Digital Equipment Corporation

Other

Charles M Riggle

Lih-Jyh Weng

John D Read

Christopher Anthony Pollard

Peter R Svendsen

Bruce Douglas Buch

Michael D Leis

Satish L Rege

Bruce A Leshay

Frank W Bernett

Patrick L Hearn

Norman A Field

Shyam C Parikh

Michael J Muchnik

Kenneth F Veseskis

William B Noe

Nersi Nazari

Richard Winfrey

A method and apparatus for bandwidth balancing of data transfer operations between a computer and a storage subsystem are disclosed. The storage subsystem contains a number of storage devices such as magnetic disk drives. Data to be stored is supplied by the computer to the storage subsystem in the form of a transfer unit through a communication channel. The storage subsystem divides the transfer unit into a number of stripes of a pre-determined size. Each stripe is allocated to a separate disk drive whose disk surfaces are formatted into a number of track bands. Each track band is composed of several contiguous tracks associated with the same data transfer rate. Each stripe is then stored on its disk drive within a selected track band. Both data storage and retrieval from each disk drive occur at the data transfer rate associated with the accessed track band. Since all the stripes in the transfer unit are transferred to their disk drives simultaneously, the transfer operation occurs at an aggregate transfer rate equal to the sum of the individual track band transfer rates. The storage subsystem selects the track bands on each disk drive in such a manner as to ensure that the data transfer operation occurs at an aggregate transfer rate within a pre-determined bandwidth range of the communication channel bandwidth.

System for selectively storing stripes of data in tracks of disks so

An improved system and method is provided for enhanced add-compare-select (ACS) implementation which is particularly adapted to time-nesting or over-lapping of the time offsets for add and compare operations. The compare operation is implemented as a sequential bottom-up procedure whereby two numerical quantities are compared by first declaring one of the quantities as a 'contingent' smaller or larger quantity. Subsequently, the least significant bits LSBs of the quantities are compared and the earlier contingent designation is retained unless the smaller of the compared bits is found to correspond to the quantity not previously designated as the 'contingent' smaller quantity, whereupon the 'contingent' designation is transferred to the previously un-designated quantity corresponding to the smaller of the compared bits. The process is iterated until all bit pairs in the compared quantities have been examined and the 'contingent' smaller quantity remaining at that point is defined to be the 'final' smaller quantity and constitutes the result of the overall compare operation. The enhanced ACS implementation compresses the computation time by almost a factor of two compared to traditional ACS implementations using top-down compare operations, without significantly affecting logic complexity of the system and eliminates the need for tie breaking.

System for enhanced implementation of add-compare-select (ACS) functions

A disk drive system which has a plurality of disks and a plurality of transducing head positioning across. The arms are mounted about common shaft for independent notation to desired radial position on the disk.

Multiple-disk storage system

A disk drive system provides all required synchronization, positioning, validation, and data functions within each disk sector. All of these functions are provided within two zones, a header section and a data section. The header section includes a preamble, a synchronization character and an address field, as well as servo information for track following. The data section of each sector includes a data preamble, a data synchronization character, a bad sector bit map, the data and data redundancy information. The header section of at least one sector in a track includes a short DC-erase field, a transitionless segment which is used in synchronization. To synchronize a read/write head to the disk, the system first detects the DC-erase field. The system next searches for the header premable and synchronization character. If it finds them within predetermined times, it then looks for a valid sector address to complete the synchronization. If the system does not detect the preamble and synchronization character within the predetermined times, it looks for another DC-erase field and continues the synchronization process. After the preamble and synchronization character are found, the desired sector is located by reading the addresses of the succeeding sectors.

High efficiency disk format and synchronization system

A balance-moment head positioner includes a rotor assembly which rotates about a central shaft. The rotor assembly supports at the same axial location a current-carrying coil, arms for the heads, and a counterweight such that the positioner is dynamically balanced with its center of gravity at the center of rotation. The coils are arranged with legs parallel to the central shaft on diametrically opposite sides of the assembly. In addition, the positioner includes magnets and an elongated U-shaped magnetic pole member with ends facing the coils at the same axial location at the arms. Thus the torque developed when current flows through the coil, and a reactive counter-torque developed by the rotating mass coincide, substantially eliminating all torsion in the shaft, and there is no net force on the shaft.

Integral balanced-moment head positioner

A data demodulator assigns a binary value to the signal recorded in a transition cell based on the amplitude of the signal in the transition cell, the amplitudes of the signal in adjacent transition cells and system experience in categorizing how recording non-linearities and system noise effect the signals. The demodulator samples a selected number of times the recorded signal in the transition cell for which a binary value is to be determined and the recorded signal in a selected number of transition cells adjacent to that cell. It then converts the amplitude of each of the signal samples to a digital symbol. Next, it concatenates portions of the digital symbols to form an interpretation word. It uses this word to enter a stored lookup table which contains binary values, and assigns to the transition cell signal the binary value associated with the word. The table associates a value with a particular word or group of words based on system experience and the demodulation of known signals into interpretation words. The binary values assigned to a number of cells may be combined to form a code word which may be decoded and corrected using soft decoding techniques.

Data demodulation system

A magnetic tachometer for generating a signal as a function of the velocity of a transducer positioning arm in a disk drive. The tachometer is formed by a pair of fixed parallel coils separated by a distance sufficient to allow a magnet, attached to a counterbalance portion of the rotary positioning arm, to move therebetween and thus to generate a voltage as a function of the velocity of the magnet. The coils are each wound over a thin armature member which is saturated by the magnet's magnetic field and they are differentially coupled to produce common mode rejection.

Magnetic tachometer for disk drives

An error detecting and correcting system implementing the Reed-Solomon (1023, 1006) code having code words whose symbols are elements in the Galois field GF(2.sup.10) generated by either the primitive polynomial x.sup.10 +x.sup.3 +1 or x.sup.10 +x.sup.7 +1. An original data word is encoded to produce a code word w(x) including a first set of checksum symbols appended thereto. Upon retrieval, the data symbols of the receive code word y(x) are encoded by the same encoder that encodes the original data word to produce a second set of checksum symbols. Both sets of checksum symbols are modulo-two summed to produce a residue R(x) from which error syndromes S.sub.i can be computed and thus enable rapid correction of errors in the received code word y(x). The system also monitors the number of non-zero symbols in the residue R(x) in order to avoid unnecessary computation of error syndromes S.sub.i and other decoding routines, such as when the received code word y(x) is otherwise uncorrectable or when the error exists only in the received checksum symbols, rather than in the data symbols. The distance between code words being (2T+ 2), the error correction routine is bypassed when the number of non-zero symbols in R(x) is less than or equal to T, which indicates that errors only reside in the checksum symbols. When the number of non-zero symbols equals (T+1), the error is uncorrectable. For determining whether a single error exists so that correction can quickly be made, the system also tests whether S.sub.i+1 /S.sub.i is constant for all error syndromes S.sub.i.

Multiple error detecting and correcting system employing Reed-Solomon

A housing (1-10) for a disk (1-12) used in a 'Winchester'-type disk mass information storage unit in a data processing system. The housing is symmetrical about the plane of the disk, and supports the disk spindle (2-33) on both sides of the disk. The housing includes two substantially identical members (1-16, 17) fitted together in a clamshell-type arrangement. The housing includes a housing extension (1-26) for receiving and supporting the read/write heads (1-14) and actuator (1-15) therefor from both ends thereof. The members forming the housing are preferably thin walled castings, and may include fins (1-104) to dissipate heat or to enhance the housing's mechanical rigidity.

Disk housing for disk drive mass storage unit

A rotary actuator assembly for positioning read/write heads in disk drive mass storage systems. The actuator assembly includes an actuator motor and one or more actuator arms secured thereto. The motor provides rotary motion of the arms, to move the heads. The stator magnet of the motor is an electromagnet wound on an arcuate core; a pair of elongate permanent magnets are attached to the rotor member, for coaction with the stator magnet. The rotor member has a U-shaped cross sectional area in a plane perpendicular to its axis of rotation; the permanent magnets are fixed to the outside, bottom corners of the U and the actuator arms are mounted across the open end of the U. The rotor is of a very small radius, to minimize inertia. The permanent magnets have a flat face facing the stator, with a channel running longitudinally down the middle, to minimize flux density variations in the permanent magnet-coil gap. The rotor balances the arm relative to the rotor's axis. The arms are of a substantially symmetrical tapered truss design, presenting low inertia; each is cantilevered to the rotor by one bolt which is passed through and pulls the arm against the rotors. Additionally, a spirol-type cylindrical spiral spring, received in a slot in the arm, prevents lateral slippage of the arm relative to the rotor. A pair of cantilevered leaf spring crash stops fixed to the stator decelerate the rotor at the ends of its range of travel.

Growing community of inventors

Colorado Springs, CO, United States of America

Charles M Riggle