Abstract: A method and apparatus are provided for digitally correcting sector pulse placement in a direct access storage device. A sector clock signal is generated equal to a sub-multiple clock rate of a predetermined servo identification clock. An individual servo identification field location is identified and compared with an expected location of said individual servo identification field to identify an error value. The sector clock signal is modified by adding or subtracting clock pulses responsive to the identified error value for generating a sector pulse.

Abstract: A rotating magnetic storage device is shown wherein the transducer and media are separated during operation by a thin film of low viscosity liquid lubricant that is recirculated from a reservoir within the head-media enclosure. Lubricant recirculation can be effected by capillary action using a wick, mechanical pumping and metering or distillation techniques. As compared to devices using air bearing separation, the transducer carrying slider has very small bearing surfaces which may be in the form of narrow rails or small depending foot elements. This enables fly heights of 2 microinches using a film of one micron thickness.

Abstract: A method and computer system having a processor are provided for selectively compressing a data communication frame for data transfer between the computer system and at least one other system. A processor work load is periodically identified for the processor. A predetermined stored threshold value is identified and compared with the identified processor work load. Responsive to an identified processor work load greater than the identified predetermined threshold value, the data communication frame is transmitted without compressing the data communication frame. Otherwise the data communication frame is compressed and then transmitted responsive to an identified processor work load less than or equal to the identified predetermined threshold value. A state of a communication line is identified for the data transfer. Responsive to an identified idle state of the communication line, the data communication frame is transmitted without compressing the data communication frame.

Abstract: Apparatus and method for providing equalization adjustment for a finite impulse response (FIR) digital filter are provided in a PRML data channel. A plurality of transformations are defined to transform effective tap weights in a digital filter into modified effective tap weight values by adding each said initial effective tap weight value with a selected cosine equalization adjustment function. A cosine expander circuit is provided for expanding 8-byte compressed equalizations into a 32-byte format and for altering the filter response by a selected cosine equalization adjustment function.

Abstract: A rotary actuator head driver for a direct access storage device includes a transducer head moving in a radial direction across a disk surface for reading and/or writing data on a plurality of tracks for storing data. A support arm supports the transducer head relative to the disk surface. The support arm is connected to a pivot defining an axis of rotation spaced from the transducer head. A head driver positions the transducer head over one track in a track following mode or moves the transducer head from one track to another track in a seek mode. The head driver generates oppositely directed forces to produce a net torque applied to the support arm through the pivot. The oppositely directed forces have a line of action orthogonal a dominant motion of the transducer head. As a result, any parasitic force is orthogonal to the dominant motion and does not contribute to track misregistration errors.

Abstract: A 10-tap finite impulse response (FIR) digital filter is provided in a partial response signaling and maximum-likelihood (PRML) data channel. A plurality of partial sums of predetermined tap weights are stored in a palette random access memory (RAM). A fixed qualifier value is received and used together with selected ones of the stored partial sums for calculating a predetermined tap weight. Predetermined filter coefficients are calculated and stored in a filter RAM using the calculated predetermined tap weight and predefined ones of the stored partial sums.

Abstract: The flow of work requests in a server driven process to process communication environment is described. Logical connections between processes and bus managers interfacing bus units to an I/O bus are assigned to connection groups for management by the bus managers. Each bus unit has its own connection groups for the logical connections. Bus unit resources are assigned to each connection group based on performance factors, and a series of bus unit messages are used to control the flow of work so that a group which has no more resources will not accept further work requests. The originator of the work requests will resequence rejected work requests and resend them when the connection group has freed up resources. A further mechanism is provided to facilitate work consistent with the server driven architecture when bus units do not have adequate DMA capabilities. Two ways of reversing control of transfer of work requests and data so that the server need not have master DMA capability are presented.

Abstract: A method and apparatus for modulating a data reference clock for rotational speed variations of the data storage disk in a direct access data storage file. A data reference clock is generated from an encoded disk rate clock of the data file. A readback signal from at least one predetermined region of the disk surface is sensed and a sync byte is detected responsive to the readback signal. An early window control signal and a late window control signal are generated corresponding to end boundaries for capturing readback encoded data signals. A load complete signal is identified responsive to the readback signal. The identified load complete signal with the early window control signal and the late window control signal and the generated data reference clock is adjusted responsive to overlapping compared signals.

Abstract: Apparatus and method for asynchronous servo identification (SID)/address mark detection are provided for data detection in a partial-response maximum-likelihood (PRML) data channel. The SID/AM pattern includes a first preamble section defined by a plurality of short magnets; a body section defined by a predetermined number of long magnets; and a delimiter section defined by a plurality of short magnets. Samples from the digital filter are applied to first and second correlation filtering paths. The first correlation filtering path includes a polarity threshold function to indicate one of a preamble section or a delimiter section of a SID/AM pattern. The second correlation filtering paths includes a second threshold function and a comparator for sequentially comparing sequential ones of the received samples responsive to identifying the second threshold value to identify a predefined sequence.

Abstract: Disclosed is a disk drive having an enclosure for housing one or more disks, an actuator arm and sliders attached to the actuator arm. The sliders include magnetic transducers for reading information related to data off the disk and for writing representations of data to the disk. In a disk drive with an in-hub, fixed spindle shaft motor, the disk or disks are attached to the hub which is rotatably mounted to the spindle shaft. The hub includes one or more fingers attached to the hub which touch the spindle shaft when the hub is stationary and when the hub is rotating. The spindle shaft may also be provided with a wear resistant layer and/or ring of conductive material to enhance the electrical pathway between the hub and the spindle shaft and provide for lower wear and a longer life.

Abstract: A dedicated servo transducer head for phase skewed position information signals in servo information tracks of a rigid magnetic disk drive unit has a width spanning up to eight servo information tracks in order to overcome problems in gain irregularity caused by abrupt phase jumps between tracks. An embodiment of the invention includes a track reading surface that varies in sensitivity across the width of the head to avoid gain loss during incremental radial movement with head edges overlying track edge fringes.

Abstract: Apparatus and method of thermal asperity compensation are provided for data detection in a partial-response maximum-likelihood (PRML) data channel. The PRML data channel includes an analog to digital converter (ADC) having a normal operating range and a filter, gain and timing control coupled to the ADC. When a thermal asperity is detected, a thermal asperity recovery mode is established responsive to the detected thermal asperity. The gain and timing control are held and the normal operating range of the ADC is adjusted responsive to the thermal asperity recovery mode. Also an AC coupling pole frequency is elevated responsive to the thermal asperity recovery mode.

Abstract: A method and apparatus are provided for controlling a digital filter function included in a data path within gain and timing control loops of a PRML magnetic recording channel. The digital filter function is removed from the data path during an acquisition mode and the digital filter function is restored to the data path during a tracking mode. A predetermined gain and a quantized delay at a synchronization frequency are provided when switching from acquisition mode to tracking mode. A unity gain and a delay of an integer number of sample periods are provided at a synchronization frequency when switching from acquisition mode to tracking mode.

Abstract: A data management mechanism for a processor system provides for management of data with minimum data transfer between processes executing work requests. Each process has storage areas for storing data associated with work requests. The data is described with descriptor elements in the work requests which indicate the location and length of segments of the data. Data is transferred to a process only if it is required for execution of a work request. Further work requests can be generated by a process executing a work request which reference the data without the process actually receiving the data The segments of data may reside in storage areas of different processors with the descriptor elements of a work request defining a logical data stream.

Abstract: The flow of work requests in a server driven process to process communication environment is described. Logical connections between processes and bus managers interfacing bus units to an I/O bus are assigned to connection groups for management by the bus managers. Each bus unit has its own connection groups for the logical connections. Bus unit resources are assigned to each connection group based on performance factors, and a series of bus unit messages are used to control the flow of work so that a group which has no more resources will not accept further work requests. The originator of the work requests will resequence rejected work requests and resend them when the connection group has freed up resources. A further mechanism is provided to facilitate work consistent with the server driven architecture when bus units do not have adequate DMA capabilities. Two ways of reversing control of transfer of work requests and data so that the server need not have master DMA capability are presented.

Abstract: Apparatus and methods are provided for encoding a predefined number of bits of binary data into codewords having a predefined number of bits for a partial-response maximum-liklihood (PRML) data channel in a direct access storage device (DASD). Rate 8/9 block codes having maximum ones and run length constraints (0,8,12,.infin.) and (0,8,6,.infin.) provide timing and gain control and reduced susceptibility to misequalization effects in PRML channels.

Abstract: A transducer head skew arrangement used in a multiple disk drive data storage system is provided. The disk drive unit includes a plurality of magnetic disk surfaces mounted for simultaneous rotation about an axis. A first sequence of servo information tracks are arrayed on a dedicated servo disk surface. At least one servo reference track is written on each data disk surface for storing servo reference information. The servo reference track is written on a first data disk surface aligned with a predefined servo information track on the dedicated servo disk surface. Then servo reference tracks are sequentially written on each successive consecutive data disk surface progressively offset radially from the last written servo reference track. A plurality of data information tracks for storing data are disposed at predetermined positions relative to the servo reference track on each the data disk surface.

Abstract: Multiple versions of a given system command are run on a single operating system by using one command processing program per command irrespective of its version. Each version of the command has a command definition which has the same format for all counterparts of the command. A command analyzer program transforms the parameters of a command character string into the correct data form for its command processing program based on the command definition for that version of the command. Multiple sets of command definitions are place in separate storage directories called libraries. The libraries are searched based on a version identifier to find the command definition for a command. The version identifier may be specified as a program attribute, entered interactively or entered by the system. Multiple programs written with different versions of commands may be run in conjunction with each other and the system will dynamically adapt to the version of each program as it is running.

Abstract: Each panel or display screen of data, is defined by a title and the tagged description of the objects, such as instruction lines, data entry lines, menu areas and command lines. Each of these objects may include other objects, described in the same syntax. Panel definitions are joined in a source file to form a panel group. Objects can be shared by all panels.

Abstract: Any number of sorted lists are efficiently partitioned into P lists, where P represents the number of processors available to sort the resulting lists. When given a large list to sort, the list is initially divided into P lists, and each processor sorts one of these lists. The lists are then exactly partitioned so that each of the elements in the new consecutive partitioned lists have values no smaller than any of the elements in the lists before it, nor larger than any of the elements in the list following it. Partitioning is done by P-1 processors. Each of the processors successively considers selected rows of elements from the sorted lists, and moves a partition boundary based on an element magnitude requirement and a partition size requirement. The new partitioned lists are then merged by the P processors, and simply strung together to provide a sorted list of all the elements.