Hitachi, Ltd.

Atsushi Tomoda

Kazutomo Ushijima

Yuuya Isoda

Hideo Aoki

Tsuyoshi Tanaka

Tomohiro Hanai

Tetsuya Yamada

Atsushi Miyamoto

Masaki Hamamoto

An index tree search method, by a computer, for searching an index tree included in a database provided by the computer which includes processors executing a plurality of threads and a memory, the index tree search method comprising: a first step of allocating, by the computer, search ranges in the index tree to the plurality of threads; a second step of receiving, by the computer, a search key; a third step of selecting, by the computer, a thread corresponding to the received search key; and a fourth step of searching, by the computer, the index tree with the selected thread using the received search key.

Index tree search method and computer

A computer for executing processing through use of a database, the computer comprising: a processor including a cache memory; and a non-volatile memory coupled to the processor, the non-volatile memory having the database constructed thereon. The computer comprises: a database management module configured to execute processing on the database; and a write processing module configured to write data stored in the cache memory into the database. The write processing module writes data that is operated in the transaction processing into the database among data stored in the cache memory, in a case of receiving a commit request for transaction processing that uses the database.

Computer

A database management system generates a log each transaction during execution of a plurality of transactions and stores the generated logs in log storage areas. The database management system records sequence numbers of logs at least in the generated logs of transactions belonging to a set of transactions of which results are different depending on a transaction execution order.

Database management system and method

Provided is a log management method in which, with a computer system provided with a processor, a memory, and a storage device, the processor executes a prescribed process and stores in the storage device a log which includes a description of the process, said method comprising a first step of the processor generating the log which includes the description of the prescribed process, a second step of the processor writing the log to a log file of the storage device, and a third step of the processor determining the end of the log region which stores the log and writing same to the log file of the storage device.

Log management method and computer system

There is provided a parallel computer system for performing barrier synchronization using a master node and a plurality of worker nodes based on the time to allow for an adaptive setting of the synchronization time. When a task process in a certain worker node has not been completed by a worker determination time, the particular worker node performs a communication to indicate that the process has not been completed, to a master node. When the communication has been received by a master determination time, the master node performs a communication to indicate that the process time is extended by a correction process time, in order to adjust and extend the synchronization time. In this way, it is possible to reduce the synchronization overhead associated with the execution of an application with a relatively large variation in the process time from a synchronization point to the next synchronization point.

Growing community of inventors

Tokyo, Japan

Atsushi Tomoda