Abstract: Embodiments are directed to efficiently managing data storage and efficiently storing data. In one scenario, a computer system receives a write request addressed to one portion of data storage within a data store. The computer system redirects the received write request to another portion of data storage which includes storage containers, each of which is a logically specified portion of the data store. The computer system determines which storage container the write request is to write to, and identifies a storage container that is to be de-staged from the second portion of data storage to the first portion of data storage. The computer system then de-stages the identified storage containers to the first portion of data storage, the data of the de-staged container being stored on the first portion of data storage in the same order in which the data was stored in the identified storage container prior to de-staging.

Abstract: In one embodiment of the present invention a cache unit organizes data stored in an attached memory to optimize accesses to compressed data. In operation, the cache unit introduces a layer of indirection between a physical address associated with a memory access request and groups of blocks in the attached memory. The layer of indirection—virtual tiles—enables the cache unit to selectively store compressed data that would conventionally be stored in separate physical tiles included in a group of blocks in a single physical tile. Because the cache unit stores compressed data associated with multiple physical tiles in a single physical tile and, more specifically, in adjacent locations within the single physical tile, the cache unit coalesces the compressed data into contiguous blocks. Subsequently, upon performing a read operation, the cache unit may retrieve the compressed data conventionally associated with separate physical tiles in a single read operation.

Abstract: Scalable data storage techniques are described. In one or more implementations, data is obtained by one or more computing devices that describes fault domains in a storage hierarchy and available storage resources in a data storage pool. Operational characteristics are ascertained, by the one or more computing devices, of devices associated with the available storage resources within one or more levels of the storage hierarchy. Distribution of metadata is assigned by the one or more computing devices to one or more particular data storage devices within the data storage pool based on the described fault domains and the ascertained operational characteristics of devices within one or more levels of the storage hierarchy.

Abstract: A system for storing data in a dynamic fashion. The system includes a storage entity. The storage entity includes portions of a plurality of different persistent storage devices. Each storage device has a set of constraints. The storage entity is configured to store data in a dynamic fashion in a layout on the persistent storage devices of the storage entity that meets the different data requirements for the data while still being within the constraints for the persistent storage devices. The storage entity is configured to change the layout for a portion of the data as requirements related to at least one of performance or resiliency for a portion of the data change while the storage entity continues to provide the data from the storage entity.

Abstract: A system architecture conserves memory bandwidth by including compression utility to process data transfers from the cache into external memory. The cache decompresses transfers from external memory and transfers full format data to naive clients that lack decompression capability and directly transfers compressed data to savvy clients that include decompression capability. An improved compression algorithm includes software that computes the difference between the current data word and each of a number of prior data words. Software selects the prior data word with the smallest difference as the nearest match and encodes the bit width of the difference to this data word. Software then encodes the difference between the current stride and the closest previous stride. Software combines the stride, bit width, and difference to yield final encoded data word. Software may encode the stride of one data word as a value relative to the stride of a previous data word.

Abstract: Embodiments are directed to efficiently managing data storage and efficiently storing data. In one scenario, a computer system receives a write request addressed to one portion of data storage within a data store. The computer system redirects the received write request to another portion of data storage which includes storage containers, each of which is a logically specified portion of the data store. The computer system determines which storage container the write request is to write to, and identifies a storage container that is to be de-staged from the second portion of data storage to the first portion of data storage. The computer system then de-stages the identified storage containers to the first portion of data storage, the data of the de-staged container being stored on the first portion of data storage in the same order in which the data was stored in the identified storage container prior to de-staging.

Abstract: Embodiments are directed to modifying storage capacity within a data store and to modifying resiliency for a data store. In one scenario, a computer system receives a request to move data. The computer system may determine that data is to be moved from an allocation on one data store to a new allocation on another data store. The computer system may create a new allocation on the other data store, where the new allocation is configured to receive data from the first data store. The computer system then moves the data to the new allocation on the second data store as data I/O requests are received at the first data store. Data store access requests are synchronized with the data movement by directing the data store access requests to the first data store, to the second data store or to both data stores depending on the type of access request.

Abstract: A system for storing data in a dynamic fashion. The system includes a storage entity. The storage entity includes portions of a plurality of different persistent storage devices. Each storage device has a set of constraints. The storage entity is configured to store data in a dynamic fashion in a layout on the persistent storage devices of the storage entity that meets the different data requirements for the data while still being within the constraints for the persistent storage devices. The storage entity is configured to change the layout for a portion of the data as requirements related to at least one of performance or resiliency for a portion of the data change while the storage entity continues to provide the data from the storage entity.

Abstract: Recovery requests are scheduled and prioritized according to priority valuations of the minimum time to next failure relative to the minimum time to next recovery for corresponding storage data sets. The prioritization can be performed by an upper layer recovery scheduler that dispatches requests to different storage units and/or by lower layers and individual storage unit schedulers. Prioritizations can be reflected in tagging on recovery requests, as determined by a first entity, and/or determined dynamically at the point of dispatch and processing.

Abstract: Scalable data storage techniques are described. In one or more implementations, data is obtained by one or more computing devices that describes fault domains in a storage hierarchy and available storage resources in a data storage pool. Operational characteristics are ascertained, by the one or more computing devices, of devices associated with the available storage resources within one or more levels of the storage hierarchy. Distribution of metadata is assigned by the one or more computing devices to one or more particular data storage devices within the data storage pool based on the described fault domains and the ascertained operational characteristics of devices within one or more levels of the storage hierarchy.

Abstract: In one embodiment of the present invention a cache unit organizes data stored in an attached memory to optimize accesses to compressed data. In operation, the cache unit introduces a layer of indirection between a physical address associated with a memory access request and groups of blocks in the attached memory. The layer of indirection—virtual tiles—enables the cache unit to selectively store compressed data that would conventionally be stored in separate physical tiles included in a group of blocks in a single physical tile. Because the cache unit stores compressed data associated with multiple physical tiles in a single physical tile and, more specifically, in adjacent locations within the single physical tile, the cache unit coalesces the compressed data into contiguous blocks. Subsequently, upon performing a read operation, the cache unit may retrieve the compressed data conventionally associated with separate physical tiles in a single read operation.

Abstract: In one embodiment of the present invention a cache unit organizes data stored in an attached memory to optimize accesses to compressed data. In operation, the cache unit introduces a layer of indirection between a physical address associated with a memory access request and groups of blocks in the attached memory. The layer of indirection—virtual tiles—enables the cache unit to selectively store compressed data that would conventionally be stored in separate physical tiles included in a group of blocks in a single physical tile. Because the cache unit stores compressed data associated with multiple physical tiles in a single physical tile and, more specifically, in adjacent locations within the single physical tile, the cache unit coalesces the compressed data into contiguous blocks. Subsequently, upon performing a read operation, the cache unit may retrieve the compressed data conventionally associated with separate physical tiles in a single read operation.

Abstract: One embodiment of the present invention sets forth a technique for processing commands received by an intermediary cache from one or more clients. The technique involves receiving a first write command from an arbiter unit, where the first write command specifies a first memory address, determining that a first cache line related to a set of cache lines included in the intermediary cache is associated with the first memory address, causing data associated with the first write command to be written into the first cache line, and marking the first cache line as dirty.

Abstract: A system for storing data in a dynamic fashion. The system includes a storage entity. The storage entity includes portions of a plurality of different persistent storage devices. Each storage device has a set of constraints. The storage entity is configured to store data in a dynamic fashion in a layout on the persistent storage devices of the storage entity that meets the different data requirements for the data while still being within the constraints for the persistent storage devices. The storage entity is configured to change the layout for a portion of the data as requirements related to at least one of performance or resiliency for a portion of the data change while the storage entity continues to provide the data from the storage entity.

Abstract: A system in which a file system may operate on a volume in which the logical address extent of the volume is divided into multiple tiers, each tier providing storage having a distinct trait set by mapping the logical addresses of the volume to appropriate underlying storage systems. A volume system exposes the volume to the file system in a manner that the file system itself has awareness of the tiers, and is aware of the trait sets of each tier. The file system may thus store file system namespaces (such as directories and files) into the tiers as appropriate for the file system namespace. A provisioning system may also be provided and be configured to provision the volume to include such tiers, and if desired, to extend the tiers.

Abstract: Some implementations may include a virtual storage system to which data is written. The virtual storage system may include a cache and multiple hard drives. Multiple queues may be associated with the multiple hard drives such that each hard drive of the multiple hard drives has a corresponding queue of the multiple queues. A set of candidate rows may be selected from the cache. For each candidate row in the set of candidate rows, destination hard drives may be identified. Each candidate row may be placed in queues corresponding to the destination hard drives. Two or more candidate rows from the multiple queues may be written substantially contemporaneously (e.g., in parallel) to two or more destination hard drives.

Abstract: A method and a processing device are provided for sequentially aggregating data to a write log included in a volume of a random-access medium. When data of a received write request is determined to be suitable for sequentially aggregating to a write log, the data may be written to the write log and a remapping tree, for mapping originally intended destinations on the random-access medium to one or more corresponding entries in the write log, may be maintained and updated. At time periods, a checkpoint may be written to the write log. The checkpoint may include information describing entries of the write log. One or more of the checkpoints may be used to recover the write log, at least partially, after a dirty shutdown. Entries of the write log may be drained to respective originally intended destinations upon an occurrence of one of a number of conditions.

Abstract: This document describes techniques for storing virtual disk payload data. In an exemplary configuration, each virtual disk extent can be associated with state information that indicates whether the virtual disk extent is described by a virtual disk file. Under certain conditions the space used to describe a virtual disk extent can be reclaimed and state information can be used to determine how read and/or write operations directed to the virtual disk extent are handled. In addition to the foregoing, other techniques are described in the claims, figures, and detailed description of this document.

Abstract: A file system independent virtualized boot block with discovery volume and cover files renders a volume visible when accessed by an accessing system which differs from a source system. For example, a downlevel operating system recognizes that data is present on a volume created in an uplevel operating system, even where the uplevel data itself may not be accessible.

Abstract: A method and a processing device are provided for sequentially aggregating data to a write log included in a volume of a random-access medium. When data of a received write request is determined to be suitable for sequentially aggregating to a write log, the data may be written to the write log and a remapping tree, for mapping originally intended destinations on the random-access medium to one or more corresponding entries in the write log, may be maintained and updated. At time periods, a checkpoint may be written to the write log. The checkpoint may include information describing entries of the write log. One or more of the checkpoints may be used to recover the write log, at least partially, after a dirty shutdown. Entries of the write log may be drained to respective originally intended destinations upon an occurrence of one of a number of conditions.