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.

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 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 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 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: 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: 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.

Abstract: The invention provides an improved method and apparatus for creating a snapshot of a file system. In a first aspect of the invention, a “copy-on-write” mechanism is used. An effective snapshot mechanism must be efficient both in its use of storage space and in the time needed to create it because file systems are often large. 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 (henceforth called “copy-on-write”). 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. In a second aspect of the invention, a record of which blocks are being used by the snapshot is included in the snapshot itself, allowing effectively instantaneous snapshot creation and deletion.

Abstract: A system and method are disclosed that provides transparent, global access to devices on a computer cluster. The present system generates unique device type (dev.sub.-- t) values for all devices and corresponding links between a global file system and the dev.sub.-- t values. The file system is modified to take advantage of this framework so that, when a user requests that a particular device, identified by its logical name, be opened, an operating system kernel queries the file system to determine that device's dev.sub.-- t value and then queries the a device configuration system (DCS) for the location (node) and identification (local address) of a device with that dev.sub.-- t value. Once it has received the device's location and identification, the kernel issues an open request to the host node for the device identified by the DCS.

Abstract: When a client computer requests data from a disk or similar device at a server computer, the client exports the memory associated with an allocated read buffer by generating and storing one or more incoming MMU (IMMU) entries that map the read buffer to an assigned global address range. The remote data read request, along with the assigned global address range is communicated to the server node. At the server, the request is serviced by performing a memory import operation, in which one or more outgoing MMU (OMMU) entries are generated and stored for mapping the global address range specified in the read request to a corresponding range of local physical addresses. The mapped local physical addresses in the server are not locations in the server's memory. The server then performs a DMA operation for directly transferring the data specified in the request message from the disk to the mapped local physical addresses.

Abstract: A system and method are disclosed for rendering devices on a cluster globally visible, wherein the cluster includes a plurality of nodes on which the devices are attached. The system establishes for each of the devices in the cluster at least one globally unique identifier enabling global access to the device. The system includes a device registrar that creates the identifiers and a global file system. The identifiers include a globally unique logical name by which users of the cluster identify the device and a globally unique physical name by which the global file system identifies the device. The registrar creates a one-to-one mapping between the logical name and the physical name for each of the devices. The system also includes a device information (dev.sub.-- info) data structure maintained by the device registrar that represents physical associations of the devices within the cluster. Each association corresponds to the physical name of a device file maintained by the global file system.

Abstract: When a client computer requests data from a disk or similar device at a server computer, the client exports the memory associated with an allocated read buffer by generating and storing one or more incoming MMU (IMMU) entries that map the read buffer to an assigned global address range. The remote data read request, along with the assigned global address range is communicated to the server node. At the server, the request is serviced by performing a memory import operation, in which one or more outgoing MMU (OMMU) entries are generated and stored for mapping the global address range specified in the read request to a corresponding range of local physical addresses. The mapped local physical addresses in the server are not locations in the server's memory. The server then performs a DMA operation for directly transferring the data specified in the request message from the disk to the mapped local physical addresses.