Hitachi, Ltd.

Hitachi Software Engineering Company, Ltd.

Kazuhiko Ohmachi

Masashi Tsuchida

Hiroyuki Kitajima

Yasuhiro Imai

Toshio Honma

Akira Yamamoto

Nobuo Kawamura

Kazuhiro Satoh

Tadashi Ohsone

Yoshiaki Yamashita

Nobuhiro Taniguchi

Junichi Kazama

Mitsuru Doki

Shinichi Fukushima

Naohumi Sakai

Hidekazu Kanao

In a data base processing method including a central processing unit for analyzing a plurality of queries made from a plurality of users so as to determine an internal processing procedure with respect to a data base, and for performing the internal processing procedure; a secondary storage for storing data capable of being shared by each user; and, an input/output controlling processing unit including at least one cache storage, for controlling a transfer of a plurality of records which are required for input/output processing operations between the central processing unit and secondary storage. The data base processing method is featured in that a decision is made whether or not a prefetching operation is carried out at a start of an input/output processing operation by judging the internal processing procedure which has been formed by the query; a retrieval range into which a record to be fetched has been stored is formed when the prefetching operation is performed; and, a prefetching record number is determined as an input/output processing demand unit, taking account of an access characteristic of the internal processing procedure and also a system characteristic containing a size of the cache storage, a performance, of the central processing unit, and a traffic of the input/output processing unit.

Method apparatus for determining prefetch operating for a data base

A method for processing a database for analyzing a query to the database from a user to determine an internal processing procedure therefor and executing it comprises the steps of preparing single or a plurality of internal processing procedures based on analysis result of the query including a variable inputted in a preprocess step and statistical information of a database system, selecting an optimum one of the internal processing procedures based on a value substituted for the variable of the query in a query execution stage and the statistical information of the database system, and executing the query in accordance with the selected internal processing procedure.

System for optimizing query processing in a relational database

After passing through the process in which the number of one page I/O requests generated in a certain unit time period, the number of multiple pages I/O requests, and the number of pages required at each of the multiple pages I/O request are input, the quantity of the I/O requests generated in the above-described unit time period is calculated in accordance with the thus-input values. The unit (the number of pages) for a multiple pages I/O actually performed by one time operation of the I/O control device corresponding to the multiple pages I/O request from users is determined in accordance with the user's instruction or specific criterion for the system based on the result of the above-described calculation, the performance of the CPU, the performance of the I/O control device, the performance of the DASD (Direct Access Storage Decive), and to which DASD each of I/O requests corresponds. The result of this determination is employed as the unit (the number of pages) for the multiple pages I/O performed by one time operation of the I/O control device. In case of the specific criterion for the system, the structure may be constituted in such a manner that it can be automatically changed in accordance with the state of change in the quantity of the I/O request.

Data I/O transaction method and system

A set of sorted columns of one table stored in a DB machine are binary-searched by a hardware in synchronism with data transfer when another table is read from a disk unit, to select a row to be joined. Where a key word sequence to be searched is stored in ascending or descending order, determination of a maximum address in an area in which key words smaller than a search key are stored and determination of a maximum address in an area in which the key words smaller than or equal to the search key are stored are parallelly executed to determine the store range in one searching.

Method and apparatus for processing data base

A method of assinging software resources such as files and programs to memory devices is disclosed which uses first storage for storing therein the access frequency to a memory device for causing each type of service request to utilize each of software resources and second storage for storing therein the processing speed, allowable utilization and storage capacity of each of memory devices and in which the allocation of each of said software resources to the memory devices is determined on the basis of information stored in the first and second storages so that a total time taken to access the memory devices is minimized while being restricted by the allowable utilization and storage capacity of each memory device.

Method of and apparatus for assigning software resources to memory

A distributed processing system having a plurality of processors and/or terminals at a plurality of locations is provided with a processing request allocator which includes a device for storing the waiting time at each processor or terminal for beginning the execution of a service request as well as a device for storing the delay in communication from each location of a processor or terminal to each other processor or terminal in the system. The processing request allocator has a device for calculating the sums of the waiting time and delay for each route through the system where a processing request from a first location is executed by a predetermined processor and the processing result is sent to a second location within the system, and a device is provided for determining the minimum value of these sums to identify the processor or terminal to which the service request is to be allocated. Since a processor or terminal is selected to perform the processing of the service request on the basis of not only the waiting time at each processor but also the communication delay, the performance of the overall system is enhanced and the response time to each service request is minimized.

Processing request allocator for assignment of loads in a distributed

Growing community of inventors

Kawasaki, Japan

Kazuhiko Ohmachi