Abstract: Diagnosis of corruption in interrelated data entities uses a graph of nodes and edges. Datum nodes represent the data entities, relationship nodes represent the relationships among the data entities. The datum nodes are connected through their relationship nodes by the edges. When corruption is detected, the relationships are analyzed and each edge connecting a datum node to a relationship node is removed from the graph when the corresponding relationship is invalid. The datum nodes that remain connected to their relationship nodes form a subgraph and the corresponding data entities are considered correct. In one aspect, if more than one subgraph is formed, the datum nodes in the largest are used. In another aspect, the data entities and relationships are analyzed to create the graph when the data entities are assumed correct. The data entities may be data and metadata of various types that can be associated with the data.

Abstract: A method for detecting errors in streaming media devices is described. In one embodiment, when a command to write a block of data to a streaming media device is received, integrity metadata associated with the data block is attached to the data block and written to the streaming media device together with the data block. Subsequently, when a read command pertaining to this data block is received, new integrity metadata is determined and compared to the attached metadata. If the new integrity metadata does not match the attached metadata, an error message is generated.

Abstract: A method for storing integrity metadata in a data storage system disk array. Integrity metadata is determined for each data stripe unit of a stripe in a disk array employing striped parity architecture. The number of physical sectors required to store the integrity metadata is determined. Sufficient data storage space, adjacent to the data stripe unit containing parity data for the stripe, is allocated for the storage of integrity metadata. The integrity metadata is stored next to the parity data. For one embodiment, a RAID 5 architecture is extended so that integrity metadata for each stripe is stored adjacent to the parity data for each stripe.

Abstract: A storage system may include a plurality of storage devices each having a plurality of addressable locations for storing data. A storage controller may be coupled to the storage devices and configured to store and retrieve data from the storage devices. An indirection map may be stored within the system having a plurality of map entries each configured to map a virtual address to a physical address on the storage devices. Each map entry may also store a checksum for data stored at the physical address indicated by the map entry. The storage controller may receive storage requests specifying a virtual address and may access the indirection map for each storage request to obtain the corresponding physical address and checksum. Dynamic striping may be employed so that new writes form new parity groups. Thus, stripes of various sizes may be supported by the storage system.

Abstract: A storage array interconnection fabric may be configured using a torus topology. A storage system including a path-redundant torus interconnection fabric is coupled to a plurality of nodes. The torus interconnection fabric may be configured to connect the plurality of nodes in an array including N rows and M columns, where N and M are positive integers. The array may be configured such that a first node in a first row of the N rows is connected to a second node in the first row and a first node in a first column of the M columns is connected to a second node in the first column. Also an ending node in the first row is connected to the first node in the first row and an ending node in the first column is connected to the first node in the first column. In addition, a first portion of the plurality of nodes is configured to communicate with a plurality of storage devices such as disk drives.

Abstract: A method for detecting a phantom write error in a data storage system is described. In one embodiment, upon receiving a read command pertaining to a data block stored on a storage medium, two version identifiers associated with the data block are compared. A first version identifier is stored within the data block and a second version identifier is stored outside of the data block. If the version identifiers do not match, the possible occurrence of a phantom write error is detected.

Abstract: A storage array interconnection fabric may be configured using a torus topology. A storage system including a path-redundant torus interconnection fabric is coupled to a plurality of nodes. The torus interconnection fabric may be configured to connect the plurality of nodes in an array including N rows and M columns, where N and M are positive integers. The array may be configured such that a first node in a first row of the N rows is connected to a second node in the first row and a first node in a first column of the M columns is connected to a second node in the first column. Also an ending node in the first row is connected to the first node in the first row and an ending node in the first column is connected to the first node in the first column. In addition, a first portion of the plurality of nodes is configured to communicate with a plurality of storage devices such as disk drives.

Abstract: A storage array interconnection fabric may be configured using multiple independent paths. A storage system including a plurality of communication paths is configured for connecting each node of a plurality of nodes forming an interconnection fabric. Each of the communications paths is an independent communications path. In addition, a first portion of the plurality of nodes is configured to communicate with a plurality of mass storage devices such as disk drives. A second portion of the plurality of nodes may be configured to communicate with a host.

Abstract: A method for embedding integrity metadata. In one exemplary embodiment, a plurality of integrity metadata segments is determined. Each integrity metadata segment is associated with a segment of user data. The user data is mapped to a plurality of physical sectors, each physical sector containing a segment of user data and the associated integrity metadata segment. For one exemplary embodiment, a common I/O data block size is determined, and its data is mapped into a number of 512-byte sectors. The number of 512-byte sectors corresponds to the number required for the common I/O data block size plus one or more additional 512-byte sectors. This creates additional space in each sector to accommodate the integrity metadata. Integrity metadata for each data segment of the common I/O size is determined. The integrity metadata for each sector is mapped to the additional space of each sector.

Abstract: Data integrity errors in a redundant storage array are handled by storing a plurality of data blocks having a horizontal redundant relationship and storing a plurality of checksums, each checksum having a vertical redundant relationship with a corresponding one of the plurality of data blocks. In response to detection of a data integrity error in at least one of the plurality of data blocks, the vertical redundant relationships between each of the checksums in the plurality of checksums and the corresponding data blocks are reviewed and the horizontal redundant relationship between the data block having the data integrity error and the remaining data blocks in the plurality of data blocks is also reviewed. The results of these reviews of the redundant relationships can be used to diagnose and repair the data integrity error.

Abstract: A method embodied as software or firmware code permits the adaptation of disk drives employing write-back caching to reduce the possibility of lost data from the write cache. In one embodiment, the method is integrated with the host operating system software employed by a host computer coupled to the disk drive. The method issues write requests to the disk drive as it receives them from the applications running on the host computer. The disk drive processes the issued requests as it is designed to, using write-back caching techniques. After each request is cached, the disk drive controller acknowledges the write request back to the host. The host delays communicating the acknowledgements back to their originating applications until the data has been actually written to the disk media. Because write-back caching does not commit cached requests to disk on a regular basis, the host software simply forces the disk drive to execute cached write requests on a regular basis using a CACHE_FLUSH command.

Abstract: A storage system may include a plurality of storage devices each having a plurality of addressable locations for storing data. A storage controller may be coupled to the storage devices and configured to store and retrieve data from the storage devices. An indirection map may be stored within the system having a plurality of map entries each configured to map a virtual address to a physical address on the storage devices. Each map entry may also store a checksum for data stored at the physical address indicated by the map entry. The storage controller may receive storage requests specifying a virtual address and may access the indirection map for each storage request to obtain the corresponding physical address and checksum. Dynamic striping may be employed so that new writes form new parity groups. Thus, stripes of various sizes may be supported by the storage system.

Abstract: A storage system comprises a storage array controller and a storage array, which includes multiple storage devices and storage device controllers. The storage array controller initiates scrubbing operation commands on data stored within a first data range using a first level of checksums for that data range. A scrubbing operation reads data from within the first data range from at least one of the storage devices, calculates a new checksum for the data, and compares the new checksum to a preexisting checksum for the data. If the new checksum doesn't equal the preexisting checksum, the data within the first data range is determined to be erroneous, and additional scrubbing operation commands are initiated for subsets of the data within the first data range using a second level of checksums for that data range.

Abstract: A storage system comprises a storage array controller and a storage array, which includes multiple disk drives and disk drive controllers. The storage array controller issues scrubbing operation commands to one or more of the disk drive controllers. In response, each disk drive controller that receives a scrubbing operation command reads data from within a data range from at least one of the disk drives, calculates a new checksum for the data, and compares the new checksum to a preexisting checksum for the data. If the new checksum doesn't equal the preexisting checksum, the data within the data range is determined to be erroneous.

Abstract: A storage device is monitored for data integrity errors and each detected data integrity error is stored in a count. When that count reaches a threshold limit, the storage device is placed into a forced failure state.

Abstract: A system may include mirroring logic, a controller, and first and second devices (e.g., data storage devices). The first and second devices may include multiple registers. The mirroring logic may be configured in a first mode wherein the mirroring logic allows the registers of the first device to be accessed from the controller and prevents the registers of the second device from being accessed from the controller. The mirroring logic may be configured in a second mode wherein the mirroring logic allows the registers of the second device to be accessed from the controller and prevents the registers of the first device from being accessed. The first and second devices may be configured via the mirroring logic such that the first and second devices are selected simultaneously. When selected simultaneously, the first and second devices may carry out a subsequently issued command substantially simultaneously.