Abstract: The present invention implements an I/O task architecture in which an I/O task requested by the storage manager, for example a stripe write, is decomposed into a number of lower-level asynchronous I/O tasks that can be scheduled independently. Resources needed by these lower-level I/O tasks are dynamically assigned, on an as-needed basis, to balance the load and use resources efficiently, achieving higher scalability. A hierarchical order is assigned to the I/O tasks to ensure that there is a forward progression of the higher-level I/O task and to ensure that resources do not become deadlocked.

Abstract: The invention provides an improved method and apparatus for creating a snapshot of a file system. A “copy-on-write” mechanism is used. The snapshot uses the same blocks as the active file system until the active file system is modified. Whenever a modification occurs, the modified data is copied to a new block and the old data is saved. In this way, the snapshot only uses space where it differs from the active file system, and the amount of work required to create the snapshot is small. A record of which blocks are being used by the snapshot is included in the snapshot itself, allowing effectively instantaneous snapshot creation and deletion. A snapshot can also be deleted instantaneously simply by discarding its root inode.

Abstract: The present invention implements an I/O task architecture in which an I/O task requested by the storage manager, for example a stripe write, is decomposed into a number of lower-level asynchronous I/O tasks that can be scheduled independently. Resources needed by these lower-level I/O tasks are dynamically assigned, on an as-needed basis, to balance the load and use resources efficiently, achieving higher scalability. A hierarchical order is assigned to the I/O tasks to ensure that there is a forward progression of the higher-level I/O task and to ensure that resources do not become deadlocked.

Abstract: The invention provides a method and system for persistent context-based behavior injection in a computing system, such as in a redundant storage system or another system having a layered or modular architecture. Behaviors that are injected can be specified to have triggering conditions, such that the behavior is not injected unless the conditions are true. Triggering conditions may include a selected ordering of conditions and a selected context for each behavior. In a system having a layered architecture, behavior injection might be used to evaluate correct responses in the face of cascaded errors in a specific context or thread, other errors that are related by context, concurrent errors, or multiple errors. Behavior injection uses non-volatile memory to preserve persistence of filter context information across possible system errors, for reporting of the results of behavior injection, and to preserve information across recovery from system errors. Multiple behavior injection threads are also provided.

Abstract: The invention provides flexible disabling of disk sets. One or more disks in a RAID sub-system may be identified as temporarily inactive. The disk or disks are then marked as inactive by setting one of a set of bits associated with each disk in the RAID subsystem. If an inactivated disk is a data disk, marking it as inactive also marks it as read only. If an inactivated disk is a parity disk, the RAID group to which it supplies parity is also inactivated and a file system must look to a mirror of the inactivated RAID sub-system for its data. When a data disk is reactivated it is marked as read/write by clearing its associated bit. When a parity disk is reactivated it is also marked as read/write by clearing its bit, however, it is not available for use until it has synchronized its operation with its mirror.

Abstract: A technique coherently suspends input/output (I/O) operations in a RAID subsystem of a storage system. A configuration tree of the RAID subsystem has a plurality of objects representing a logical configuration of storage devices coupled to the system. According to the technique, a “freeze” condition may be imposed on an object of the configuration tree to suspend I/O operations directed to that object. In order to freeze, I/O operations underway (“in flight”) in the RAID subsystem and directed to the object need to complete sufficiently so as to reach a recoverable state in the event the subsystem subsequently fails prior to an I/O restart procedure. Once a freeze condition has been imposed, new I/O requests directed to the object are inserted onto a freeze list of pending requests at the RAID subsystem and are blocked from processing until the object is “unfrozen” (i.e., the freeze condition is lifted).

Abstract: The present invention implements an I/O task architecture in which an I/O task requested by the storage manager, for example a stripe write, is decomposed into a number of lower-level asynchronous I/O tasks that can be scheduled independently. Resources needed by these lower-level I/O tasks are dynamically assigned, on an as-needed basis, to balance the load and use resources efficiently, achieving higher scalability. A hierarchical order is assigned to the I/O tasks to ensure that there is a forward progression of the higher-level I/O task and to ensure that resources do not become deadlocked.

Abstract: The invention features a method for controlling storage of data in a plurality of storage devices each including storage blocks, for example, in a RAID array. The method includes receiving a plurality of write requests associated with data, and buffering the write requests. A file system defines a group of storage blocks, responsive to disk topology information. The group includes a plurality of storage blocks in each of the plurality of storage devices. Each data block of the data to be written is associated with a respective one of the storage blocks, for transmitting the association to the plurality of storage devices.

Abstract: A storage operating system is configured to assign volume block numbers (VBNs) to a volume. The system has a plurality of disks, and each disk of the plurality of disks is assigned disk block numbers (DBNs). A raidmap is configured to map the VBNs to the DBNs of the plurality of physical disks, the mapping for a particular disk stored in a disk label for the particular disk. The disk label for the particular disk is then written to the particular disk.

Abstract: In one embodiment, a first storage device and a second storage device form a mirror. When the first storage device loses synchronization with the second storage device, data present in the second storage device but not in the first storage device are identified. The identified data are then copied to the first storage device. In one embodiment, a method of rebuilding data in a storage device includes the act of replacing a failed storage device with a replacement storage device. Up-to-date data for the failed storage device, which may be stored in a corresponding mirror, may then be copied to the replacement storage device. Thereafter, the replacement storage device and any other storage devices that have lost synchronization with their mirror are resynchronized.

Abstract: A method for precognitive fetching, involving receiving an original request, performing pre-fetching analysis using the original request to obtain a pre-fetch request, forwarding the pre-fetch request to a storage subsystem, and receiving a response to the pre-fetch request from the storage subsystem.

Abstract: A technique maps the capacity of storage devices, such as disks, into any RAID group of a volume of a storage system regardless of the location of the RAID group within a volume block number (VBN) space of the volume. The technique separates disks and mapped VBN ranges, allowing for flexibility in the description and extension of RAID group capacities, while providing disk addition policies that support location-independent disk insertion into RAID groups. The technique also provides a disk label structure that supports the provision of multiple VBN ranges within a RAID group and within individual disks. Moreover, the technique provides file system support for allocation and topology management of the multiple mapped VBN ranges within disks and RAID groups, as well as noncontiguous VBN ranges across the RAID groups in the volume.

Abstract: The invention provides a method and system for recovery of file system data in file servers having mirrored file system volumes. The invention makes use of a “snapshot” feature of a robust file system (the “WAFL File System”) disclosed in the Incorporated Disclosures, to rapidly determined which of two or more mirrored volumes is most up-to-date, and which file blocks of the most recent mirrored volume have been changed from each one of the mirrored file systems. In a preferred embodiment, among a plurality of mirrored volumes, the invention rapidly determines which is the most up-to-date by examining a consistency point number maintained by the WAFL File System at each mirrored volume. The invention rapidly pairwise determines what blocks are shared between that most up-to-date mirrored volume and each other mirrored volume, in response to a snapshot of the file system maintained at each mirrored volume and are stored in common pairwise between each mirrored volume and the most up-to-date mirrored volume.

Abstract: An improved method and apparatus for creating a snapshot of a file system. A record of which blocks are being used by a snapshot is included in the snapshot itself, allowing effectively instantaneous snapshot creation and deletion. The state of the active file system is described by a set of metafiles; in particular, a bitmap (henceforth the “active map”) describes which blocks are free and which are in use. The inode file describes which blocks are used by each file, including the metafiles. The inode file itself is described by a special root inode, also known as the “fsinfo block”. The system begins creating a new snapshot by making a copy of the root inode. This copy of the root inode becomes the root of the snapshot.

Abstract: The invention provides a method and system for persistent context-based behavior injection in a computing system, such as in a redundant storage system or another system having a layered or modular architecture. Behaviors that are injected can be specified to have triggering conditions, such that the behavior is not injected unless the conditions are true. Triggering conditions may include a selected ordering of conditions and a selected context for each behavior. In a system having a layered architecture, behavior injection might be used to evaluate correct responses in the face of cascaded errors in a specific context or thread, other errors that are related by context, concurrent errors, or multiple errors. Behavior injection uses non-volatile memory to preserve persistence of filter context information across possible system errors, for reporting of the results of behavior injection, and to preserve information across recovery from system errors. Multiple behavior injection threads are also provided.

Abstract: A method and apparatus for a reliable data storage system using block level checksums appended to data blocks. Files are stored on hard disks in storage blocks, including data blocks and block-appended checksums. The block-appended checksum includes a checksum of the data block, a VBN, a DBN, and an embedded checksum for checking the integrity of the block-appended checksum itself. A file system includes file blocks with associated block-appended checksum to the data blocks. The file blocks with block-appended checksums are written to storage blocks. In a preferred embodiment a collection of disk drives are formatted with 520 bytes of data per sector. For each 4,096-byte file block, a corresponding 64-byte block-appended checksum is appended to the file block with the first 7 sectors including most of the file block data while the 8th sector includes the remaining file block data and the 64-byte block-appended checksum.

Abstract: The invention provides flexible disabling of disk sets. One or more disks in a RAID subsystem may be identified as temporarily inactive. The disk or disks are then marked as inactive by setting one of a set of bits associated with each disk in the RAID subsystem. If an inactivated disk is a data disk, marking it as inactive also marks it as read only. If an inactivated disk is a parity disk, the RAID group to which it supplies parity is also inactivated and a file system must look to a mirror of the inactivated RAID subsystem for its data. When a data disk is reactivated it is marked as read/write by clearing its associated bit. When a parity disk is reactivated it is also marked as read/write by clearing its bit, however, it is not available for use until it has synchronized its operation with its mirror.

Abstract: The invention features a method for controlling storage of data in a plurality of storage devices each including storage blocks, for example, in a RAID array. The method includes receiving a plurality of write requests associated with data, and buffering the write requests. A file system defines a group of storage blocks, responsive to disk topology information. The group includes a plurality of storage blocks in each of the plurality of storage devices. Each data block of the data to be written is associated with a respective one of the storage blocks, for transmitting the association to the plurality of storage devices.

Abstract: The invention provides a method and system for reducing RAID parity computation following a RAID subsystem failure. Ranges of RAID stripes are assigned to bits in a bitmap that is stored on disk. When writes to the RAID are in progress, the bit associated with the range of stripes in the bitmap is set. When a failure occurs during the write process, the bitmap is analyzed on reboot to determine which ranges of stripes where in the process of being written, and the parity data for only those ranges of stripes is recomputed. Efficiency is increased by use of an in-memory write counter that tracks multiple writes to each stripe range. Using the write counter, the bitmap is written to disk only after each cycle of its associated bitmap bit being set to a value of 1 and then returning to zero. The invention may be installed, modified, and removed at will from a RAID array, and this may be accomplished while the system is in operation.

Abstract: The invention provides a method and system for recovery of file system data in file servers having mirrored file system volumes. The invention makes use of a “snapshot” feature of a robust file system (the “WAFL File System”) disclosed in the Incorporated Disclosures, to rapidly determined which of two or more mirrored volumes is most up-to-date, and which file blocks of the most recent mirrored volume have been changed from each one of the mirrored file systems. In a preferred embodiment, among a plurality of mirrored volumes, the invention rapidly determines which is the most up-to-date by examining a consistency point number maintained by the WAFL File System at each mirrored volume.