JPH0395661A - Process assignment device for parallel computer - Google Patents

Process assignment device for parallel computer

Info

Publication number
JPH0395661A
JPH0395661A JP23136889A JP23136889A JPH0395661A JP H0395661 A JPH0395661 A JP H0395661A JP 23136889 A JP23136889 A JP 23136889A JP 23136889 A JP23136889 A JP 23136889A JP H0395661 A JPH0395661 A JP H0395661A
Authority
JP
Japan
Prior art keywords
processor
local memory
processing
allocation
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
JP23136889A
Other languages
Japanese (ja)
Inventor
Haruo Yokota
治夫 横田
Hajime Kitagami
北上 始
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 JP23136889A priority Critical patent/JPH0395661A/en
Publication of JPH0395661A publication Critical patent/JPH0395661A/en
Pending legal-status Critical Current

Links

Landscapes

  • Multi Processors (AREA)

Abstract

PURPOSE:To increase the overall processing speed by assigning a process with preference to a processor of a local memory when it is decided that the data stored in a single local memory can be applied to the process. CONSTITUTION:A processor assignment table 34 of a processor assignment device 31 stores the assignment situation of processes to the processors. A local memory state store part 35 stores the store state of the data stored in each local memory. Then an assignment part 36 of the device 31 assigns the processes to the processors by reference to the table 34. At the same time, a process is assigned with preference to the processor of a relevant local memory when it is decided that the data stored in the single local memory can be applied to the processor based on the data of the part 35. As a result, the useless transfer of data can be omitted and the overall processing speed is improved.

Description

【発明の詳細な説明】 (概要) 並列計算機の処理割り当て装置に係り、特に複数のプロ
セサを有し、各プロセサにローカルメモリを設けてなる
並列計算機の各プロセサへ処理を割り当てる処理割り当
て装置に関し、 データの転送処理をなるべく少なくすることができるよ
うにすることを目白勺とし、 上記のような処理割り当て装置を、プロセサへの処理の
割り当て状況を格納したプロセサ割り当てテーブルと、
各ローカルメモリが格納するデータの格納状況を格納す
るローカルメモリ状況格納部と、上記プロセサ割り当て
テーブルに基づいて処理を各プロセサに割り当てると共
に上記ローカルメモリ状況格納部のデータに基づき、処
理が一のローカルメモリの格納データを使用できると判
断した時には当該ローカルメモリの属するプロセサに優
先的に処理を割り当てる割り当て都とから構戊する。
DETAILED DESCRIPTION OF THE INVENTION (Summary) This invention relates to a processing allocation device for a parallel computer, and particularly to a processing allocation device that allocates processing to each processor of a parallel computer having a plurality of processors and each processor having a local memory. The aim is to reduce data transfer processing as much as possible, and the above-mentioned processing allocation device is equipped with a processor allocation table that stores the status of processing allocation to processors,
A local memory status storage unit that stores the storage status of data stored in each local memory, and a local memory status storage unit that allocates processing to each processor based on the above-mentioned processor allocation table. When it is determined that data stored in a memory can be used, processing is preferentially assigned to a processor to which the local memory belongs.

(産業上の利用分野) 本発明は、並列計算機の処理割り当て装置に係り、特に
複数のプロセサを有し、各プロセサにローカルメモリを
設けてなる並列計算機の各プロセサへ処理を割り当てる
処理割り当て装置に関する。
(Industrial Application Field) The present invention relates to a processing allocation device for a parallel computer, and more particularly to a processing allocation device for allocating processing to each processor of a parallel computer having a plurality of processors, each of which is provided with a local memory. .

?従来の技術) 近年処理の高速化を計るため、第6図に示すように複数
のプロセサ11−,〜11−Nを共通バス13で結合し
、共有メモリ14にアクセスできるようにすると共に各
プロセサ11−1〜11−Nにもローカルメモリ12−
■〜12−Nを配置し、処理割り当て装置15により各
プロセサ11−1〜l1−Nに処理を割り当てるように
した計算機システムかある。このような計算機システム
にあっては処理を割り当てられたプロセサは共有メモリ
14からデータを読み出し、自己のローカルメモリに処
理環境を構築し、割り当てられた処理を実行する。
? Prior Art) In recent years, in order to speed up processing, a plurality of processors 11-, to 11-N are connected by a common bus 13, as shown in FIG. 6, so that they can access a shared memory 14, and each processor 11-1 to 11-N also have local memory 12-
There is a computer system in which processors 11-1 to 12-N are arranged, and a processing assignment device 15 assigns processing to each of the processors 11-1 to 11-N. In such a computer system, a processor assigned a process reads data from the shared memory 14, builds a processing environment in its own local memory, and executes the assigned process.

このような計算機システムにおいて、処理割り当て装置
にあっては一般には処理要求は要求された順に一列の待
ち行列(キューと称する)を作戊して、処理を空いてい
る(処理を実行していない)プロセサに順次割り当てる
ようにしていた。
In such a computer system, the processing allocation device generally creates a queue (referred to as a queue) for processing requests in the order in which they are requested, and waits for idle processing (processing that is not being executed). ) was allocated sequentially to the processors.

〔発明が解決しようとする課題〕[Problem to be solved by the invention]

ところでこのような計算機システムにおいて、あるプロ
セサがある処理を行なうため、自己のローカルメモリに
処理環境を構築し処理を開始し、その処理の途中に、優
先順位を有する他の処理を実行するため当該処理を中断
し、他の処理を実行することがある。このような場合に
は、その中断された処理を改めて実行するとき、従来の
処?割り当て装置においては当該処理を他のプロセサに
割り当てることがあり、このような場合には、割り当て
られたプロセサは改めて共宥メモリから、データを読み
取り、自らのローカルメモリに処理環境を構築しなけれ
ばならず、処理の無駄を招くこととなる。このような事
態は同一プロセサで再処理をする場合において、既にデ
ータを共有メモリに転送してしまっている場合も同様で
ある。
By the way, in such a computer system, in order to perform a certain process, a certain processor builds a processing environment in its own local memory and starts the process, and in the middle of that process, in order to execute another process that has priority, the processor Processing may be interrupted and other processing may be executed. In such a case, when re-executing the interrupted process, do you use the conventional method? The allocation device may allocate the processing to another processor, and in such a case, the allocated processor must read the data again from the shared memory and create a processing environment in its own local memory. This results in wasteful processing. This situation also occurs when reprocessing is performed using the same processor and the data has already been transferred to the shared memory.

また、第7図に示すように複数のローカルメモリ16−
■〜16−Nを有する複数のプロセサ17−1〜17−
.をネットワーク18で結合し処理割り当て装置19て
処理を割り当てることが可能としたシステムにおいて、
第8図に示すように、例えば、プロセサAのローカルメ
モリに処理a及び処理bの処理環境が、また、他のプロ
セサBのローカルメモリに処理C及び処理dの環境が格
納されているとき、処理aがプロセサBに割り付けられ
た場合にはプロセサAのローカルメモリからプロセサB
のローカルメモリに可処理環境aを転送する必要があり
、処理に時間がかかるという問題がある。
Further, as shown in FIG. 7, a plurality of local memories 16-
■Multiple processors 17-1 to 17- having ~16-N
.. In a system in which it is possible to connect processes through a network 18 and allocate processes using a process allocation device 19,
As shown in FIG. 8, for example, when the processing environments of processing a and processing b are stored in the local memory of processor A, and the processing environments of processing C and processing d are stored in the local memory of another processor B, If process a is assigned to processor B, it is transferred from processor A's local memory to processor B.
There is a problem in that it is necessary to transfer the processable environment a to the local memory of the computer, and the processing takes time.

そこで本発明はデータの転送処理をなるべく少なくする
ことができる並列計算機の処理割り当て装置を提供する
ことを目的とする。
Therefore, an object of the present invention is to provide a processing allocation device for a parallel computer that can reduce data transfer processing as much as possible.

?課題を解決するための手段〕 本発明にあって,上記の課題を解決するための手段は、
第1図に示すように、複数のプロセサ1一■〜1−nを
有し、各プロセサ1−1〜1−n,にローカルメモリ2
−1〜2−nを設けてなる並列計算機の各プロセサ1−
1〜1−nへ処理を割り当てる処理割り当て装置3であ
って、プロセサ1−■〜1−nへの処理の割り当て状況
を格納したプロセサ割り当てテーブル4と、各ローカル
メモリ2−1〜2−nが格納するデータの格納状況を格
納するローカルメモリ状況格納郡5と、上記プロセサ割
り当てテーブル4に基づいて処理を各プロセサ1−1〜
1−nに割り当てると共に上記ローカルメモリ状況格納
部5のデータに基づき、処理が一のローカルメモリ1.
の格納データを使用できると判断した時には当該ローカ
ルメモリ2.の属するプロセサ1−1に優先的に処理を
割り当てる割り当て都6とからなるものである。
? Means for Solving the Problems] In the present invention, means for solving the above problems are as follows:
As shown in FIG. 1, it has a plurality of processors 1-1 to 1-n, each of which has a local memory 2.
- Each processor 1- of a parallel computer provided with 1 to 2-n
A process allocation device 3 that allocates processes to processors 1 to 1-n, which includes a processor allocation table 4 that stores the status of process allocation to processors 1-1 to 1-n, and each local memory 2-1 to 2-n. Each processor 1-1 to 1-1 performs processing based on the local memory status storage group 5 that stores the storage status of data stored in
1-n, and based on the data in the local memory status storage unit 5, processing is performed on one local memory 1.
When it is determined that the data stored in the local memory 2. can be used, the local memory 2. and an allocation capital 6 that preferentially allocates processing to the processor 1-1 to which the processor 1-1 belongs.

(作用) 本発明によれば、処理割り当て装置の、プロセサ割り当
てテーブルにはプロセサへの処理の割り当て状況が格納
され、ローカルメモリ状況格納部には各ローカルメモリ
が格納するデータの格納状況が格納されている。そして
、割り当て部はプロセサ割り当てテーブルに基づいて処
理を各プロセサに割り当てると共に上記ローカルメモリ
状況格納部のデータに基づき、処理が一のローカルメモ
リの格納データを使用できると判断した時には当該ロー
カルメモリの属するプロセサに優先的に処理を割り当て
るから、無駄なデータの転送はしなくてもすみ、全体と
して処理が迅速なものとなる。
(Operation) According to the present invention, the processor allocation table of the processing allocation device stores the allocation status of processes to processors, and the local memory status storage section stores the storage status of data stored in each local memory. ing. Then, the allocation unit allocates the process to each processor based on the processor allocation table, and when it is determined that the process can use the data stored in one local memory based on the data in the local memory status storage unit, the allocation unit allocates the process to each processor based on the processor allocation table. Since processing is assigned to the processor with priority, there is no need to transfer unnecessary data, and overall processing becomes faster.

(実施例) 以下本発明に係る並列計算機の処理割り当て装置の実施
例を図面に基づいて説明する。
(Embodiment) Hereinafter, an embodiment of a processing allocation device for a parallel computer according to the present invention will be described based on the drawings.

第2図乃至第5図は本発明に係る並列計算機の処理割り
当て装置の実施例を示すものである。本実施例において
、処理割り当て装置31には複数のプロセサ(図示して
いない)への処理の割り当て状況を格納したプロセサ割
り当てテーブル34と、各ローカルメモリ(図示してい
ない)が格納するデータの格納状況を格納するローカル
メモリ状況格納部35と、上記プロセサ割り当てテーブ
ル34に基づいて処理を各プロセサに割り当てると共に
上記ローカルメモリ状況格納部35のデータに基づき、
処理が一のローカルメモリの格納データを使用できると
判断した時には当該ローカルメモリの属するプロセサに
優先的に処理を割り当てる割り当て部36とからなる。
2 to 5 show an embodiment of a processing allocation device for a parallel computer according to the present invention. In this embodiment, the processing allocation device 31 includes a processor allocation table 34 that stores the allocation status of processing to a plurality of processors (not shown), and a storage of data stored in each local memory (not shown). A local memory status storage unit 35 that stores the status, and processing is assigned to each processor based on the processor allocation table 34, and based on the data of the local memory status storage unit 35,
It consists of an allocation section 36 which, when it is determined that data stored in a local memory can be used for processing, preferentially allocates the processing to the processor to which the local memory belongs.

本実施例においてローカルメモリ状況格納部35は、第
3図に示すように、各ローカルメモリが属するプロセサ
番号に対するポインタにどの処理環境が格納されている
かを格納している。この例においてはプロセサ番号lの
ローカルメモリには処理98、プロセサ番号2のローカ
ルメモリには処理99及び処理100、プロセサ番号3
のローカルメモリにはどの処理も格納されず、プロセサ
番号nのローカルメモリには処理123の環境が構築さ
れていることを示している。
In this embodiment, the local memory status storage unit 35 stores which processing environment is stored in a pointer to a processor number to which each local memory belongs, as shown in FIG. In this example, the local memory of processor number l has process 98, the local memory of processor number 2 has processes 99 and 100, and processor number 3 has process 99 and process 100.
This shows that no process is stored in the local memory of processor number n, and an environment for process 123 is built in the local memory of processor number n.

また、本実施例においてプロセサ割り当てテーブルは、
第4図に示すように、各プロセサの処理状況を格納して
いる。この例ではプロセサ1及びプロセサ3は空状態(
Ready状態)であり、プロセサ2は処理99をまた
、プロセサnは処理123を実行中であることが格納さ
れている。
Furthermore, in this embodiment, the processor allocation table is
As shown in FIG. 4, the processing status of each processor is stored. In this example, processor 1 and processor 3 are in an empty state (
Ready state), and it is stored that processor 2 is executing process 99 and processor n is executing process 123.

そして割り当て部36はこれらの情報を基に評価関数を
用いて処理要求に対して処理を実行させるプロセサを決
定してプロセサ番号を出力することによって処理を割り
当てる。
Based on this information, the allocation unit 36 uses an evaluation function to determine a processor to execute a process in response to a process request, and allocates the process by outputting a processor number.

次に本実施例に係る並列計算機の処理割り当て装置の作
動を説明する。
Next, the operation of the processing allocation device for a parallel computer according to this embodiment will be explained.

第5図は本発明に係る並列計算機の処理割り当て装置の
作動を示すフローチャートである。
FIG. 5 is a flowchart showing the operation of the processing allocation device for a parallel computer according to the present invention.

まず、処理の要求が来ると、割り当て都は要求処理の環
境がどのプロセサ上にあるか知るためにローカルメモリ
状況格納部を見る(STI)。
First, when a processing request comes, the allocator looks at the local memory status store (STI) to find out which processor the request processing environment is on.

どのプロセサにもその処理を実行する環境が無いときに
は通常の処理を行ない、どこかのプロセサに処理環境が
格納されている時には、プロセサ割り当てテーブルを参
照して、そのプロセサに他の処理が割り当てられている
かどうかを判定し(ST2),他の処理が割り当てられ
ていないときにはそのプロセサに処理を割り当てる。こ
のとき当該プロセサに他の処理が割り当てられていると
きには、なるべくローカルメモリの内容を書換なくとも
よいプロセサを探し(ST3),当該プロセサに処理を
割振る。
If no processor has an environment to execute the process, normal processing is performed, and if a processing environment is stored in some processor, another process is assigned to that processor by referring to the processor allocation table. (ST2), and if no other processing is assigned, the processing is assigned to that processor. At this time, if another process is assigned to the processor, a processor that does not require rewriting the contents of the local memory is searched for (ST3), and the process is assigned to the processor.

例えば第3図及び第4図に示す状態において処理98の
要求があった時には、割り当て部はプロセサ3ではなく
、ローカルメモリに処理98の環境を有するプロセサ1
に処理を割り当てる。
For example, when there is a request for the process 98 in the state shown in FIGS. 3 and 4, the allocation unit is not the processor 3, but the processor 1 that has the environment of the process 98 in the local memory.
Assign processing to.

これにより無駄なデータの伝送は行なわなくともすむ。This eliminates the need for unnecessary data transmission.

従って本実施例によれば、処理は当該処理を行なうこと
ができる環境を有するプロセサに優先して割り当てられ
るので、無駄なデータの転送はしなくてもすみ、全体と
して処理が迅速なものとなる。
Therefore, according to this embodiment, processing is prioritized and assigned to a processor that has an environment that can perform the processing, so there is no need to transfer unnecessary data, and the processing as a whole becomes faster. .

尚、上記実施例では共膚メモリの格納内容についても制
御を行なう必要があるが、これについては共有メモリも
ローカルメモリの1つとして一般化して考えることがで
きる。また、各プロセサがローカルメモリではなくキャ
ッシュメモリを持つ場合があるが、この場合にはキャッ
シュメモリを小さい容量のローカルメモリと考え本発明
を適用することができる。
In the above embodiment, it is also necessary to control the contents stored in the common memory, but in this regard, the shared memory can also be generalized as one type of local memory. Further, each processor may have a cache memory instead of a local memory, and in this case, the present invention can be applied by considering the cache memory as a small-capacity local memory.

更に上記実施例においては並列計算機の処理割り当て装
置として、各プロセサとは別の装置を想定して説明した
が、一のプロセサを処理割り当て装置としても作動させ
、ローカルメモリにローカルメモリ状況格納部とプロセ
サ割り当てテーブルを格納して処理することもできる。
Furthermore, although the above embodiment has been explained assuming that a device separate from each processor is used as a processing allocation device for a parallel computer, one processor is also operated as a processing allocation device, and the local memory is provided with a local memory status storage section. Processor allocation tables can also be stored and processed.

〔発明の効果〕〔Effect of the invention〕

以上説明したように、本発明によれば、複数のプロセサ
を有し、各プロセサにローカルメモリを設けてなる並列
計算機の各プロセサへ処理を割り当てる処理割り当て装
置を、プロセサへの処理の割り当て状況を格納したプロ
セサ割り当てテーブルと、各ローカルメモリが格納する
データの格納状況を格納するローカルメモリ状況格納部
と、上記プロセサ割当テーブルに基づいて処理を各プロ
セサに割り当てると共に上記ローカルメモリ状況格納部
のデータに基づき、処理が一のローカルメモリの格納デ
ータを使用できると判断した時には当該ローカルメモリ
の属するプロセサに優先的に処理を割当てる割り当て部
を設けるものとしたから、処理が一のローカルメモリの
格納データを使用できると判断した時には当該ローカル
メモリの属するプロセサに優先的に処理を割り当て、無
駄なデータの転送はしなくてもすみ、全体として処理が
迅速なものとなるという効果を奏する。
As explained above, according to the present invention, a process allocation device that allocates processes to each processor of a parallel computer having a plurality of processors and each processor is provided with a local memory can be used to monitor the status of process allocation to the processors. The stored processor allocation table, a local memory status storage unit that stores the storage status of data stored in each local memory, and a local memory status storage unit that allocates processing to each processor based on the processor allocation table and the data in the local memory status storage unit. Based on this, when a process determines that data stored in one local memory can be used, an allocation unit is provided that assigns the process preferentially to the processor to which the local memory belongs. When it is determined that the local memory can be used, processing is preferentially assigned to the processor to which the local memory belongs, eliminating the need for unnecessary data transfer and resulting in faster processing overall.

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

第1図は本発明の原理図、第2図は本発明に係る並列計
算機の処理割り当て装置の実施例を示す図、第3図は第
2図に示した並列計算機の処理割り当て装置のローカル
メモリ状況格納部を示す図、第4図は第2図に示した並
列計算機の処理割り当て装置のプロセサ割り当てテーブ
ルを示す図、第5図は第2図に示した並列計算機の処理
割り当て装置の作動を示すフローチャート、第6図及び
第7図は本発明が適用される並列計算機を示す図、第8
図は第7図に示した並列計算機のローカルメモリの格納
状態を示す図である。 1−1〜L.・・・プロセサ 2−1〜lo・・・ローカルメモリ 3・・・処理割り当て装置 4・・・プロセサ割り当てテーブル 5・・・ローカルメモリ状況格納部 6・・・割り当て部 本発明の厭樒■ 第 IEI 第 2 図 35 笛 3 図 ラ;方已イタ・j のイ乍動 第5図 プロセサ枦1り当て干一っ゜ルのr’3’l−第4図 11−2 1 木金明か゛i&IJ1!どバろシヌ寸ム第6図 17−2 7−N ローカルメモリのA人2危 第 8 図 −400
Fig. 1 is a diagram showing the principle of the present invention, Fig. 2 is a diagram showing an embodiment of the processing allocation device for a parallel computer according to the invention, and Fig. 3 is a local memory of the processing allocation device for a parallel computer shown in Fig. 2. FIG. 4 shows the processor allocation table of the processing allocation device for the parallel computer shown in FIG. 2. FIG. 5 shows the operation of the processing allocation device for the parallel computer shown in FIG. 2. 6 and 7 are flow charts showing a parallel computer to which the present invention is applied, and FIG.
This figure shows the storage state of the local memory of the parallel computer shown in FIG. 7. 1-1~L. ...Processors 2-1 to lo...Local memory 3...Processing allocation device 4...Processor allocation table 5...Local memory status storage section 6...Allocation section IEI 2nd Figure 35 Flute 3 Figure A; Direction Ita・j's I movement Figure 5 Processor holder 1 Hitting one round r'3'l-Figure 4 11-2 1 Wooden metal light i & IJ1 ! Dobaroshinu Dimensions Figure 6 17-2 7-N Local Memory Person A 2nd Crisis Figure 400

Claims (1)

【特許請求の範囲】 複数のプロセサ(1_−_1〜1_−_n)を有し、各
プロセサ(1_−_1〜1_−_n)にローカルメモリ
(2_−_1〜2_−_n)を設けてなる並列計算機の
各プロセサ(1_−_1〜1_−_n)へ処理を割り当
てる処理割り当て装置(3)であって、 プロセサ(1_−_1〜1_−_n)への処理の割り当
て状況を格納したプロセサ割り当てテーブル(4)と、 各ローカルメモリ(2_−_1〜2_−_n)が格納す
るデータの格納状況を格納するローカルメモリ状況格納
部(5)と、 上記プロセサ割り当てテーブル(4)に基づいて処理を
各プロセサ(1_−_1〜1_−_n)に割り当てると
共に上記ローカルメモリ状況格納部(5)のデータに基
づき、処理が一のローカルメモリ(2_−_i)の格納
データを使用できると判断した時には当該ローカルメモ
リ(2_−_i)の属するプロセサ(1_−_i)に優
先的に処理を割り当てる割り当て部(6)と、 からなることを特徴とする並列計算機の処理割り当て装
置。
[Claims] A parallel system having a plurality of processors (1_-_1 to 1_-_n) and each processor (1_-_1 to 1_-_n) provided with a local memory (2_-_1 to 2_-_n). A process allocation device (3) that allocates processes to each processor (1_-_1 to 1_-_n) of a computer, comprising a processor allocation table (3) that stores the status of process allocation to the processors (1_-_1 to 1_-_n). 4); a local memory status storage unit (5) that stores the storage status of data stored in each local memory (2_-_1 to 2_-_n); (1_-_1 to 1_-_n), and when it is determined that the process can use the data stored in one local memory (2_-_i) based on the data of the local memory status storage unit (5), the local memory A processing allocation device for a parallel computer, comprising: an allocation unit (6) that preferentially allocates processing to a processor (1_-_i) to which (2_-_i) belongs.
JP23136889A 1989-09-08 1989-09-08 Process assignment device for parallel computer Pending JPH0395661A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP23136889A JPH0395661A (en) 1989-09-08 1989-09-08 Process assignment device for parallel computer

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP23136889A JPH0395661A (en) 1989-09-08 1989-09-08 Process assignment device for parallel computer

Publications (1)

Publication Number Publication Date
JPH0395661A true JPH0395661A (en) 1991-04-22

Family

ID=16922525

Family Applications (1)

Application Number Title Priority Date Filing Date
JP23136889A Pending JPH0395661A (en) 1989-09-08 1989-09-08 Process assignment device for parallel computer

Country Status (1)

Country Link
JP (1) JPH0395661A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011198122A (en) * 2010-03-19 2011-10-06 Fujitsu Ltd Device and method for assigning process, and computer program
JP2014006657A (en) * 2012-06-22 2014-01-16 Fujitsu Ltd Job management program, job management method, and job management device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011198122A (en) * 2010-03-19 2011-10-06 Fujitsu Ltd Device and method for assigning process, and computer program
JP2014006657A (en) * 2012-06-22 2014-01-16 Fujitsu Ltd Job management program, job management method, and job management device

Similar Documents

Publication Publication Date Title
US9965324B2 (en) Process grouping for improved cache and memory affinity
JP2561760B2 (en) Process in information processing system and information processing network
JP6755935B2 (en) Shared memory controller and how to use it
US5884077A (en) Information processing system and method in which computer with high load borrows processor of computer with low load to execute process
US5784698A (en) Dynamic memory allocation that enalbes efficient use of buffer pool memory segments
JPH02247768A (en) Dispersed processing system
EP0523863B1 (en) Digital data processor for high level instructions
JPH0570177B2 (en)
JPH0330897B2 (en)
JPH06236344A (en) Method and apparatus for arbitration between plurality of data transfer requests
EP0317481B1 (en) Remote storage management mechanism and method
US11093291B2 (en) Resource assignment using CDA protocol in distributed processing environment based on task bid and resource cost
EP0317468A2 (en) Bus flow control system
JPH04354038A (en) Address conversion register control system
JPH0395661A (en) Process assignment device for parallel computer
JP2003517681A (en) Shared resource arbitration method and apparatus
JP3733402B2 (en) Processor resource selection method, processor resource selection system therefor, and computer-readable program recording medium
JP2008276322A (en) Information processing device, system, and method
JP2780662B2 (en) Multiprocessor system
JP2005266841A (en) Computer system equipped with exclusive processor
JPH0612395A (en) Task allocating method in multiprocessor system
JPH1078942A (en) Multiprocessor system
JPH02245864A (en) Multiprocessor system
JP2830780B2 (en) Multiprocessor system
US6959361B2 (en) Distributed caching mechanism for pending memory operations within a memory controller