Seagate Technology Incorporated

Western Digital Technologies, Inc.

Mark Alan Heath

Mark A Gaertner

Bang Cong Nguyen

Jon David Trantham

Brian Thomas Edgar

Monty Aaron Forehand

Kenny T Coker

Jimmie Ray Shaver

Travis D Fox

Kenny Troy Coker

David Christopher Pruett

Kenneth Lawrence Barham

Merle E Thowe

Gina R Danner

John Garrett Lingo

D Christopher Pruett

Praveen Viraraghavan

A method for managing a memory stack provides mapping a part of the memory stack to a span of fast memory and a part of the memory stack to a span of slow memory, wherein the fast memory provides access speed substantially higher than the access speed provided by the slow memory.

Memory stacks management

A target location for an actuator of a disk drive is determined for prepositioning the actuator before receiving a command from a host. The target location is determined based on at least one of a characteristic of the disk drive, recent command activity of the host, and information provided by the host. The actuator is moved toward the target location to preposition the actuator.

Actuator prepositioning for disk drive

A data storage device is disclosed comprising a non-volatile memory, such as a disk of a disk drive or a non-volatile semiconductor memory. First and second write commands are received from a host, and in response first and second write data are requested from the host. The first write data is stored in the non-volatile memory, and after storing the first write data, a first acknowledge is transmitted to the host to acknowledge completion of the first write command. After transmitting the first acknowledge, a third write command is received from the host. The second write data is stored in the non-volatile memory, and after storing the second write data, a second acknowledge is transmitted to the host acknowledging completion of the second write command. After transmitting the second acknowledge, third write data is requested from the host associated with the third write command.

Data storage device overlapping host data transfer for a write command with inter-command delay

An embedded system with reduced susceptibility to single event upset effects. The system includes an instruction memory that can store at least one instruction set. The instruction memory utilizes a parity checking error-detection scheme. The system also includes a non-volatile memory that can store a copy of the at least one instruction set, and a data memory that can store at least one data sequence. The data memory utilizes an error correction coding (ECC) scheme. A controller, which is responsive to the instruction memory, the non-volatile memory, and the data memory, replaces the at least one instruction set in the instruction memory with the copy of the at least one instruction set from the non-volatile memory, if a parity error is detected in connection with the at least one instruction set in the instruction memory. The controller also operates in conjunction with the data memory to implement the ECC scheme.

Embedded system with reduced susceptibility to single event upset effects

A system and method for accessing data stored on surfaces of discs in a disc drive utilizing more than one serpent per recording zone. The serpents are traversed using a defect descriptor table and one or more physical zone tables (PZTs). An Address Translation Logic Unit is configurable to use one or more PZTs to access multiple serpents per recording zone.

Physical zone table for use with surface-based serpentine format

A disc drive with a control processor programmed with an execution time dependent command schedule optimization method to effect data throughput with a host device. The disc drive includes a head disc assembly executing commands scheduled by the control processor. The control processor selects and schedules a next optimum command from among commands analyzed by the control processor during a time the head disc assembly is executing a current command. The steps utilized by the control processor to select and schedule the next optimum command include executing a first command with the head disc assembly, determining a computation time for a second command, storing the computation time as a computation time estimate, and using the stored computation time estimate to determine the number of commands in a command queue for analysis to provide a level of command schedule optimization commensurate with the available time.

Execution time dependent command schedule optimization

A disc storage system having a host computer interface adapted to coupled to a host computer, a disc storage medium having a disc surface and a spindle motor coupled to the disc adapted to rotate the disc. The disc includes spare data regions and permanent data regions. A transducer is positioned for reading and writing data on the disc surface. The system further includes a controller adapted to write data on the spare data regions to thereby provide a non-volatile write cache.

Disc storage system having non-volatile write cache

A system for controlling movement of a head over tracks on a disc in a disc drive includes a servo positioner for positioning the head over a desired track, a sectoring circuit for determining when the head is over a data sector on the desired track, a formatter for controlling data transfers to and from the desired track, and a controller for providing the servo positioner with a track address indicative of the desired track. A track change system is coupled to the formatter and the servo positioner, and provides a direct communication channel between the servo positioner and the formatter. Thus, the data transfer can be carried out utilizing communications between the formatter and the servo positioner, without waiting for the microcontroller to be interrupted.

System for reducing controller delays during sequential data transfers in a disc drive

A method formats disc surfaces in a disc drive for storage of data in a group of tracks on the disc surfaces. The disc drive performs seek operations to seek among cylinders on the disc surfaces as a function of time according to an actual seek profile. An approximation of the actual seek profile is determined such that the actual seek profile falls within the approximation. The disc surfaces are formatted according to the approximation of the actual seek profile.

System and method for formatting disc surfaces

A data storage system and method of scheduling commands in such a data storage system are provided in which commands are stored in a command sort queue and a scheduled command queue. Commands in the command sort queue are sorted and assigned a priority. Eventually, commands in the command sort queue are transferred to the scheduled command queue. Commands in the scheduled command queue are executed without further sorting. The desired queue depth or size of the scheduled command queue is determined as a function of both the queue depth of the command sort queue and a command execution rate value which is indicative of the rate at which commands in the scheduled command queue are executed. The desired queue depth may be dynamically determined using the queue depth of the command sort queue and the command execution rate value as inputs to a look-up table. The data storage system may include a small computer system interface (SCSI) disc (or &ldquo;disk&rdquo;) drive which executes commands from a host system.

Disc drive data storage system and method for dynamically scheduling queued commands

An apparatus and associated method are disclosed for optimizing the data transfer performance of a disc drive in which user data is stored in data blocks arranged on tracks of at least one rotating disc. When a selected, uninterrupted amount of time passes since the most recently performed disc drive operation, the disc drive enters an idle mode and selectively evaluates the data blocks for uncorrected read errors. During the idle mode, at such time that a selected data block is found having an uncorrected error, a sequence of corrective operations are performed in an attempt to resolve the error and the selected data block is reallocated when at least selected one of the corrective actions are unsuccessful.

Data transfer optimization through sector evaluation and reallocation

The present invention is a method for decreasing rotational latency in systems which include a data retrieval element and a rotating medium. The method requires that the rotating medium be divided into a plurality of discrete angular regions. For each discrete angular region, a number of cylinders that may be traversed by the data retrieval element during a single revolution of the medium is identified. A rotational latency assigned to each discrete angular region based on a current location of the retrieval element. A command queue array is searched for a command that addresses a location within a range of cylinders determined by the number of cylinders in the discrete angular region having the smallest rotational latency. Finally, the command addressing a location in the range of cylinders in the selected discrete angular region of the medium is executed.

Growing community of inventors

Oklahoma City, OK, United States of America

Mark Alan Heath