Fujitsu Corporation

Yoshifumi Ujibashi

Motoyuki Kawaba

Kouichi Kumon

Yui Noma

Kohta Nakashima

Minoru Nakamura

Ryusuke Nishikawa

A processor obtains an input dataset and an output dataset. The processor begins a search process that is a process of searching for a data conversion path of converting the input dataset via one or more intermediate datasets to the output dataset and that includes generating a plurality of intermediate datasets from the input dataset with different data conversion methods, filtering the plurality of intermediate datasets to select a next search intermediate dataset, and generating another intermediate dataset from the next search intermediate dataset. The processor outputs two or more intermediate datasets generated in the course of the search process and receives selection information indicating a selection of part of the two or more intermediate datasets. The processor controls the filtering method for selecting the next search intermediate dataset on the basis of the selection information.

Search control method and search control apparatus

A non-transitory computer-readable recording medium storing a data transformation program that causes a processor to execute a process. The process includes generating a plurality of first programs, each of the first programs transforming first input data and outputting first output data, contents of the transforming by the plurality of the first programs being different from each other, and among a plurality of pieces of a second input data different from the first input data, outputting the second input data that maximizes an entropy of a plurality of pieces of second output data, where each of the first programs transforms the second input data to the second output data.

Non-transitory computer-readable recording medium, data transformation device, and data transformation method

An information processing apparatus is disclosed. A processor acquires an upper limit and a lower limit of a probability of a false positive for each of multiple tests based on data-after-aggregation pertinent to a presence or absence of a specific event occurrence acquired by multiple testing, and sets a value from multiple upper limits being acquired. The processor calculates the probability of the false positive with respect to each of tests having lower limits less than the value, and acquires a set of probabilities of the false positive.

Information processing apparatus and method

In a search apparatus, a processing unit obtains input data and output data and stores these data in a storage unit. The processing unit classifies a plurality of data conversion methods into a plurality of groups. With respect to each group, the processing unit produces intermediate data from the input data using the data conversion methods belonging to the group, and evaluates the intermediate data using an evaluation function corresponding to the group, in order to thereby search for a combination of data conversion methods within the group. The processing unit determines a combination of data conversion methods that is able to convert the input data into the output data on the basis of the search results obtained for the plurality of groups.

Computer-readable recording medium storing search program and search method

An information processing apparatus includes a memory and a processor coupled to the memory and configured to extract data having x times y bits from genome data in which a mutation pattern at each gene mutation position is represented as x bits, such that the extracted data is constituted by y data pieces each having x bits at y respective mutation positions, x and y being integers greater than or equal to 1, respectively, to refer to an addend table that stores a plurality of addend data associated with respective x-times-y-bit data to identify addend data corresponding to the extracted data, and to use an SIMD instruction to add the identified addend data to count data at positions corresponding to the y mutation positions in the genome data, the count data indicating counts of occurrences of mutation patterns in the genome data.

Information processing apparatus and method of collecting genome data

A variant information processing device for processing genetic information includes a processor configured to create variant storage data, from variant information of each of a plurality of target individuals to be processed, where the variant information includes information of variant locus and variant pattern associated with the variant locus. The variant locus corresponds to a portion where the genetic information varies among the plurality of target individuals, the variant pattern corresponds to the genetic information of the portion, and the variant storage data includes an array region with each a first storage region with a fixed bit length and a second storage region with the fixed bit length. The code associated with the variant pattern at each of the variant locus is stored in first storage region or both of the first and second storage regions depending on the length of variant pattern associated with the code.

Variant information processing device and method

A variant master table includes respective columns for variant ID, variant pattern, and pattern code. A composite primary key constraint for the variant master table is set by the columns for variant ID and pattern code. The respective processors reads variant information, verifies whether or not variant patterns for variant positions have been entered into the table, and for unentered variant patterns, attempts to insert a row of information into the table that includes the unentered variant pattern, a corresponding variant ID, and a pattern code allocated based on the variant master table. In cases in which the row of information violates a composite primary key constrain, the respective processor reattempts to insert an updated row of information in which the pattern code has been changed. The processors encode the variant information based on information entered into the variant master table.

Information processing device, method, and computer-readable recording medium

A database system includes a storage device which stores a database storing a plurality of data groups, range information including a minimum and a maximum of an appointed item, a total value and a number of the data, and an information processing device comprises a processor configured to, in response to deletion of a first data, update the total value and the number of the data, calculate a difference between a first total value, based on the number of the data and at least one of the maximum and the minimum in the range information, and a second total value which is updated, as minimum or maximum, judge at least one of whether or not the minimum which is calculated exceeds the minimum in the range information and whether or not the maximum which is calculated is less than the maximum in the range information, and update the range information.

Database system, information processing device and database program

A plurality of compute nodes are divided into a plurality of groups. A maximum available resource amount determining unit determines, for each of the plurality of groups, the available resource amount of the compute node having the greatest available resource amount among the compute nodes belonging to the group as the maximum available resource amount of the group. An excluding unit compares the resource consumption of a job with the maximum available resource amount of each of the plurality of groups, and excludes a group whose maximum available resource amount is less than the resource consumption from search objects. A searching unit searches for a compute node whose available resource amount is greater than or equal to the resource consumption, from the compute nodes belonging to a group that is not excluded from the search objects.

Job scheduling apparatus and method therefor

An information processing apparatus includes a detection unit and a frequency control unit and is designed to activate power saving control and perform information processing. The detection unit detects an idle time of a processor caused by job scheduling. The frequency control unit reduces the operating frequency of the processor for executing a job scheduled to be executed immediately before the idle time, below the maximum frequency. At this time, the frequency control unit reduces the operating frequency such that the execution time of the job is extended to the execution start time of a waiting job, which is included in the idle time.

Information processing apparatus and power saving control method

A disclosed priority control program recorded in a computer-readable medium causes a computer to execute, in job allocation for computational resources, a first step of lowering a job allocation priority of a user based on an estimated utilization amount of a job associated with the user, the job allocation priority indicating a degree of priority of the user in obtaining an allocation of the computational resource, and the estimated utilization amount being an amount of the computational resources estimated to be used for the job and being submitted to and recorded in a memory device on a job-to-job basis; and a second step of increasing the job allocation priority over time at a restoration rate which corresponds to a user-specific amount of the computational resources available for the user per unit time, the user-specific amount being recorded in the memory device on a user-to-user basis.

Dynamically changing job allocation priority based on a restoration rate for each user and time frame

A computer produces a first schedule of jobs including ongoing jobs and pending jobs which is to cause a plurality of computing resources to execute the pending jobs while preventing suspension of the ongoing jobs running on the computing resources. The computer also produces a second schedule of the jobs which allows the ongoing jobs to be suspended and rescheduled to cause the computing resources to execute the suspended jobs and pending jobs. Based on the produced first and second schedules, the computer calculates an advantage factor representing advantages to be obtained by suspending jobs, as well as a loss factor representing losses to be caused by suspending jobs. The computer chooses either the first schedule or the second schedule, based on a comparison between the advantage factor and loss factor.

Method of scheduling jobs and information processing apparatus implementing same

A job scheduling apparatus determines an assignment order, which is the order in which jobs are assigned to a computational resource, on the basis of priority levels and being associated with the assignment order. The apparatus assigns the jobs to the computational resource on the basis of the assignment order. The apparatus reduces the priority levels for the jobs that have been assigned to the computational resource. The apparatus increases the priority levels with time. Regarding a priority level among the priority levels, if, at a future time, which is a fixed time period from the start of execution of the jobs, an amount of an increase in the priority level is expected to be equal to or larger than an amount of a reduction in the priority level for a job, assignment of the job to the computational resource is executed.

Growing community of inventors

Kawasaki, Japan

Yoshifumi Ujibashi