JPS61145674A - Object parallel processing system - Google Patents

Object parallel processing system

Info

Publication number
JPS61145674A
JPS61145674A JP26787984A JP26787984A JPS61145674A JP S61145674 A JPS61145674 A JP S61145674A JP 26787984 A JP26787984 A JP 26787984A JP 26787984 A JP26787984 A JP 26787984A JP S61145674 A JPS61145674 A JP S61145674A
Authority
JP
Japan
Prior art keywords
task
processing
tasks
parallel
physical processors
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
JP26787984A
Other languages
Japanese (ja)
Inventor
Hitoshi Matsumoto
均 松本
Hideki Sato
秀樹 佐藤
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.)
Fujitsu Ltd
Original Assignee
Fujitsu 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 Fujitsu Ltd filed Critical Fujitsu Ltd
Priority to JP26787984A priority Critical patent/JPS61145674A/en
Publication of JPS61145674A publication Critical patent/JPS61145674A/en
Pending legal-status Critical Current

Links

Landscapes

  • Multi Processors (AREA)

Abstract

PURPOSE:To execute the fast processing of an object group and to use effectively resources by providing a task control mechanism and a task assigning mechanism, assigning a task to plural physical processors, and processing the results in parallel. CONSTITUTION:A model control processor 1 first generates an object group 30 through a task control mechanism 3 as task aggregates 60, 61...6n to a virtual memory device 5. Next, a task assigning mechanism 4 executes the prescribed assignment concerning tasks 70, 71...7n to be processed corresponding to physical processors 20, 21...2n from the task of the device 5. Next, the processors 20...2n executes the prescribed task communication by means of message transmission lines 10, 11... and functions in parallel. Thus, since the object processing is not a sequence processing but a parallel processing, the whole processing can be executed earlier than conventional. Plural physical processors are used, the efficient assignment can be executed and resources can be effectively used.

Description

【発明の詳細な説明】 し産業上の利用分野] 本発明はオブジェクト指向モデルにおけるオブジェクト
をプロセッサのタスクとして考え、タスク間通信により
オブジェクトを並列的に処理させる方式に関する。
DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method in which objects in an object-oriented model are considered as tasks of a processor, and objects are processed in parallel through inter-task communication.

[従来の技術] 人間が目で認識できる事物の紙・電話以外に会議・研究
なども「オブジェクト」としてとらえオブジェクトで構
成されるモデル内で、オブジェクト間を「メツセージ」
通信により事物処理乃至計算を行うことができる。
[Conventional technology] In addition to things that humans can recognize with the naked eye, such as paper and telephones, conferences and research are also considered "objects" and messages are sent between objects within a model made up of objects.
Things can be processed or calculated through communication.

電子計算機上で前記のようにオブジェクトをプログラミ
ング言語によりプログラミング化するという概念ができ
つつある。計算機の資源として計算を行う単位として「
タスク」を考える。通常1つの計算機は一時に1つのタ
スクができるのみであるから、オブジェクト間でメツセ
ージを通信するときに「タスク」の順次処理を行うこと
となる。
The concept of programming objects using a programming language on an electronic computer as described above is emerging. As a unit of calculation as a computer resource, "
Think about "tasks". Normally, one computer can only perform one task at a time, so when messages are communicated between objects, the "tasks" are processed in sequence.

オブジェクトの存在する場所的環境が分散している場合
、互いに協力し合う形でオブジェクト処理例えば知識を
表現するためのデータ処理などの仕事を進めている。1
つのタスクにおいて予定した仕事が出来上がるまで場合
によっては、他のタスク処理を開始することを待ってい
て、出来上がった仕事の結果により、それを受付て、次
の処理を開始することが多い。
When objects are located in dispersed locations, they cooperate with each other to process objects, such as data processing to express knowledge. 1
In some cases, the task waits until the scheduled work for one task is completed to start processing another task, and depending on the result of the completed work, it is often accepted and the next processing is started.

したがって複数のタスクと、複数のオブジェクトを考え
たとき、両者間に上下の関係はないが、そのオブジェク
ト数はタスクの処理の経過と共に増減することとなる。
Therefore, when considering a plurality of tasks and a plurality of objects, there is no vertical relationship between them, but the number of objects increases or decreases as the processing of the tasks progresses.

[発明が解決しようとする問題点] 従来は電子計算機によりタスクを処理するとき、一時に
単純には■タスクの処理ができ、シーケンシャルの処理
が進行するのみであった。複数のタスクをまとめたもの
が、更に複数存在するようなモデルでは、複数個のタス
クが殆ど同時に発生しているため、全体の処理が遅くな
る欠点があった。
[Problems to be Solved by the Invention] Conventionally, when tasks were processed by an electronic computer, only one task could be processed at a time, and the processing proceeded sequentially. In a model in which there are multiple sets of multiple tasks, the multiple tasks occur almost simultaneously, which has the disadvantage of slowing down the overall processing.

[問題点を解決するための手段] 前述の問題点を解決するため本発明の採用した手段は、
タスク管理機構によりモデル管理プロセッサにおける仮
想記憶装置に対しオブジェクトまたはオブジェクト群を
タスクの集合体として生成し、次にタスク割当機構によ
り前記タスクを複数の物理プロセッサに割当て、該タス
クで相互間通信することにより前記オブジェクト群の動
作が並列処理されることである。
[Means for solving the problems] The means adopted by the present invention to solve the above-mentioned problems are as follows:
A task management mechanism generates an object or a group of objects as a collection of tasks in a virtual storage device in a model management processor, and then a task allocation mechanism allocates the tasks to a plurality of physical processors, and the tasks communicate with each other. Accordingly, the operations of the object group are processed in parallel.

[作用] 本発明によりタスクは複数の物理プロセッサに割当られ
、並列処理されるため、オブジェクト群の早い処理が可
能である。
[Operation] According to the present invention, tasks are assigned to a plurality of physical processors and processed in parallel, so that objects can be processed quickly.

[実施例1 第1図は本発明の詳細な説明するためのプロセッサを含
むブロック図である。第1図において、1はモデル管理
プロセッサ、20,2、−2nは複数の物理プロセッサ
、3はタスク管理機構、30はオブジェクト群、4はタ
スク割当機構、5は仮想記憶装置、60.61−・・6
nは仮想記憶装置のタスク、70.71゜−・7nは要
処理タスク、8は業務モデル例を全体的に示すものであ
る。80,8、−8nは業務、90,9、−・9nはタ
スク0.クズり1〜タスクn、 10.11− はタス
ク間通信のメソセージ伝送路を示している。
[Embodiment 1] FIG. 1 is a block diagram including a processor for explaining the present invention in detail. In FIG. 1, 1 is a model management processor, 20, 2, -2n are a plurality of physical processors, 3 is a task management mechanism, 30 is an object group, 4 is a task allocation mechanism, 5 is a virtual storage device, 60.61- ...6
n indicates a virtual storage device task, 70.71°-.7n indicates a task to be processed, and 8 indicates an overall business model example. 80, 8, -8n are tasks, 90, 9, -.9n are tasks 0. 10.11- indicates a message transmission path for inter-task communication.

まずモデル管理プロセッサ1はタスク管理機構3をして
、オブジェクト群30につきタスク集合体60゜6t−
−−6nとして、仮想記憶装置に生成させる。このとき
オブジェクト群中の成るオブジェクトとタスク60〜6
nとの対応は適宜で良い。
First, the model management processor 1 uses the task management mechanism 3 to create a task set of 60°6t- for each object group 30.
--6n, to be generated in the virtual storage device. At this time, the objects in the object group and tasks 60 to 6
The correspondence with n may be made as appropriate.

次にタスク割当機構4は物理プロセッサ20,2、−・
−2nと対応する要処理タスク?0.71、−7nにつ
いて、仮想記憶装置のタスクから所定の割当てを行う。
Next, the task allocation mechanism 4 uses the physical processors 20, 2, -.
-2n and corresponding task to be processed? For 0.71 and -7n, predetermined allocation is performed from the virtual storage device task.

物理プロセッサ20−・の処理能力・タスクの大きさな
どにより、?0,7、−7nが割当てられる。
Depending on the processing power of the physical processor 20, the size of the task, etc.? 0, 7, -7n are assigned.

次に物理プロセッサ20−・は所定のタスク間通信をメ
ツセージ伝送路10.1、・−・−により行いながら、
並列的に動作する。
Next, the physical processors 20-- perform predetermined inter-task communication via the message transmission paths 10.1, .
Works in parallel.

オブジェクト群30が単一のオブジェクトのみである場
合も処理は同様に行う。
The same processing is performed when the object group 30 includes only a single object.

業務モデル例8では業務80〜8nが、業務遂行上タス
ク0 = nを生成した上でタスクOにおいて業務Oが
、タスクlにおいて業務1が実行されると考え、業務O
からメツセージ伝送路10.11−により他のタスクへ
通信し、更にメツセージ伝送路12が他の業務に通信を
行う。この通信は単方向伝送路を示したが、一般的には
双方向伝送路による通信となる。この伝送路により業務
上相手方に実行依頼すること、データ転送すること、処
理結果の返送待ちであることなどを通信し、更に相手方
から返送データを受けたりする。
In business model example 8, businesses 80 to 8n generate task 0 = n in business execution, and consider that business O is executed in task O and business 1 is executed in task l, and business O is
Message transmission lines 10, 11- are used to communicate with other tasks, and message transmission line 12 is used to communicate with other tasks. Although this communication shows a unidirectional transmission path, it is generally a bidirectional transmission path. This transmission path is used to communicate information such as requesting execution to the other party for business purposes, transferring data, and waiting for return of processing results, and also receiving data returned from the other party.

このオブジェクト処理としてオフィスにおける業務モデ
ルを例示すると第2図のようになる。第2図において研
究員X氏、7氏に家族手当を支給するという業務の、上
半分はクラスとし、モデル世界のオブジェクトであり、
下半分は現実世界のオブジェクトでインスタンスという
。下半分では、家族手当支給、2というオブジェクトを
処理するため、タスクは日付、当事者、室長、−1善頬
書式を定めて処理して行くことで、他のタスク、ここで
は研究室のタスクに当事者のX氏、7氏が居るので、第
1図と同様のタスク管理機構・割当機構によりタスクと
プロセッサが割付られ、当事者が具体化して処理が進め
られる。
An example of a business model in an office as this object processing is shown in FIG. In Figure 2, the upper half of the task of paying family allowances to researchers Mr. X and Mr. 7 is a class, which is an object in the model world.
The lower half is a real-world object called an instance. In the lower half, in order to process the object 2, ``Family Allowance Payment,'' the task is processed by setting the date, party, director, and -1 Zenjo format, and then processing it to other tasks, in this case, the laboratory task. Since there are parties involved, Mr.

[発明の効果] このようにして本発明によると、従来のようにオブジェ
クト処理がシーケンス処理ではなく、並列処理となるか
ら、全体的処理は従来より早くできる。また複数の物理
プロセッサを使用し、効率の良い割当を行うことができ
るため、資源を有効に利用することが可能である。
[Effects of the Invention] As described above, according to the present invention, object processing is not sequence processing as in the past, but parallel processing, so that overall processing can be performed faster than in the past. Furthermore, since a plurality of physical processors can be used and efficient allocation can be performed, resources can be used effectively.

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

第1図は本発明の詳細な説明するためのプロセッサを含
むブロック図、 第2図は第1図の業務モデルを例示する図である。 1−モデル管理プロセッサ 20.2、−−一物理プロセッサ3 0−・オブジェクト群 4−・タスク割当機構 5〜・仮想記憶装置 60.6、− タスク ?0.7、−・要処理タスク 8−業務モデル例 特許出願人    富士通株式会社 代理人     弁理士 鈴木栄祐 第 lrl!J
FIG. 1 is a block diagram including a processor for explaining the present invention in detail, and FIG. 2 is a diagram illustrating the business model of FIG. 1. 1-Model management processor 20.2, ---Physical processor 3 0--Object group 4--Task allocation mechanism 5--Virtual storage device 60.6,--Task? 0.7, - Processing Task 8 - Business Model Example Patent Applicant Fujitsu Limited Agent Patent Attorney Eisuke Suzuki lrl! J

Claims (1)

【特許請求の範囲】[Claims] タスク管理機構によりモデル管理プロセッサにおける仮
想記憶装置に対しオブジェクトまたはオブジェクト群を
タスクの集合体として生成し、次にタスク割当機構によ
り前記タスクを複数の物理プロセッサに割当て、該タス
クで相互間通信することにより前記オブジェクト群の動
作が並列処理されることを特徴とするオブジェクト並列
処理方式。
A task management mechanism generates an object or a group of objects as a collection of tasks in a virtual storage device in a model management processor, and then a task allocation mechanism allocates the tasks to a plurality of physical processors, and the tasks communicate with each other. An object parallel processing method characterized in that the operations of the object group are processed in parallel.
JP26787984A 1984-12-19 1984-12-19 Object parallel processing system Pending JPS61145674A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP26787984A JPS61145674A (en) 1984-12-19 1984-12-19 Object parallel processing system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP26787984A JPS61145674A (en) 1984-12-19 1984-12-19 Object parallel processing system

Publications (1)

Publication Number Publication Date
JPS61145674A true JPS61145674A (en) 1986-07-03

Family

ID=17450894

Family Applications (1)

Application Number Title Priority Date Filing Date
JP26787984A Pending JPS61145674A (en) 1984-12-19 1984-12-19 Object parallel processing system

Country Status (1)

Country Link
JP (1) JPS61145674A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05224963A (en) * 1991-09-10 1993-09-03 Internatl Business Mach Corp <Ibm> Process control system for real-time system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05224963A (en) * 1991-09-10 1993-09-03 Internatl Business Mach Corp <Ibm> Process control system for real-time system

Similar Documents

Publication Publication Date Title
US5062037A (en) Method to provide concurrent execution of distributed application programs by a host computer and an intelligent work station on an sna network
EP0362106B1 (en) Method for scheduling execution of distributed application programs at preset times in an SNA LU 6.2 network environment
EP0362105B1 (en) Method for processing program threads of a distributed application program by a host computer and an intelligent work station in an SNA LU 6.2 network environment
US4901231A (en) Extended process for a multiprocessor system
US5201049A (en) System for executing applications program concurrently/serially on different virtual machines
US5925098A (en) Apparatus and method for dispatching client method calls within a server computer system
TW406242B (en) System and method for maximizing usage of computer resources in scheduling of application tasks
US4961137A (en) Method for operating a multiprocessor system for therein establishing a global binary assertion and a multiprocessor apparatus comprising synchronization means for such establishing, in particular for effecting a garbage collection operation
US8046759B2 (en) Resource allocation method and system
CN100553214C (en) Mobile virtual environment system
CN104123265A (en) Method and system for communication among multiple kernels
CN114816694A (en) Multi-process cooperative RPA task scheduling method and device
CN115951974B (en) Management method, system, equipment and medium of GPU virtual machine
JPS61145674A (en) Object parallel processing system
CN114745377B (en) Edge cloud cluster service system and implementation method
CN107832154B (en) Multi-process processing method, processing device and application
KR20050057834A (en) Method for mass data processing through scheduler in multi processor system
JPS6077259A (en) Load dispersion system
KR20100048843A (en) Robot used software component apparatus and thread processing method using by it
US5392426A (en) Method and apparatus for use in program operation, control and control block management and storage
JP2590872B2 (en) Task scheduling method
CN114116181B (en) Distributed data analysis task scheduling system and method
JPH05204875A (en) Scheduling system for thread
CN110515737A (en) Data management task operation method and device
CN113703931A (en) Multitask scheduling method and device