JPH02297670A - Data base retrieving system - Google Patents

Data base retrieving system

Info

Publication number
JPH02297670A
JPH02297670A JP1118983A JP11898389A JPH02297670A JP H02297670 A JPH02297670 A JP H02297670A JP 1118983 A JP1118983 A JP 1118983A JP 11898389 A JP11898389 A JP 11898389A JP H02297670 A JPH02297670 A JP H02297670A
Authority
JP
Japan
Prior art keywords
data
retrieval
processing
parallel
storage devices
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP1118983A
Other languages
Japanese (ja)
Inventor
Kouji Zaiki
材木 幸治
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Panasonic Holdings Corp
Original Assignee
Matsushita Electric Industrial Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Matsushita Electric Industrial Co Ltd filed Critical Matsushita Electric Industrial Co Ltd
Priority to JP1118983A priority Critical patent/JPH02297670A/en
Publication of JPH02297670A publication Critical patent/JPH02297670A/en
Pending legal-status Critical Current

Links

Landscapes

  • Multi Processors (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

PURPOSE:To retrieve a data base in large scale at high speed and to reduce a main memory area as well by distributedly storing an attribute and an index to data in each storage devices and executing the retrieval parallelly by respective processors. CONSTITUTION:The attribute and index of the data are distributedly stored in respective storage devices 4-6 of a parallel processing system computer, which is composed of plural processors 1-3 and the storage devices 4-6, and the retrieval of the data is executed parallely by the respective processors 1-3. Thus, since the plural processors 1-3 parallely execute the retrieval by utilizing the parallel processing system computer, the data base in the large scale can be retrieved at high speed. Since operation to directly handle the substance of the data is not included in retrieval processing, time for data transfer to a main memory until the start of the retrieval and the main memory area can be reduced.

Description

【発明の詳細な説明】 産業上の利用分野 本発明は並列処理装置を用いたデータベース検索方式に
関すム 従来の技術 従来 データベースの検索に用いられてきた方法(よ 
基本的に逐次処理であり、一時に一つの操作 すなわ板
 索引の探索と論理的比較が行なわれも −例として、
第4図版 従来から行なわれてきた単一処理装置を用い
た検索方法を示す。これ(よ 処理装置16と記憶装置
17とから成る処理ユニット15. 検索されるべきデ
ータを格納する補助記憶装置10とから構成されも 検
索されるデータの実体24.25.26.27.28.
29は補助記憶装置10に格納され 検索されるデータ
を管理するための表18を記憶装置17に格納し 処理
装置16は照会されたキーワードか収 この表18をサ
ーチして、該当データを補助記憶装置10から得も 従
って、該当データの検索は処理装置16により逐次的に
行なわれも発明が解決しようとする課題 しかし 上述した従来の方法で(上 データの検索が逐
次的に行なわれ またデータベースのデータの実体その
ものを検索の対象とするた八 データベースの大規模化
に伴賎 データベースの−検索に長時間かかってしまう
という問題があつ九 本発明は上述の課題に鑑みて試さ
れたもので高速にデータベースが検索できるデータベー
ス検索方式を提供することを目的とすも 課題を解決するための手段 本発明は上述の課題を解決するた八 データの属性・索
引を複数の処理装置と記憶装置から成る並列処理方式計
算機の各前記記憶装置に分散して格納して、各前記処理
装置が並行してデータの検索を行なうという構成を備え
たものであも作用 本発明は上述の構成によって並列処理方式計算機を利用
して、複数の処理装置が検索を並行して行なうた六 大
規模データベースの検索を高速に行うことができも ま
た 検索処理にはデータの実体を直接扱う操作は含まれ
ないム 検索を開始するまでの主記憶へのデータ転送時
間と主記憶域の節減が図れも 実施例 第1図は本発明のデータベース検索方式を示す流れ図で
ある。第2図は本発明の一実施例に使用される並列処理
方式計算機の模式図であム この並列処理方式計算機は
 処理ユニッ) 7,8.9とアクセス機構11、及び
補助記憶装置10よりなも処理ユニット71ヨ  処理
装置(1)lと記憶装置(1)4からなり、処理ユニッ
ト8(友 処理装置(2)2と記憶装置(2)5からな
り、処理ユニット9(よ 処理装置(3)3と記憶装置
(3)6からなも 処理装置1.2,3が実行する命令
コードはそれぞれ 記憶装置4,5.6に格納され 処
理装置1,2.3は独立に並行して実行できも 従って
、この並列処理方式計算機(よ 多重命令多重データ処
理方式(MIMD)並列計算機であも この並列処理方
式計算機を用いて、データベース検索を行なう場合の模
式図を第3図に示す。
[Detailed Description of the Invention] Industrial Application Field The present invention relates to a database search method using a parallel processing device.
It is basically a sequential process, one operation at a time, i.e. searching the board index and logical comparison - for example,
Figure 4 shows a conventional search method using a single processing device. A processing unit 15 consisting of a processing device 16 and a storage device 17.A processing unit 15 consisting of a processing device 16 and a storage device 17.A processing unit 15 consisting of an auxiliary storage device 10 that stores the data to be retrieved.The entity of the data to be retrieved 24.25.26.27.28.
29 is stored in the auxiliary storage device 10; a table 18 for managing the data to be searched is stored in the storage device 17; Therefore, the problem to be solved by the invention is that the search for the relevant data is carried out sequentially by the processing unit 16. However, in the conventional method described above (above) The present invention was tried in view of the above-mentioned problems, and has been developed in view of the above-mentioned problems. In order to solve the above-mentioned problems, the present invention aims to provide a database search method that can search a database in a database. The present invention also has a configuration in which the data is distributed and stored in each of the storage devices of the parallel processing computer, and each of the processing devices searches the data in parallel. By using a computer, multiple processing units can perform searches in parallel, which makes it possible to search large-scale databases at high speed. Embodiment Fig. 1 is a flowchart showing the database search method of the present invention. Fig. 2 is a flowchart showing the database search method of the present invention. This is a schematic diagram of a parallel processing computer to be used. and storage device (1) 4; The instruction codes executed by processing units 1, 2, and 3 are stored in storage units 4, 5, and 6, respectively, and processing units 1, 2, and 3 can be executed independently and in parallel. Even if it is a multiple-instruction multiple-data processing (MIMD) parallel computer, a schematic diagram of a case where a database search is performed using this parallel processing computer is shown in FIG.

第1図に示した流れ図に従って、本発明の検索方式につ
いて以下説明すも まず、ステップ19として、第4図
に示したデータを管理するための属性の表18を第3図
に示すよう?、Q  12.13.14の3つの部分に
分割し 各処理ユニット7.8.9の記憶装置4,5.
6に分散して格納する。次へ ステップ20として、補
助記憶装置10にデータの実体24゜25、26.27
.28.29が格納されているアドレスを、8表12.
13.14のエントリ欄に設定すム 次&ζ ステップ
21として、ホスト計算機 第2図では処理ユニット7
.8.9のうちの1つがホスト計算機に相当する力丈 
ここから検索されるべきデータのキーワードを各処理装
置 即板 そのホスト計算機以外の処理装置に転送し 
各処理装置は検索処理を開始する。すなわ杖 各処理ユ
ニッ) ?、8.9の処理装置1,2.3ζよ それぞ
れ記憶装置4,5.6に格納されてい本 データの管理
表12.13.14を並列に検索処理を行なう。また 
このとき各処理装置1 、2 、3 i友  補助記憶
装置lOにあるデータの実体24.25.26.27.
28.29を直接参照することはな(〜続いてステップ
22として、例えばホスト計算機を処理ユニット7とし
 検索の語気 データの実体26が存在するエントリ3
が照会されたとすると、検索処理を終了した処理装置 
即ち処理装置2がエントリ3の値をホスト計算機である
処理二ニット7に転送すム 更にこの啄 ホスト計算機
である処理ユニット7力丈 他の処理ユニット9く 検
索が終了したことを知らせも これにより、全ての処理
ユニットは検索処理を終了すも 最後番へステップ23
として、ホスト計算機である処理ユニット7(上 処理
ユニット8から送られたエントリ3によって、補助記憶
装置10に格納されているデータの実体26のアドレス
を知り、このデータの実体26を取り出す。以上の本発
明の検索方式により、並列処理方式計算機を利用して、
複数の処理装置が並行して検索を行bx  また検索処
理にはデータの実体を直接扱わないた八 データベース
の検索の高速化と主記憶域の節減が図れる。な耘 本実
施例で(よ 処理ユニットを3つの場合を示した力(そ
の数が多くなるほど検索の高速化がより図れることは言
うまでもな(− 発明の詳細 な説明したように本発明によれは 大規模データベース
の検索の高速化と主記憶域の節減が図れる効果があも
The search method of the present invention will be described below according to the flowchart shown in FIG. 1. First, in step 19, a table 18 of attributes for managing the data shown in FIG. 4 is created as shown in FIG. , Q 12.13.14 into three parts, each processing unit 7.8.9 storage device 4, 5.
It is distributed and stored in 6. Next, as step 20, data entities 24, 25, 26, 27 are stored in the auxiliary storage device 10.
.. The address where 28.29 is stored is 8 Table 12.
13. Set in the entry field of 14 Next & ζ As step 21, the host computer
.. 8. One of the nine is the strength equivalent to the host computer.
From here, the keywords of the data to be searched are immediately transferred to each processing device to a processing device other than the host computer.
Each processing device starts a search process. Sunawa Cane (each processing unit)? , 8.9's processing units 1 and 2.3ζ perform parallel search processing on the data management tables 12.13.14 stored in the storage devices 4 and 5.6, respectively. Also
At this time, the data entities 24, 25, 26, 27, .
28. Do not directly refer to 29 (~Subsequently, in step 22, for example, the host computer is set as the processing unit 7, and the entry 3 in which the search word data entity 26 exists
is queried, the processing device that finished the search process
That is, the processing unit 2 transfers the value of entry 3 to the processing unit 7, which is the host computer, and also notifies the processing unit 7, which is the host computer, that the search has ended. , all the processing units finish the search process, but go to the last step 23
As a result, the address of the data entity 26 stored in the auxiliary storage device 10 is learned from the entry 3 sent from the processing unit 7 (upper processing unit 8) which is the host computer, and this data entity 26 is retrieved. According to the search method of the present invention, using a parallel processing computer,
Since multiple processing devices perform searches in parallel, the data entities are not directly handled in the search processing, which speeds up database searches and saves main storage space. In this embodiment, the present invention shows a case where there are three processing units (it goes without saying that the higher the number, the faster the search can be achieved). It has the effect of speeding up searches of large-scale databases and saving main storage space.

【図面の簡単な説明】[Brief explanation of drawings]

第1図は本発明のデータベース検索方式を示す流れ医 
第2図は本発明に用いる並列処理方式計算機の模式図 
第3図は検索されるデータを管理する様子を示した模式
図 第4図は検索の従来例として用いた模式図である。 1.2.3・・・・処理装置 4.5.6・・・・記憶
装置7.8.9・・・・処理ユニット、10・・・・補
助記憶装置11・・・・アクセス機!  12.13.
14・・・・管理表18・・・・管理表、24〜29・
・・・データの実俟代理人の氏名 弁理士 粟野重孝 
はか1名第1図 7、8. q−兎埋ユニット l0−−−楕功記憔4表1 第4図 1<
FIG. 1 is a flowchart showing the database search method of the present invention.
Figure 2 is a schematic diagram of a parallel processing computer used in the present invention.
FIG. 3 is a schematic diagram showing how data to be searched is managed. FIG. 4 is a schematic diagram used as a conventional example of searching. 1.2.3...Processing device 4.5.6...Storage device 7.8.9...Processing unit, 10...Auxiliary storage device 11...Access machine! 12.13.
14... Management table 18... Management table, 24-29.
...Name of the data attorney: Shigetaka Awano, patent attorney
1 person Figure 1 7, 8. q-Usagi unit 10 --- Ellipse record 4 Table 1 Figure 4 1 <

Claims (2)

【特許請求の範囲】[Claims] (1)並列に実行できる複数の処理装置と記憶装置から
成る並列処理方式計算機において、前記各記憶装置にデ
ータに対する属性・索引を分散して格納し、前記各処理
装置が並行して検索を行なうことを特徴とするデータベ
ース検索方式。
(1) In a parallel processing computer consisting of a plurality of processing devices and storage devices that can execute in parallel, attributes and indexes for data are stored in a distributed manner in each of the storage devices, and each of the processing devices performs searches in parallel. A database search method characterized by:
(2)並列処理方式計算機の各記憶装置に表形式でデー
タの属性、索引及び前記データの実体が存在するアドレ
ス・ポインタが格納され、前記データの実体は一箇所で
管理されることを特徴とする特許請求の範囲第1項記載
のデータベース検索方式。
(2) Data attributes, indexes, and address pointers where the data entities exist are stored in table format in each storage device of the parallel processing computer, and the data entities are managed in one place. A database search method according to claim 1.
JP1118983A 1989-05-12 1989-05-12 Data base retrieving system Pending JPH02297670A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP1118983A JPH02297670A (en) 1989-05-12 1989-05-12 Data base retrieving system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP1118983A JPH02297670A (en) 1989-05-12 1989-05-12 Data base retrieving system

Publications (1)

Publication Number Publication Date
JPH02297670A true JPH02297670A (en) 1990-12-10

Family

ID=14750107

Family Applications (1)

Application Number Title Priority Date Filing Date
JP1118983A Pending JPH02297670A (en) 1989-05-12 1989-05-12 Data base retrieving system

Country Status (1)

Country Link
JP (1) JPH02297670A (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06274548A (en) * 1993-03-22 1994-09-30 A T R Jido Honyaku Denwa Kenkyusho:Kk Similarity degree calculating device
JPH07319923A (en) * 1994-04-04 1995-12-08 At & T Global Inf Solutions Internatl Inc Method and equipment for processing of parallel database of multiprocessor computer system
JPH1196168A (en) * 1997-09-17 1999-04-09 Hitachi Information Systems Ltd Parallel retrieval/statistic processing system, method therefor, and storage medium recording program for realizing the method
US5978803A (en) * 1995-11-08 1999-11-02 Fujitsu Limited Information retrieval system
JP2003108165A (en) * 2001-09-28 2003-04-11 Daiichikosho Co Ltd Karaoke collective system with featured melody retrieval processing system
JP2010218131A (en) * 2009-03-16 2010-09-30 Fujitsu Ltd Search method, search program, and search device
US8423562B2 (en) 2011-06-06 2013-04-16 Fujitsu Limited Non-transitory, computer readable storage medium, search method, and search apparatus

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06274548A (en) * 1993-03-22 1994-09-30 A T R Jido Honyaku Denwa Kenkyusho:Kk Similarity degree calculating device
JPH07319923A (en) * 1994-04-04 1995-12-08 At & T Global Inf Solutions Internatl Inc Method and equipment for processing of parallel database of multiprocessor computer system
US5978803A (en) * 1995-11-08 1999-11-02 Fujitsu Limited Information retrieval system
JPH1196168A (en) * 1997-09-17 1999-04-09 Hitachi Information Systems Ltd Parallel retrieval/statistic processing system, method therefor, and storage medium recording program for realizing the method
JP2003108165A (en) * 2001-09-28 2003-04-11 Daiichikosho Co Ltd Karaoke collective system with featured melody retrieval processing system
JP2010218131A (en) * 2009-03-16 2010-09-30 Fujitsu Ltd Search method, search program, and search device
US8423562B2 (en) 2011-06-06 2013-04-16 Fujitsu Limited Non-transitory, computer readable storage medium, search method, and search apparatus

Similar Documents

Publication Publication Date Title
JPH09319757A (en) Information retrieving system
JP2006172067A (en) Database management method, system and program
JPH02297670A (en) Data base retrieving system
CN109753533A (en) A kind of multi-source relevant database client development approach and device
JPH07334402A (en) Data base as main memory
JPH09218880A (en) Document data retrieval system
JP3260743B2 (en) Information search method
JPH0628401A (en) Retrieval system
JPH10269225A (en) Data base dividing method
von Bultzingsloewen et al. Kardamom—a dataflow database machine for real-time applications
JPH02116936A (en) Reorganizing system
Walker An information retrieval package for microcomputers
JP2740238B2 (en) File exclusive control unit
JPH01120621A (en) Retrieving/updating system for data base
JPH01282635A (en) Index maintaining system
JPS633351A (en) Buffer retrieving control system
Papadopoulos et al. Access Methods.
JPH0581115A (en) Data processing method for distributed database managing system
JPH03282749A (en) Temporary update data control system
WO2002091236A1 (en) Database integration system
JPS63285631A (en) Updating and processing system for index file
JPH0455952A (en) Data transfer system for data base system
Tsichlas Apostolos N. Papadopoulos
JPH0375873A (en) Data base processing system
JPS63220365A (en) Image data retrieving system