Abstract: A system, method, and medium provide communication between a volume server and a volume owner in a distributed storage management system. A volume server may manage configuration information for a particular volume of storage. A volume owner may manage input/output (I/O) for the volume. The volume server may lease the configuration information to the volume owner using a communications protocol. The communications protocol comprises a plurality of operations for management of the lease. The volume owner may use the configuration information to manage the I/O for the volume.

Abstract: A system and method for data storage in an array. A system includes a client coupled to a storage subsystem. The storage subsystem comprises data storage locations addressable as rows and columns in an array. Each column comprises a separate storage device. Each row includes redundant data. For a given row, a coordinating storage device receives data from the client, coordinates computation and storage of redundant data, and forwards data to other storage devices. In response to receiving data targeted for storage in a given storage location, a non-volatile, temporary storage device that is associated with the separate storage device that includes the given storage location buffers the received data. The coordinating storage device conveys a write completion message to the client in response to detecting that the data has been buffered in the non-volatile, temporary storage devices. At least two storage devices are coordinating storage devices in separate rows.

Abstract: A system using consistent virtual addresses to communicate in cooperative multi-layer virtualization environments includes a volume server, one or more physical storage devices, a front-end layer and a back-end layer of virtualization participants. The volume server may be configured to aggregate storage in the one or more physical storage devices into a logical volume, and to provide configuration information on the logical volume to a front-end virtualization participant and to a back-end virtualization participant. The front-end virtualization participant may be configured to send a storage request including a target virtual address identifying a block of the logical volume to the back-end virtualization participant. The back-end virtualization participant may be configured to use the configuration information to identify a physical data block at a physical storage device corresponding to the target virtual address, and to perform a physical I/O operation on the physical data block.

Abstract: A method, system, computer system, and computer-readable medium for maintaining up-to-date, consistent copies of primary data without the need to replicate modified data when the data were modified as a result of an operation that is not an application-driven write operation captured during replication. Selected storage management operations are performed on the primary and secondary data stores at points in time when the data are the same to ensure that the data stored within the data stores remain consistent. These selected storage management operations include operations that produce modified data stored in the primary data store, where a portion of the modified data are not replicated to a secondary node. Other types of storage management operations are selected to be performed on both the primary and secondary data stores, where the operations do not directly change data in the primary data store, but may affect data stored in the primary data store.

Abstract: One goal of consistency interval replication is to achieve a consistent copy of data generated by independent streams of writes from nodes in a clustered/distributed environment. Two writes to the same block from different nodes may arrive at a replication target in a different order from the order in which they were written to primary storage. A consistency interval coordinator may analyze a list of blocks modified during a consistency interval to determine conflict blocks written to by two different nodes during the same consistency interval. Conflict resolution may involve a node reading data for a conflict block from primary storage and forwarding it to the replication target or a node completing a suspended in-progress write for the conflict block. Once the conflicts have been resolved, the replication target may checkpoint the data modified during the interval and nodes may resume writes to the conflict blocks for the new interval.

Abstract: A volume server for flexibly combining mirroring, striping and concatenation in virtual storage devices includes virtualization functionality and a layout table. The virtualization functionality may be configured to aggregate storage within one or more block devices into a logical volume. The layout table may include multiple extent entries, where each extent entry represents a mapping of an extent of block device storage to a virtual address range of the logical volume, including an indication of a stripe of the logical volume to which the extent is mapped. An overlap in virtual address range between a first and a second extent entry may indicate mirroring of the overlapping blocks at each extent. In response to a storage request, the virtualization functionality may obtain a block device address corresponding to the logical volume virtual address of the storage request from the layout table.

Abstract: Systems and methods for implementing volume sets in a storage system. According to a first embodiment, a system may include a volume server, a first and a second client computer system, and a plurality of physical block devices. The volume server may be configured to aggregate storage in the plurality of physical block devices into a plurality of logical volumes, wherein a given logical volume includes storage from at least two physical block devices, to distribute a first subset including at least two of the plurality of logical volumes to the first client computer system for input/output as a first volume set configured to be accessed as a single logical device, and to distribute a second subset including at least two of the plurality of logical volumes to the second client computer system for input/output as a second volume set configured to be accessed as a single logical device.

Abstract: A method, system, computer system, and computer-readable medium for maintaining up-to-date, consistent copies of primary data without the need to replicate modified data when the data were modified as a result of an operation that is not an application-driven write operation captured during replication. Selected storage management operations are performed on the primary and secondary data stores at points in time when the data are the same to ensure that the data stored within the data stores remain consistent. These selected storage management operations include operations that produce modified data stored in the primary data store, where a portion of the modified data are not replicated to a secondary node. Other types of storage management operations are selected to be performed on both the primary and secondary data stores, where the operations do not directly change data in the primary data store, but may affect data stored in the primary data store.

Abstract: Various embodiments of systems and methods are disclosed for tracking valid regions of a working volume. State information identifies which regions of the working volume are currently valid. When the volume is created, the state information can be initialized to a value that identifies all regions of the volume as being invalid. The invalid regions do not need to be synchronized, since there will not be any need to reconstruct the data within those regions to a particular value. Accordingly, volume initialization, which synchronizes redundant data (e.g., RAID parity or a mirrored copy) with application data in the invalid regions, can be delayed. As the volume is accessed by an application, the redundant data associated with the regions being accessed is synchronized, and the state information is updated to indicate that those regions are valid.

Abstract: After a first device (e.g., a host node in a shared data clustered system that stores a two-way mirrored volume) in a network creates or modifies a description of a layout for a data volume, the first device transmits separate copies of the data volume layout description to a pair of second devices, respectively, for storage in respective memories thereof. The first device may be a host node, and the pair of second devices may be first and second data storage systems. The first device and the pair of second devices are configured so that I/O transactions are transmitted between the first device and either of the pair of second devices. In other words, the first device is contained in a network layer that is different from the network layer that contains the pair of second devices.

Abstract: A system for write forwarding in a storage environment employing distributed virtualization includes a first and a second storage server providing access to a first and a second physical storage device, respectively, a virtual device server and a virtual device client. The virtual device server may be configured to aggregate storage in the first and second physical storage device into a virtual storage device and make the virtual storage device accessible to the virtual device client for I/O operations. An update requested by the virtual device client may require a first physical write operation at the first physical storage device and a second physical write operation at the second physical device. The virtual device client may be configured to send an update request to the first storage server, rather than to both the first and the second storage servers.

Abstract: Systems and methods for performing snapshots in a storage environment employing distributed block virtualization. In one embodiment, the system may include a volume server, a first and a second host computer system, and a plurality of physical block devices. The volume server may be configured to aggregate storage in the plurality of physical block devices into a plurality of logical volumes, where a particular logical volume includes storage from at least two physical block devices. The volume server may further be configured to make a first and a second subset of the logical volumes available to the first and second host computer systems for input/output, respectively. The first subset and the second subset may be at least partially nonoverlapping, and a second logical volume may be configured to store a snapshot of data stored in a first logical volume logical volume.

Abstract: Disclosed is an apparatus or method performed by a computer system for creating a hierarchy of data volumes. Each data volume in the hierarchy is a point-in-time (PIT) copy of another data volume in the hierarchy or a PIT copy of a data volume V. In one embodiment of the apparatus or method, the contents of a first data volume in the hierarchy can be refreshed to the contents of a second data volume in the hierarchy such that the first data volume becomes a PIT copy of the second data volume. Before the first data volume is fully refreshed to the contents of the second data volume, data of the first data volume can be read or modified.

Abstract: A method, system, and language to express storage requirements. The language provides keywords and rules corresponding to commands for configuring a set of storage devices to provide requested capabilities of a logical volume. The language also has keywords and constructs for defining capabilities. The language supports direct inheritance of a capability, where a template specifies another template that contains rules to be used to provide a given capability. The language also supports indirect inheritance of a capability, where a template requires a capability but does not provide an implementation of the capability. In addition, the language is processed to “merge” rules by selecting a single storage device that conforms to more than one rule when possible. Merging rules enables a minimum number of storage devices to be used to meet a given logical volume configuration and set of capabilities.

Abstract: Systems and methods for performing recoverable mirroring in a storage environment employing asymmetrically distributed block virtualization. In one embodiment, the system may include a volume server, a first and a second host computer system, and a plurality of physical block devices. The volume server may be configured to aggregate storage in the plurality of physical block devices into a plurality of logical volumes, where a particular logical volume includes storage from at least two physical block devices, to make a first subset of the logical volumes available to the first host computer system for input/output, and to make a second subset of the logical volumes available to the second host computer system for input/output. The first subset and the second subset may be at least partially nonoverlapping, a given logical volume may be configured as a mirrored logical volume including a plurality of copies of a given data block.

Abstract: Various systems and methods for performing coordinated distributed write logging are provided. A method may involve one of several hosts, each of which has an associated log, requesting permission to perform a write to data in a storage volume from a coordinator. The coordinator coordinates access to the storage volume between the hosts. Prior to receipt of a response from the coordinator granting permission to perform the write to the storage volume, the host may initiate logging the write data for the write to a respective log. The host may signal completion of the write to an application that initiated the write in response to both logging the write data to the respective log and receiving the response from the coordinator. The host may perform one or more underlying block operations to update the storage volume according to the write subsequent to signaling completion of the write to the application that initiated the write.

Abstract: A system employing asymmetric distributed block virtualization includes a volume server, a plurality of volume clients, and one or more physical block devices. The volume server aggregates storage in the physical block devices into a logical volume and makes the logical volume accessible to the volume clients for input/output (I/O) operations. In order to manage different I/O access requirements (such as read-only access versus read-write access) of the volume clients, the volume server maintains a different I/O access policy for each volume client to control the kinds of input/output operations that the volume client is allowed to perform on the logical volume.

Abstract: A method for coordinating access to a storage volume between several hosts, each of which has an associated log, may involve: one of the hosts requesting permission from a coordinator to perform a write to a region of the storage volume; the host storing write data for the write to an associated log; and the coordinator sending a response granting permission to perform the write to the host, where the response indicates whether an old version of data in the region of the storage volume should be relocated to a COW (Copy On Write) snapshot volume. The host may signal completion of the write to an application that initiated the write prior to performing the write to the region of the storage volume and subsequent to both storing the write data to the associated log and receiving the response from the coordinator.

Abstract: Various embodiments of systems and methods are disclosed for initially synchronizing redundant data (e.g., a mirror, a replica, or a set of parity information) with an original volume. State information identifies which regions of the original volume are currently valid, and only valid regions of the original volume are used to generate the values of the redundant data during the initial synchronization. For example, if the redundant data is a set of parity information, synchronizing the redundant data involves calculating one or more parity values based on the valid regions of the volume. If the redundant data is a duplicate copy (e.g., a mirror or replica) of the volume, synchronizing the redundant data involves copying the valid regions of the volume to the duplicate copy of the volume. If the original volume includes any invalid regions, unnecessary copying and/or processing for those regions can be avoided during the initial synchronization.

Abstract: A virtual copy of data stored in a first memory is created in a second memory. Creating the virtual copy includes, in one embodiment, creating first and second tables in memory each one of which comprises a plurality of multibit entries. Each entry of the first table corresponds to a respective memory region of the first memory. Each entry of the second table corresponds to a respective memory region of the second memory. The first bit of the first and second tables indicates whether the corresponding memory region of the first and second memories, respectively, contains valid data. The second bit of the first and second tables indicates whether data in the corresponding memory region of the first and second memories, respectively, has been modified since the creation of the first and second tables, respectively.