on data storage systems. A slice is selected from a first storage container of a storage tier. The storage tier includes a second storage container. The slice is relocated from the first storage container of the storage tier to the second storage container of the storage tier, where I/O load of the first storage container is greater than I/O load of the second storage container.

Abstract: A method is used in managing storage space of storage tiers. Storage space utilization characteristics of a storage tier of a set of storage tiers in a storage system are evaluated. Based on the evaluation, reservation of headroom storage space in the storage tier is adjusted. The headroom storage space indicates an amount of the storage space reserved as unused storage space in the storage tier.

Abstract: Techniques are described for storing data. A plurality of storage tiers are provided including a first set and a second set of storage tiers of physical devices. Data stored on any physical device in the first set is stored in an encrypted form. Data stored on any physical device in the second set is not stored in an encrypted form. A first value is specified for a first setting that is any of a tiering preference and tiering requirement indicating that at least one data portion of a logical device is to be stored on physical device(s) of a storage tier storing data in an encrypted form. Responsive to specifying the first value as the first setting, the at least one data portion of the logical device currently stored on physical device(s) of the second set are relocated to physical device(s) of the first set.

Abstract: A method of verifying that a storage system is complying with best practices guide includes a software program running in host servers for determining the host server configuration parameters, and comparing the host server configuration to the storage system configuration. If a newly added best practices rule is found to be different than either the current configuration of the storage system or the host server configuration, then an alert may be generated, and sent to a storage system user or manager. In addition, a suggested change notice may be generated and transmitted to the storage system user or manager. There may also be a suggested change notice sent to the host server. Best practices rules may be stored in a data base and updated frequently.

Abstract: A storage processor of a data storage system includes a data relocation task scheduler that provides system limit control on the execution of data relocation requests. The data relocation task scheduler also provides fairness and concurrency limit enforcement by scheduling relocation requests from different provisioning domains, such as storage pools, in a fair manner. For example, the data relocation task scheduler includes stackable sets of pending task queues that store incoming data relocation requests. The data relocation task scheduler applies an iteration scheme to remove data relocation requests from the pending task queues thereby providing scheduling fairness among the queues and among the different provisioning domains within the data storage system.

Abstract: Systems and methods for processing user queries and identifying a set of documents relevant to the user query from a database using multi ranker search are described. In one implementation, the retrieved documents can be paired to form document pairs, or instance pairs, in a variety of combinations. Such instance pairs may have a rank order between them as they all have different ranks. A classifier, hyperplane, and a base ranker may be constructed for identifying the rank order relationships between the two instances in an instance pair. The base ranker may be generated for each rank pair. The systems use a divide and conquer strategy for learning to rank the instance pairs by employing multiple hyperplanes and aggregate the base rankers to form an ensemble of base rankers. Such an ensemble of base rankers can be used to rank the documents or instances.

Abstract: Systems and methods for processing user queries and identifying a set of documents relevant to the user query from a database using multi ranker search are described. In one implementation, the retrieved documents can be paired to form document pairs, or instance pairs, in a variety of combinations. Such instance pairs may have a rank order between them as they all have different ranks. A classifier, hyperplane, and a base ranker may be constructed for identifying the rank order relationships between the two instances in an instance pair. The base ranker may be generated for each rank pair. The systems use a divide and conquer strategy for learning to rank the instance pairs by employing multiple hyperplanes and aggregate the base rankers to form an ensemble of base rankers. Such an ensemble of base rankers can be used to rank the documents or instances.