International Business Machines Corporation

Stephen F Edel

Kenneth Wayne Boyd

Gregory E McBride

Kenneth Fairclough Day, Iii

Harry M Yudenfriend

David Michael Shackelford

John Jay Wolfgang

Kenneth F Day, Iii

Michael L Workman

Mantle M Yu

Kenneth G Yamamoto

Ich Van Pham

A method, system, and computer program product for providing continuous data protection for write data supplied from a host incorporating a processor device to data storage in accordance with write input/output (I/O) of an owning application is provided. The write data is identified as one of a synchronous candidate and an asynchronous candidate. The write data identified as the one of the synchronous candidate and the asynchronous candidate is transferred to at least one sidefile. The write data transferred to the at least one sidefile is secured in a nonvolatile storage journal. The nonvolatile storage journal is preset with a timeout limit applicable to write data identified as the synchronous candidate. If the timeout limit is exceeded, a loss mitigation operation to safeguard the write data identified as the synchronous candidate is performed.

Synchronous and asynchronous continuous data protection

Provided are a method, system, and article of manufacture for copying writes from primary storages to secondary storages across different networks. A failure notification is communicated, using a first network protocol, from a primary device in a first group of at least one primary device and at least one corresponding primary storage managed by the at least one primary device to a control system in response to the primary device determining that a write to the primary storage cannot be copied to a corresponding secondary storage. A failure notification is communicated, using a second network protocol, from a primary device in a second group of at least one primary device and at least one corresponding primary storage managed by the at least one primary device to the control system in response to the primary device determining that a write to the corresponding primary storage cannot be copied to a corresponding secondary storage. A freeze command is received, from the control system using the first network protocol at the at least one primary device in the first group. A freeze command is received, from the control system using the second network protocol at the at least one primary device in the second group. The copying of writes from the primary storages in the first and second groups to the corresponding secondary storages is suspended in response to receiving the freeze commands.

Copying writes from primary storages to secondary storages across different networks

An apparatus, system, and method are disclosed for switching a volume address association in a point-in-time copy relationship. A copy module creates a point-in-time copy structure of a source volume at a target volume. A migration module copies data from the source volume to the target volume. A detection module detects data corruption in the source volume. A switch module switches a logical volume address from a source volume address of the source volume to a target volume address of the target volume in response to detecting the data corruption, redirecting data reads and writes to the target volume. In one embodiment, a tracking module tracks a current target volume copy of source volume data and redirects an access for source volume data without the current target volume copy to the source volume.

Apparatus, system, and method for switching a volume address association in a point-in-time copy relationship

Provided are a method, system, and article of manufacture for managing the copying of writes from primary storages to secondary storages across different networks. Information is maintained on a first group of at least one primary storage, at least one primary device managing Input/Output (I/O) access to the at least one primary storage, and at least one corresponding secondary storage to which writes to the at least one primary storage are copied and a second group of at least one primary storage, at least one primary device managing Input/Output (I/O) access to the at least one primary storage, and at least one corresponding secondary storage to which writes to the at least one primary storage are copied. A failure notification is received from one of the primary devices in the first or second group of a failure to copy a write from the primary device to the secondary device. A freeze command is issued over a first network, using a first network protocol, to the at least one primary device in the first group to stop copying writes to the corresponding at least one secondary device. A freeze command is issued over a second network, using a second network protocol, to the at least one primary device in the second group to stop copying writes to the corresponding at least one secondary device.

Managing the copying of writes from primary storages to secondary storages across different networks

A data recording disk file with a voice coil motor (VCM) actuator and a digital servo control system for positioning the read/write heads to selected concentric data tracks includes means for calculating the factor relating acceleration of the heads to VCM input current. This permits each disk file to compensate for any variation in the nominal VCM physical parameters, thus enabling VCMs to be manufactured to less demanding tolerances. The disk file also calculates a unique acceleration factor for each head position during movement of the heads from one track to another, thereby improving the arrival time of the heads to the target track. During an initialization procedure, the disk file performs a track seek, during which VCM current and head velocity information are acquired and used by the microprocessor in the digital servo control system to calculate a value representing an average actual acceleration factor unique to that disk file. This value, together with predetermined constants representing the characteristic relationship of acceleration factor to head position in a disk file with a nominal VCM, are used by the microprocessor to calculate the actual acceleration factor for each head position during a track seek.

Disk file digital servo control system with compensation for variation

A disk file digital servo control system incorporates means for measuring the time between samples of read/write head position error signals (PES), and uses a value of that measured time as part of the computation of the digital control signal. The digital servo control system has a microprocessor which runs a control signal algorithm, a portion of which is a state estimator for estimating the position, velocity and acceleration of the head. The state estimator requires the use of estimator constants determined from physical parameters of the disk file, including the PES sampling time. By using the actual measured PES sampling time, rather than a constant value of nominal PES sampling time, the estimator constants can be modified to improve the accuracy of the computed control signal. This results in improved performance of the control system to move the head to the target track whenever there are variations in the nominal PES sampling time caused by variations in the disk file drive motor speed.

Growing community of inventors

San Jose, CA, United States of America

Stephen F Edel