Hitachi, Ltd.

Hitachi Microcomputer Engineering, Ltd.

Kenzo Kurihara

Tomohiro Murata

Mitsuaki Niida

Takashi Kobayashi

Minoru Kosuge

Takashi Doi

Masaharu Akatsu

Akira Yamamoto

Shigeo Homma

Shigeo Honma

Yoshihiro Uchiyama

Kazuhiro Kawashima

Michio Miyazaki

Yoshiaki Kuwahara

Norihisa Komoda

Isao Tsushima

Kichizo Akashi

Kenichi Kageura

Keiichi Hara

Soichi Oyama

Masaru Teranishi

Ai Tanaka

Noriko Komori

A plurality of processors of a disk controller are divided into a plurality of clusters. In normal operation, data is transferred between a host computer and a disk drive in a cluster non-restricted processing mode in which a job can be distributedly processed among processors belonging to any clusters without regard to cluster. In response to a program version upgrading command from an operator, the cluster mode is shifted from the cluster non-restricted processing mode to a cluster restricted processing mode in which the cluster to which processors which are permitted to distributedly process one job belong is restricted. When the shifting to the cluster restricted processing mode is completed, one cluster is selected as a cluster to be maintained and versions of the programs to be executed by the processors belonging to the cluster to be maintained are upgraded. When the versions of the programs of all processors of the cluster to be maintained have been upgraded, the acceptance of a new job by the processors belonging to the cluster to be maintained is resumed and the execution of the new programs is started. The selection of the cluster, the version upgrading of the programs and the start of the execution of the new programs are repeated sequentially for all clusters. When the version upgrading of the programs of all processors is completed, the cluster mode is returned from the cluster restricted processing mode to the cluster non-restricted process mode.

Multi-processor system, disk controller using the same and nondisruptive maintenance method thereof

A plurality of processors of a disk controller are divided into a plurality of clusters. In normal operation, data is transferred between a host computer and a disk drive in a cluster non-restricted processing mode in which a job can be distributedly processed among processors belonging to any clusters without regard to cluster. In response to a program version upgrading command from an operator, the cluster mode is shifted from the cluster non-restricted processing mode to a cluster restricted processing mode in which the cluster to which processors which are permitted to distributedly process one job belong is restricted. When the shifting to the cluster restricted processing mode is completed, one cluster is selected as a cluster to be maintained and versions of the programs to be executed by the processors belonging to the cluster to be maintained are upgraded. When the versions of the programs of all processors of the cluster to be maintained have been upgraded, the acceptance of a new job by the processors belonging to the cluster to be maintained is resumed and the execution of the new programs is started. The selection of the cluster, the version upgrading of the programs and the start of the execution of the new programs are repeated sequentially for all clusters. When the version of the programs of all processors is completed, the cluster mode is returned from the cluster restricted processing mode to the cluster non-restricted processing mode.

System for processing an I/O request using an old program while loading a new program which upon completion used to process the I/O request

A plurality of processors of a disk controller are divided into a plurality of clusters. In normal operation, data is transferred between a host computer and a disk drive in a cluster non-restricted processing mode in which a job can be distributedly processed among processors belonging to any clusters without regard to cluster. In response to a program version upgrading command from an operator, the cluster mode is shifted from the cluster non-restricted processing mode to a cluster restricted processing mode in which the cluster to which processors which are permitted to distributedly process one job belong is restricted. When the shifting to the cluster restricted processing mode is completed, one cluster is selected as a cluster to be maintained and version of the programs to be executed by the processors belonging to the cluster to be maintained are upgraded. When the version of the programs of all processors of the cluster to be maintained have been upgraded, the acceptance of a new job by the processors belonging to the cluster to be maintained is resumed and the execution of the new programs is started. The selection of the cluster, the version upgrading of the programs and the start of the execution of the new programs are repeated sequentially for all clusters. When the version upgrading of the programs of all processors is completed, the cluster mode is returned from the cluster restricted processing mode to the cluster non-restricted processing mode.

System for selecting one cluster of processors and upgrading the version

A disk controller itself performs the serial inspection process in the following way. It divides a track into a current track group directly available for an external computer and a spare track group available for the disk controller independently of the external computer before registering the groups in it. It also automatically selects a track available in the spare track group as an alternative track. It further copies to the alternative track data recorded on a current track to be inspected. In turn, the disk controller writes a test pattern on the current track. With this, it detects a defect position on a surface of a recording medium of the current track. It also registers the defect position detected on the surface of the recording medium to the current track. It further copies the data having been copied on the alternative track to the current track.

Method and system for a high-reliable disk system

A disc controller, which is connected to a plurality of channels for supplying access requests and discs. The controller includes a plurality of storage paths which control transfer of data between a cache, the channels and the discs, and a control memory which controls the respective operations of the plurality of storage paths. The control memory stores information which is used to select a storage path in accordance with predetermined standards for storage path selection. A disc or the cache is accessed using the selected storage path.

Disc access control method for cache-embedded disc control apparatus

In order for a disk control unit with built-in cache to write non-reflective data in the cache to a disk drive without stopping extension of a read/write command from host computer in the event of a fault in the cache or its backup memory, the track data in normal one of the cache and its backup memory is written to a certain physical track of the disk drive upon judging at read/write command execution that the accessed track is a non-reflective track, and a read/write command from the host computer is implemented for the physical track of the disk drive which has completed the non-reflective track write operation.

Method of fault handling for a disk control unit with built-in cache

An information medium storage and management system includes storage compartments for storing a plurality of information media; a plurality of recording/reproducing apparatus each capable of accessing each of a plurality of information media, each capable of recording at least one file; handling apparatus for mounting a mount information medium on a mount recording/reproducing apparatus and demounting a demount information medium from a demount recording/reproducing apparatus; and a controller for controlling the handling apparatus to select as the mount information medium one of the plurality of information media in accordance with the use history data of each information medium, to select as the mount recording/reproducing apparatus one of unoccupied recording/reproducing apparatus of the plurality of recording/reproducing apparatus, and to cause the mount information medium to use the mount recording/reproducing apparatus for a predetermined time period.

Method and system for storing and managing a plurality of information

When the mount and demount operation of information media are achieved between an information medium storage shelf and a plurality of recording/playback apparatuses by use of a plurality of handling devices, in order to execute the operations in a concurrent fashion without causing an interference between the handling devices for an increased operation efficiency, the area of the storage shelf and at least one of the recording/playback apparatuses are respectively subdivided into a plurality of subareas so as to process the obtained subareas by the plurality devices respectively associated therewith.

Automatic storage system and method for information recording media

In a scheduling system, constraints changing based on situations and scheduling know-how are implemented according to the knowledge engineering method, whereas computations for the concrete, optimal allocation are conducted according to the mathematical programming method. As a result, there can be achieved a scheduling which can easily cope with changes in the scheduling know-how and constraints and which has a high maintainability and a high computation speed.

Growing community of inventors

Tokyo, Japan

Kenzo Kurihara