JPS6331818B2 - - Google Patents
Info
- Publication number
- JPS6331818B2 JPS6331818B2 JP57032826A JP3282682A JPS6331818B2 JP S6331818 B2 JPS6331818 B2 JP S6331818B2 JP 57032826 A JP57032826 A JP 57032826A JP 3282682 A JP3282682 A JP 3282682A JP S6331818 B2 JPS6331818 B2 JP S6331818B2
- Authority
- JP
- Japan
- Prior art keywords
- job
- queue
- processor
- time
- tij
- 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.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multi Processors (AREA)
Description
【発明の詳細な説明】
(1) 発明の技術分野
本発明は疎結合計算機システムにおける各プロ
セツサのジヨブクラス毎の処理時間、ジヨブキユ
ーの処理数を把握することにより、イニシエータ
からのジヨブ取り出し要求時、該プロセツサで実
行すれば最も早く終了すると思われるジヨブを選
択する方式に関する。[Detailed Description of the Invention] (1) Technical Field of the Invention The present invention grasps the processing time and number of job queues processed for each job class of each processor in a loosely coupled computer system. This relates to a method for selecting a job that is expected to finish the fastest if executed by a processor.
(2) 従来技術と問題点
従来、ジヨブキユーを共有する疎結合計算機シ
ステムにおいては、ジヨブがどのプロセツサで実
行されるかは、その時のプロセツサの動作状況に
依存していた。従つて高速プロセツサと低速プロ
セツサから疎結合計算機システムが構成されてい
る場合、ジヨブのターンアラウンドが実行される
プロセツサに依存し、バラツキが出るとともに、
ジヨブキユーのプライオリテイが余りジヨブ制御
機能を果さなくなるという欠点があつた。(2) Prior Art and Problems Conventionally, in loosely coupled computer systems that share job queues, which processor executes a job depends on the operating status of the processor at that time. Therefore, when a loosely coupled computer system is configured with a high-speed processor and a low-speed processor, job turnaround will depend on the processor being executed, and there will be variations.
There was a drawback that the priority of the job queue did not perform much of a job control function.
(3) 発明の目的
本発明は前記欠点を解消し、ジヨブキユー上の
順にジヨブを終了させる方式を提供することを目
的とする。(3) Object of the Invention It is an object of the present invention to solve the above-mentioned drawbacks and to provide a method for terminating jobs in the order on the job queue.
(4) 発明の構成
該目的は疎結合プロセツサからなるシステムに
おいて、ジヨブキユー上のジヨブに関して、該ジ
ヨブの実行時間と該ジヨブのジヨブキユー上での
待ち時間の和が最小になる、あるプロセツサに該
ジヨブを割り付けることを特徴とするジヨブ選択
方式により達成される。(4) Structure of the Invention The object is to, in a system consisting of loosely coupled processors, assign a job to a processor that minimizes the sum of the execution time of the job and the waiting time of the job on the job queue. This is achieved by a job selection method characterized by allocating .
(5) 発明の実施例 以下、本発明を図面を用いて詳細に説明する。(5) Examples of the invention Hereinafter, the present invention will be explained in detail using the drawings.
第１図は本発明の一実施例を示す全体ブロツク
図である。 FIG. 1 is an overall block diagram showing one embodiment of the present invention.
図において、１１はシステムＡ、１２はシステ
ムＢ、１３はシステムＣ、１４はジヨブキユー
（ジヨブクラスに分けられ複数個存在する。）Iij
はｉシステムのｊ番目のイニシエータ、JOB1〜
JOB3はジヨブ名称である。 In the figure, 11 is system A, 12 is system B, 13 is system C, and 14 is job queue (divided into job classes, there are multiple job classes) Iij
is the j-th initiator of the i system, JOB 1 ~
JOB 3 is the job name.
第２図は、本発明の一実施例を示すジヨブ選択
機能のフローチヤートである。 FIG. 2 is a flowchart of a job selection function illustrating one embodiment of the present invention.
図において、２１はTijの計算部（システムｉ
においてジヨブｊを実行した時の実行終了時刻、
２２はTijの最小システム決定部である。 In the figure, 21 is the calculation unit of Tij (system i
The execution end time when job j is executed in
22 is the minimum system determining unit of Tij.
各システム（システムＡ１１、システムＢ１
２、システムＣ１３）にはイニシエータ（Iij）
が複数個存在している。ここでシステムとは疎結
合プロセツサをいう。またイニシエータとジヨブ
クラスの対応はｎ：ｎ対応（ｎは任意の整数）で
あるが、一般にはｎ：１対応が多く、以下はｎ：
１対応で説明する。 Each system (system A11, system B1
2. System C13) has an initiator (Iij)
There are multiple. Here, the system refers to a loosely coupled processor. Also, the correspondence between initiators and job classes is n:n (n is any integer), but in general there is often n:1 correspondence, and the following is n:
This will be explained using 1 correspondence.
各システムにおけるジヨブ処理状況を動的に把
握し、システム毎に該クラスに対する平均実行時
間を把握する場合、ジヨブキユー１４上の１ジヨ
ブがシステムに取り上げられる時間をＱとすると
Ｑ＝Ｅ／Ｉとなる。但し、Ｅはジヨブが実行され
てから終了するまでの時間、Ｉはイニシエータの
数とする。 When dynamically grasping the job processing status in each system and grasping the average execution time for the class for each system, if the time taken for one job on the job queue 14 to be picked up by the system is Q, then Q=E/I. . Here, E is the time from execution to completion of the job, and I is the number of initiators.
次にジヨブ選択機能について、システムＣのイ
ニシエータからジヨブが要求された時、如何にし
てジヨブキユー１４上の各ジヨブ（JOBj（ｊ＝１
〜３））を選択するかについて述べる。 Next, regarding the job selection function, when a job is requested from the initiator of system C, how can each job (JOBj (j=1
~3)) We will discuss whether to choose.
尚、各ジヨブはｊの正昇順にキユーされている
ものとする。 It is assumed that each job is queued in ascending order of j.
ここで、Eiはシステムｉにおけるジヨブの実行
時間、Qiは、システムｉにおけるジヨブの取出
し時間、Aiは今回のジヨブ選択で、システムｉ
において該ジヨブを処理するまでに、処理しなけ
ればいけないジヨブ数であり、初期値はAc＝０，
Ai＝0.5（ｉ≠ｃ）JOBjを実行した時の実行終了
時刻（現時刻を０とする。）として説明する。 Here, Ei is the job execution time in system i, Qi is the job retrieval time in system i, Ai is the current job selection, and system i
It is the number of jobs that must be processed before processing the job in , and the initial value is Ac=0,
Ai=0.5 (i≠c) The execution end time when JOBj is executed (the current time is assumed to be 0) will be explained.
まず、ジヨブキユー１４上のジヨブに対して順
に、Tijの計算部２１でTij＝Ei＋AiQiとして計
算する。次にJOBjにとつてTijが最小のシステム
ｍをTijの最小システム決定部２２により求める。
そしてｍ＝ｃならJOBjを選択する。又ｍ≠ｃな
らAmに１を加えて、ジヨブキユー上の次のジヨ
ブつまりJOBj＋１に前記手順（Tijの最小を求め
る）を繰返す。 First, the Tij calculation unit 21 calculates Tij=Ei+AiQi for the jobs on the job queue 14 in order. Next, the system m with the minimum Tij for JOBj is determined by the minimum system determination unit 22 of Tij.
Then, if m=c, select JOBj. If m≠c, add 1 to Am and repeat the above procedure (finding the minimum Tij) for the next job on the job queue, that is, JOBj+1.
(6) 発明の効果
以上説明したように、本発明によれば実行され
る疎結合プロセツサシステムに依存せず、略ジヨ
ブキユーの順に各ジヨブの実行が終了するという
効果がある。(6) Effects of the Invention As explained above, according to the present invention, the execution of each job is completed in approximately the order of the job queue, regardless of the loosely coupled processor system in which it is executed.
第１図は本発明の一実施例を示す全体ブロツク
図である。第２図は、本発明の一実施例を示すジ
ヨブ選択機能のフローチヤートである。
記号の説明、１１はシステムＡ、１２はシステ
ムＢ、１３はシステムＣ、１４はジヨブキユー
（ジヨブクラスに分けられ複数個存在する。）Iij
はシステムのｊ番目のイニシエータ、JOB1〜
JOB3はジヨブ名称。
FIG. 1 is an overall block diagram showing one embodiment of the present invention. FIG. 2 is a flowchart of a job selection function illustrating one embodiment of the present invention. Explanation of symbols: 11 is system A, 12 is system B, 13 is system C, 14 is job queue (divided into job classes, there are multiple jobs) Iij
is the jth initiator of the system, JOB 1 ~
JOB 3 is the job name.
Claims (1)
て、ジヨブキユー上のジヨブに関して、該ジヨブ
の実行時間と該ジヨブのジヨブキユー上での待ち
時間の和が最小になる、あるプロセツサに該ジヨ
ブを割り付けることを特徴とするジヨブ選択方
式。1. In a system consisting of loosely coupled processors, a job characterized by assigning a job to a certain processor that minimizes the sum of the execution time of the job and the waiting time of the job on the job queue with respect to the job on the job queue. Selection method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP3282682A JPS58151664A (en) | 1982-03-02 | 1982-03-02 | Job selecting system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP3282682A JPS58151664A (en) | 1982-03-02 | 1982-03-02 | Job selecting system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS58151664A JPS58151664A (en) | 1983-09-08 |
| JPS6331818B2 true JPS6331818B2 (en) | 1988-06-27 |
Family
ID=12369626
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP3282682A Granted JPS58151664A (en) | 1982-03-02 | 1982-03-02 | Job selecting system |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS58151664A (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS6448119A (en) * | 1987-08-19 | 1989-02-22 | Hitachi Ltd | Semiconductor device |
| US5121494A (en) * | 1989-10-05 | 1992-06-09 | Ibm Corporation | Joining two database relations on a common field in a parallel relational database field |
| JP2001101149A (en) * | 1999-09-30 | 2001-04-13 | Nec Corp | Distributed parallel data processor, recording medium recording distributed parallel data processing program and distributed parallel data processing system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS57146367A (en) * | 1982-02-05 | 1982-09-09 | Hitachi Ltd | Decentralized processing system |
-
1982
- 1982-03-02 JP JP3282682A patent/JPS58151664A/en active Granted
Also Published As
| Publication number | Publication date |
|---|---|
| JPS58151664A (en) | 1983-09-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3989911B2 (en) | Virtualization of global interrupt queue | |
| CN109086135B (en) | Resource scaling method and device, computer equipment and storage medium | |
| JP7461947B2 (en) | Latency-aware dynamic priority changing in a processor - Patents.com | |
| US6473780B1 (en) | Scheduling of direct memory access | |
| JPS62229358A (en) | Processor selection system | |
| JP4912927B2 (en) | Task allocation apparatus and task allocation method | |
| JPS6331818B2 (en) | ||
| CN110308988B (en) | Dynamic scheduling method and system applied to heterogeneous computing platform | |
| KR101639947B1 (en) | Hadoop preemptive deadline constraint scheduling method, execution program thereof method and recorded medium of the program | |
| CN115167973B (en) | Data processing system of cloud computing data center | |
| JPH0612395A (en) | Task allocation method in multiprocessor system | |
| Yabuuchi et al. | Multi-resource low-latency cluster scheduling without execution time estimation | |
| JPH02281338A (en) | Task control method | |
| JP3374480B2 (en) | Data processing device | |
| JPH0769840B2 (en) | Lock management controller | |
| KR0164139B1 (en) | Red queue management method to improve the real-time of operating system | |
| KR940003846B1 (en) | Scheduling Scheme of Real-time Processing of Electronic Switching System | |
| JPH0460745A (en) | Load diffusion system for multi-cpu system | |
| JPH08272739A (en) | Method for improving on-line response | |
| JPS63317833A (en) | Priority control processing system | |
| CN85104907A (en) | Hierarchy multprocessor system and control method thereof | |
| JPH0877029A (en) | Processing request execution order control method based on load factor | |
| JPH0447357A (en) | Virtual computer system control method | |
| JPH08329022A (en) | I / O processing load balancing control method for multiprocessor system | |
| JPH09128321A (en) | User interface device |