US 8,898,677 B2 · Data arrangement calculating system, data arrangement calculating method, master unit and data arranging method

Abstract

A data arrangement calculating system including a master unit and a plurality of slave units connected with said master unit. The master unit includes a data arranging section and a job allocating section. The data arranging section includes a data dividing section and an arranging section configured to arrange a first block of the blocks in a first slave unit of the plurality of slave units as an owner block, and arrange the replica block of a second block of the blocks next to the first block in the first slave unit. The first slave unit includes a data retaining section configured to retain said first block and the replica block of said second block and a job executing section. The job executing section executes the sliding window calculation by using the first block and the replica block of the second block.

Drawings

Drawings are temporarily unavailable. Open the PDF at USPTO

Drawings

Claims16

Dependent claims

A data arrangement calculating system comprising:

a master unit which has data of records arranged in order based on key data of said records and a job which executes sliding window calculation every predetermined window width;
and
a plurality of slave units connected with said master unit,
wherein said master unit comprises:
a data arranging section configured to divide said data to arrange in said plurality of slave units;
and
a job allocating section configured to allocate said job to each of said plurality of slave units,
wherein said data arranging section comprises:
a data dividing section configured to generate blocks and replica blocks of said blocks by dividing said data;
and
an arranging section configured to arrange a first block of said blocks in a first slave unit of said plurality of slave units as an owner block, and arrange the replica block of a second block of said blocks next to said first block in said first slave unit,
wherein said first slave unit comprises:
a data retaining section configured to retain said first block and the replica block of said second block;
and
a job executing section configured to receive said job and execute said job to carry out the sliding window calculation every said predetermined window width to said first block, and
wherein said job executing section executes the sliding window calculation by using said first block and the replica block of said second block when the predetermined window width extends over said first block and the replica block of said second block.

13 more elements

A data arrangement calculating method comprising:

a master unit dividing data of records arranged in order based on key data of said records;
said master unit arranging said data which has been divided, in a plurality of slave units;
said master unit allocating a job which executes sliding window calculation every predetermined window width to each of said plurality of slave units;
and
each of said plurality of slave units executing the sliding window calculation based on the allocated job,
wherein said dividing comprises:
dividing said data to generate blocks and replica blocks of said blocks,
wherein said arranging comprises:
arranging a first block of said blocks as an owner block in a first slave unit of said plurality of slave units;
and
arranging the replica block of said second block next to said first block in said first slave unit,
wherein said executing comprises:
said first slave unit receiving said job and executing the sliding window calculation every said predetermined window width to said first block, and
wherein said executing comprises:
executing the sliding window calculation by using said first block and the replica block of said second block, when said predetermined window width extends over said first block and the replica block of said second block.

12 more elements

A method of arranging data in a non-transitory computer readable medium comprising:

dividing data of records arranged in order defined based on key data of said records;
arranging the divisions of said data in a plurality of slave units;
allocating a job which executes a sliding window calculation every predetermined window width to each of said plurality of slave units;
and
updating each of said plurality of slave units when receiving data containing a new record,
wherein said dividing comprises:
generating blocks and replica blocks of said blocks from said data, and
wherein said arranging comprises:
arranging a first block of said blocks in a first slave unit of said plurality of slave units as an owner block;
arranging the replica block of a second block next to said first block in said first slave unit;
generating arrangement data by relating said first block to said first slave unit as the owner block, and relating the replica block of said second block to said first slave unit, and
wherein said allocating comprises:
referring to said arrangement data to recognize a kind of the key data defining an order of first records which are contained in said first block;
and
allocating a plurality of first key data corresponding to the kind of the key data and said job to said first slave unit, and
wherein said updating comprises:
referring to said arrangement data to insert said new record in said second block to generate an insertion block, based on a second key of the same kind as that of the key data contained in said new record;
determining whether or not a size of said insertion block is within a first threshold value:
dividing said insertion block to generate a fourth block and a fifth block when the size of said insertion block is larger than said first threshold value;
arranging said fourth block as an owner block in a new second slave unit;
arranging a replica block of said fifth block next to said fourth block in said second slave unit;
re-arranging a replica block of said fourth block in said first slave unit which retains the replica block of said second block;
and
re-arranging said fifth block as an owner block in a third slave unit which retains said second block as the owner block.

21 more elements

Dependencies are inferred from the claim text.

Description8,594 words

Summary1,823 words

#This application is a National Stage of International Application No. PCT/JP2010/070854filed Nov. 24, 2010, claiming priority based on Japanese Patent Application No. 2009- 277521 filed Dec. 7, 2009, the contents of all of which are incorporated herein by reference in their entirety.

#The present invention is related to a data arrangement calculating system, especially to a data arrangement calculating system which executes a sliding window calculation.

#As an example of a data arrangement calculating system, Bigtable which manages large-scale data is described in Non-Patent Literature 1. The Bigtable is a system in which a table is divided into tablets which are distributed to and managed by a plurality of servers. The Bigtable has a feature that the table can have the tremendous size and at the same time can be distributed to many servers. In such a system in which data is divided in a suitable size, and managed over a plurality of servers, the sliding window calculation becomes possible according to the key data. The sliding window calculation is a calculating method in which a series of data arranged in an order such as time series data is sequentially calculated every specific section. For example, the sliding window calculation is used to calculate a moving average of stock market and to calculate a place where a user stays and a stay time from a time series of position data of a user.

Brief description of the drawings9 figures

#The objects, effect, features of the present invention would become clearer from exemplary embodiments in conjunction with the attached drawings:

#FIG. 1 is a block diagram showing a configuration example of a data arrangement calculating system 10 according to a first exemplary embodiment of the present invention;

#FIG. 2 is a block diagram showing a hardware configuration example of a master unit 100 and slave units 200 in the exemplary embodiment of the data arrangement calculating system 10;

Detailed description6,528 words

#Hereafter, a data arrangement calculating system, a data arrangement calculating method, a master unit, a data arranging method, a data arrangement program and a storage medium according to exemplary embodiments of the present invention will be described with reference to the attached drawings.

#[First Exemplary Embodiment]

#A first exemplary embodiment of the present invention will be described. FIG. 1 is a block diagram showing a configuration of a data arrangement calculating system 10 according to the first exemplary embodiment of the present invention. Referring to FIG. 1, the data arrangement calculating system 10 is provided with a master unit 100 and a plurality of slave units 200. The master unit 100 has data which have a plurality of records arranged in the order of key data of a predetermined key, and a job which executes a sliding window calculation every predetermined window width. The plurality of slave units 200 are connected with the master unit 100.

Facts

Application

13/514,229Filed 24 Nov 2010

Priority

JP 2009-2775217 Dec 2009

Published

US 2012/0246661 A127 Sep 2012

Granted

25 Nov 2014

Term ends

25 Jun 2031

Estimated · how

20 years from earliest known nonprovisional domestic filing plus recorded patent term adjustment, including 213 days of adjustment. Maintenance-fee lapses, terminal disclaimers and statutory extensions have not been checked. The continuity chain may be incomplete; this is not an enforceability determination.

Classification4 codes

G06F 9/50
Multiprogramming arrangements: allocation of resources, e.g. of the central processing unit [CPU]
G06F 16/2471
Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries: distributed queries
G06F 40/00
Handling natural language data
G06F
Computing and data processing

Timeline

7 Dec 2009Priority claimedJP 2009-277521
24 Nov 2010Application filedUS 13/514,229
7 Jun 2012Assigned to NEC Corporationassignment of assignor's interest
27 Sep 2012PublishedUS 2012/0246661 A1
25 Nov 2014GrantedUS 8,898,677 B2
12 Oct 2026TodayWithin its term
25 Jun 2031Term endsEstimated, includes 213 days of adjustment

Data arrangement calculating system, data arrangement calculating method, master unit and data arranging method

Life of this patent

Data arrangement calculating system, data arrangement calculating method, master unit and data arranging method

Life of this patent

Life of this patent

Abstract

Drawings

Drawings

Claims16

Description8,594 words

Drawings

Classification4 codes

Life of this patent

Description8,594 words