Hitachi, Ltd.

Naoki Ikawa

Tetsuya Hashimoto

Kenta Takahashi

Kei Takeda

Shinichi Kawamoto

Atsushi Hirata

Kazuhiro Matsuya

Naohiro Suzuki

Junya Sawazaki

Kosuke Shindo

In a data retrieval/distribution apparatus which retrieves designated information from the collected data in accordance with a retrieval/distribution request sent from a data retrieval/distribution client, and sends a retrieval result to the data retrieval/distribution client, the data retrieval/distribution client sends, to the data retrieval/distribution server, the retrieval/distribution request including a first retrieval condition using a virtual data conversion definition, which is a virtualization of data conversion definitions as a definition for converting the collected data into a distribution format, a processing unit of the data retrieval/distribution apparatus generates a second retrieval condition using each of the data conversion definitions from the first retrieval condition, executes retrieval processing according to the generated second retrieval condition, and sends a retrieval result of the retrieval processing to the data retrieval/distribution client as a retrieval result according to the first retrieval condition.

Data retrieval/distribution method and apparatus, and storage medium

Information processing system for rapidly performing analysis of semistructured data while preserving fault-tolerance of data for a store request acquiring the number which has been stored of the same and data structures set beforehand, refer to data structure management information to determine a data store unit, instructs replica creation of the data with regard to the data stored in the data store unit in which any of the replicas have been stored, transmits an instruction, for performing a data operation, to a data structure operation unit, whereupon a processing unit, in accordance with the content of an analysis request, performs analysis processing by way of either data stored in any of the data storage units after a data structure operation or data which have not been subject to a data structure operation.

Information processing system and control method thereof

Disclosed is a time-series management device capable of filtering time-series data having a possibility of matching a specified search pattern and reading in the data from a storage device when performing a time-series analysis. A data accumulation unit () creates a time-series index having a feature value of a data series calculated at a specific regular time interval. In addition, a data search unit () makes a decision as to the feature value for each regular time interval included in the time-series index using an evaluation formula of a specified search condition, identifies a time period of a complying data series group, and performs a time-series analysis for only the data series of the identified time period.

Time-series data management device, system, method, and program

An operation management program collects a query status table of a migration-source agent processing program and creates a server status table. The operation management program, based on the server status table, computes a migration cost for each query, and selects the query with the smallest migration cost as a migration query. Then, the operation management program migrates the selected query, using the optimum migration method.

Stream data processing method and system

A load balancing method for continuation requests transmitted from client moves session information to continuation request to another business server and embeds a character string for making a request transmission apparatus identify the business server of destination in to a response to the continuation request to return the response to the client, so that the business server of destination to which subsequent requests transmitted from the client are transferred is switched.

Load balancing method and system

A computer system having a plurality of computers for responding to a request from a client and a running management computer for managing said plurality of computers, wherein: the running management computer includes a storage unit and a processing unit; the storage unit stores a target value of performance information on a response time to the request, for each URL pattern of the request; and the processing unit monitors the performance information on the response time to the request, for each URL pattern of the request, and judges whether or not the monitored performance information read from the storage unit satisfies the target value of the performance information, to execute load distribution by moving a Web content corresponding to the URL pattern of the request not satisfying the target value of the performance information, to another computer.

Growing community of inventors

Tokyo, Japan

Naoki Ikawa