Seagate Technology Incorporated

David Peter DeCenzo

Clark Edward Lubbers

Erik Riedel

Fred Wayne Flournoy

Stephen J Sicola

Michael D Walker

David Louis Spengler

Bill A Pagano

William Pagano

Thomas M Lawlor

Apparatus for storing and retrieving data such as in a computer network. A plurality of local data storage units each comprise a local control circuit and a housing which encloses a moveable data transducer adjacent a data storage medium. A shared resource module physically interconnects each of the local units, and includes shared circuitry that supplies a resource required by the local control circuits to transfer data to and from the data storage media. In some preferred embodiments, the resource comprises a programming instruction set that is utilized by a programmable processor in each of the local control circuits, such as servo code used to provide data transducer positional control. Alternatively, the resource comprises a shared buffer memory space utilized by read/write channels in each of the local control circuits. Preferably, data are stored across the data storage media using RAID techniques (redundant array of independent discs/devices).

Data storage subgroup with local and shared resources

A data storage system is provided with a storage domain having an active zone subportion of storage space, and a command queue controller configured for short-stroking an actuator directed to the active zone. A method is provided for virtualizing a storage space to store user data in a first domain and redundant data in a second domain, partitioning each of the domains into active and inactive subportion storage zones, and executing a retrieve access command from one of the first and second domains depending on which domain's active zone contains an LBA associated with the retrieve access command.

Command queue ordering by flipping active write zones

Method and apparatus for flushing cached writeback data to a storage array. Sets of writeback data are accumulated in a cache memory in an array with a view toward maintaining a substantially uniform distribution of the data across different locations of the storage array. The arrayed sets of data are thereafter transferred from the cache memory to the storage array substantially at a rate at which additional sets of writeback data are provided to the cache memory by a host. Each set of writeback data preferably comprises a plurality of contiguous data blocks, and are preferably written (flushed) to the storage in conjunction with the operation of a separate access command within a selected proximity range of the data with respect to the storage array. A stripe data descriptor (SDD) is preferably maintained for each set of writeback data in the array.

Dynamic adaptive flushing of cached data

Disclosed is a storage system and method that provides multi-path bus and component interconnection and isolation in a data storage system. A plurality of data storage devices in a removable assembly are connected to a fabric that is configurable to connect some or all of the data storage devices (or 'drives') to a drive controller and configurable to isolate one or more data storage devices from the drive controller. Multiple controllers, fabrics, and interconnecting buses may be employed to provide redundancy in the event of a connector, bus, or controller failure. Computer program code operating in a host, interface controller, and/or drive controller configures the fabric to isolate failed devices and may be employed to optimize data transfer rates. Data storage devices may be multi-ported. The fabric may comprise any device or devices capable of configurably interconnecting data storage devices to one or more controllers and may comprise multiplexers, cross point switches, port bypass controllers. Fabrics may also provide translation or conversion of one bus or interface format to another format.

Multipath redundant storage system architecture and method

A data storage system and associated method is provided within an enclosure supporting a self-contained plurality of discrete data storage devices configured for connecting with a network device via a network. The data storage system includes a redundant array of independent drives (RAID) container services module in the enclosure that allocates and manages a storage space of the data storage devices for storing primary and redundant data, and a policy engine in the enclosure that continuously and qualitatively characterizes the network load to the data storage system and manages a power distribution to each of the data storage devices based on a predicted utilization that differentiates between access commands for primary data and access commands for redundant data.

Power management in a storage array

Method and apparatus for controllably applying input power to storage devices in a multi-device array. The array preferably includes a circuit which changes input power state of the storage devices in a time-staggered sequence. Preferably, the change of input power state transitions the array from an operationally deactivated state to an operationally activated state, or vice versa. The storage devices preferably store data in accordance with a selected RAID configuration. A power supply preferably supplies respective first and second direct current (dc) voltages to each of the storage devices, and the circuit respectively connects the dc voltages to the storage devices in a selected sequential order. The circuit further preferably changes an input power state of a single storage device of the array in accordance with a selected time-varying profile. The profile is preferably applied to a gate of a switching device to control a source-drain conduction path.

Individual storage device power control in a multi-device array

An apparatus and associated method for deriving an error code value to characterize an error event that occurs during execution of an access command, memorializing the error event in an indexable error table, and logging an association of the error code value and an index in the error table related to the error event.

Self-defining counters

Disclosed is a data storage architecture that provides data regeneration following the failure of two storage domains and that provides conversion for RAID-6 to RAID-5 or RAID-0 without moving or copying any data. A storage grid comprising at least eight domains and twelve rows contains user data in a first contiguous portion and may contain row parity data in a second continuous portion and may contain vertical parity data in a third contiguous portion. In one embodiment data is arranged in row and vertical sets each comprising four data blocks such that each user data block is a member of one row set and is a member of one vertical set. In another embodiment sets comprise two blocks each.

Dual redundant data storage format and method

A carrier and associated methodology for removably supporting a plurality of data storage devices as a multiple disc array in a distributed storage system. The carrier defines attachment features for fixing each of the data storage devices to the carrier preventing movement of the data storage devices in relation to the carrier in response to a vibration stimulus. This reduction in vibration stimulus allows the individual data storage devices in the multiple disc array to increase the number of track per storage area. This benefit can be used to increase capacity, increase performance and increase reliability of the multiple disc array.

Storage array with enhanced RVI suppression

A data storage array is provided such as for a distributed storage system or a consumer electronics device wherein increased data storage capacity is provided at the cost of increased power consumption in relation to industry standard power consumption rates. The array has a plurality of data storage devices, each having an actuator operably disposed in a data storing and retrieving relationship with a data storage medium. The data storage medium is rotated at a speed of substantially greater than 10,000 revolutions per minute, and each of the data storage mediums has a data storage capacity of substantially greater than 100 gigabytes. The increased heat created by the relatively faster rotation speed is handled through improved airflow made possible by a multiple device array constructed in accordance with embodiments of the present invention.

Growing community of inventors

Pueblo, CO, United States of America

David Peter DeCenzo