Abstract: A video data record/playback system includes a plurality of magnetic disk devices that sequentially store video data in scattered form, a plurality of servers 5 that are respectively connected to the plurality of magnetic disk devices and access the data, a control device 9 that is connected to the plurality of servers 6 and sequentially records video data of a video camera 8 to the magnetic disk devices as data, and a monitor device 4 that is connected to the plurality of servers 6, sequentially reads data stored in the disks, and displays the read data.

Abstract: A client-server system which improves efficiency of each server by distributing loads to a plurality of servers having a plurality of storages for sequentially storing data which are distributed in each of the plurality of storages; a plurality of servers, each of which is connected to the plurality of storages, for accessing the data distributed and stored in each of the plurality of storages; an administration apparatus which is connected to the plurality of servers for administrating the data sequentially stored in the plurality of storages, and the plurality of servers; and a client which is connected to both the administration apparatus and the plurality of servers, the clients specifies a server which is connected to a storage where a head block of the data is stored by inquiring to the administration apparatus, and accesses the data in the plurality of servers according to the order of the data storage sequence from the specified server.

Abstract: A data allocator which allocates data and redundant data for recovering the data. The data allocator has k storage groups, each of which includes m array files. Each array file includes multiple memory areas for storing the data and the redundant data on a block basis. A redundant group is formed which consists of k memory areas, each of which is selected from the same row of each one of k storage groups, and one of the k memory areas stores the redundant data and the remaining (k-1) memory areas store the data. The redundant data are uniformly allocated to m.sup.k sets of memory areas (D.sub.ij1, D.sub.2j2, . . . , D.sub.kjk) of the same rows of the storage groups, where j.sub.i (j.sub.i =1-m) is the number of the array file in each of the storage groups, and D.sub.kjk is the memory area belonging to k-th storage group and j.sub.k -th array file of the k-th storage group.

Abstract: A data transfer for high-speed data transfer between two memories comprises three buffers for temporarily storing data being transferred, a source side controller for switching the buffers and for transferring the data in terms of blocks in such a manner that the leading buffer coincides with the address boundary of the source memory, and a destination side controller for switching the buffers and for controlling the transfer of data in terms of blocks. By switching over the buffers one after another, the data transfer from the source memory to the data transfer apparatus and the data transfer from the data transfer apparatus to the destination memory can take place simultaneously.

Abstract: The invention relates to a client-server system which improves efficiency of each server by distributing loads to a plurality of servers. The a client-server system comprises a plurality of storages for sequentially storing data which are distributed in each of the plurality of storages; a plurality of servers, each of which is connected to the plurality of storages, for accessing the data distributed and stored in each of the plurality of storages; an administration apparatus which is connected to the plurality of servers for administrating the data sequentially stored in the plurality of storages, and the plurality of servers; and a client which is connected to both the administration apparatus and the plurality of servers, the clients specifies a server which is connected to a storage where a head block of the data is stored by inquiring to the administration apparatus, and accesses the data in the plurality of servers according to the order of the data storage sequence from the specified server.

Abstract: A memory device with a redundant disk array. In order to quicken the processing of data writing into the redundant disk array, a first cache memory for accessing the redundant disk array, a second cache memory for storing old data of the first cache memory when the data of the first cache memory are renewed, and a third cache memory for storing old CK data read out of the redundant disk array when the data of the first cache memory are renewed are provided. New CK data are produced from new data for the renewal, old data stored in the second cache memory and old CK data stored in the third cache memory, and the new data for the renewal and the new CK data are stored into data storage means and CK data storage means of the redundant disk array.

Abstract: A join processor for relational databases, which includes a main processor ( 1) for processing information; a main memory (2) connected to the main processor to store information; a plurality of auxiliary processors (3) controlled by the main processor; a plurality of auxiliary memories (4) each connected to the corresponding auxiliary processor; and a device for enabling said auxiliary processors to perform in parallel dividing and storing relations of relational databases in the auxiliary memories in units of record and extracting from the auxiliary memories and sorting join fields of the relations to be joined.

Abstract: In a data processing system for management of a relational data base stored among a plurality of disc storage units, a method and apparatus for horizontally partitioning a physical page on the basis of tuples includes a master processor, a master disc storage unit coupled to the master processor, a plurality of slave processors controlled by the master processor, and a plurality of slave disc storage units, one coupled to each of the slave processors. The master disc storage unit stores, in the form of a B-tree structure, a clustered index for either an attribute or a relation to be processed in the relational data base. The plurality of slave disc storage units store divisionally a relation in the data base which is partitioned on the basis of a page for a clustered index thereof in such a manner that the plurality of slave processors may execute, in parallel, a plurality of processings on a cluster of tuples, as defined in their range in connection with a given key value of the clustered index.