JPS5819958A - Degenerating method for computer - Google Patents
Degenerating method for computerInfo
- Publication number
- JPS5819958A JPS5819958A JP56118849A JP11884981A JPS5819958A JP S5819958 A JPS5819958 A JP S5819958A JP 56118849 A JP56118849 A JP 56118849A JP 11884981 A JP11884981 A JP 11884981A JP S5819958 A JPS5819958 A JP S5819958A
- Authority
- JP
- Japan
- Prior art keywords
- task
- execution
- queue
- tasks
- computer
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
【発明の詳細な説明】
この発明は計算機の負荷が異常に高くなった場合に計算
機を効率よく動作させる為の計算機の縮退方法に関する
ものである。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a computer degeneration method for efficiently operating a computer when the load on the computer becomes abnormally high.
計算機のシステム設計においては、通常の運用上支障が
ない程度の処理能力を持ったシステムに設計するのであ
る一!、外部条件が変動した場合或は異常事態が発生し
た場合等に計算機の負荷が異常に高くなって計算機で処
理しきれなくなり、システムの重故障を招来する場合が
あった。When designing a computer system, the system should be designed to have enough processing power to cause no problems in normal operation! When external conditions change or when an abnormal situation occurs, the load on the computer becomes abnormally high and the computer cannot handle the processing, resulting in a serious system failure.
この発明は従来の装置における上述の欠点を除去するた
めになされたもので、計算機が実行すべきタスク(ta
sk ) にそのタスクの種類によりあらかじめ実行
優先度を付けておき、計算機の負荷が異常に高くなった
場合は実行優先度の低いタスクの実行を取止めることに
よって、システムの重故障を回避することができる計算
機の縮退方法を提供することを目的としている。This invention was made in order to eliminate the above-mentioned drawbacks of conventional devices.
sk)) is assigned an execution priority in advance according to the type of task, and if the load on the computer becomes abnormally high, the execution of tasks with lower execution priority is stopped, thereby avoiding serious system failures. The purpose is to provide a computer degeneracy method that allows
以下、図面についてこの発明の詳細な説明する。図面は
この発明の一実施例を示すブロック図で、(1)はタス
クの待行列(task queue ) s (2)
はタスク応答スケジュラ(task response
5ehedulsr(3)Fiタスク実行スケジュラ
(task exaeutionscheduler
)である。Hereinafter, the present invention will be described in detail with reference to the drawings. The drawing is a block diagram showing an embodiment of the present invention, in which (1) is a task queue (2)
is a task response scheduler.
5ehedulsr (3) Fi task execution scheduler
).
次に動作について説明する。計算機システムの仕事の処
理単位をタスクと呼び、1つのタスクの実行要求が発生
するとタスク応答スケジュラ(2)はこれをタスクの待
行列(1)に登録する。すなわち、記憶に書込む。この
登録は実行要求が発せられタスク応答スケジュラ(2)
kよって応答された順位、すなわち実行要求発生順に行
われかつ当該タスクに対しあらかじめ定められた実行優
先度を付して行われる。この実行優先度の最も簡単な例
としては優先度1と2に分け、優先度1のタスクは計算
機の負荷如何に関せず必ず実行すべきタスク、優先度2
のタスクは実行し々くても重故障を招来することのない
タスクとする。計算機の負荷が所定値以上になったこと
はタスクの待行列(1)K登録されたタスクの総数が所
定値(以下Nで表す)以上になったことKよって判定す
る。Next, the operation will be explained. A processing unit of work in a computer system is called a task, and when a request for execution of one task occurs, the task response scheduler (2) registers it in the task queue (1). That is, write to memory. This registration is executed by the task response scheduler (2) when an execution request is issued.
The task is executed in the order responded by k, that is, in the order in which the execution request was generated, and the task is executed with a predetermined execution priority given to the task. The simplest example of this execution priority is to divide it into priorities 1 and 2, with priority 1 tasks being tasks that must be executed regardless of the computer load, and priority 2 tasks.
These tasks should not cause serious failures even if they are executed frequently. Whether the load on the computer has exceeded a predetermined value is determined by the fact that the total number of tasks registered in the task queue (1) K has exceeded a predetermined value (hereinafter referred to as N).
タスク応答スケジュラ(2)はタスクの待行列(11を
)、調べ、登録されているタスクのうち実行要求発生順
の最先のタスクをタスク実行スケジュラ(3)により実
行させる。実行が終了するとその終了したタスクはタス
クの待行列(1)から消去される。The task response scheduler (2) examines the task queue (11), and causes the task execution scheduler (3) to execute the task that is first in the order of execution requests among the registered tasks. When execution is completed, the completed task is deleted from the task queue (1).
タスクの待行列(1)に登録されているタスクの総数が
Nになると、タスク応答スケジュラ(2)はタスクの待
行列(1)K登録されているタスクのうちから最も実行
優先度が低くかつ実行要求発生順が最後のタスクを消去
する。したがって、タスクの待行列(11に登録されて
いるタスクの総数は(N−1)K保たれる。先に説明し
た例のように実行優先度が1と2の2段階に分れている
とし、タスクの待行列(1)に登録されているタスクの
総数が(N−1)のときに優先度2のタスク(以下タス
ク人とする)の実行要求が発生するとこのタスクAFi
タスクの待行列(1)に−たん登録されるが、その時タ
スクの総数はNとなるので、タスク応答スケジュラ(2
)は優先度2で最後に登録されたタスクAを消去し、す
なわちタスクAは登録されないことと同じ結果になる。When the total number of tasks registered in the task queue (1) reaches N, the task response scheduler (2) selects the task with the lowest execution priority from among the K registered tasks in the task queue (1). Delete the last task in the order of execution requests. Therefore, the total number of tasks registered in the task queue (11) is maintained at (N-1)K.As in the example explained earlier, the execution priority is divided into two levels, 1 and 2. Then, when the total number of tasks registered in the task queue (1) is (N-1), when a request for execution of a task with priority 2 (hereinafter referred to as task person) occurs, this task AFi
The tasks are registered in the task queue (1), but at that time the total number of tasks is N, so the task response scheduler (2)
) deletes task A, which was last registered with priority 2, which has the same result as task A not being registered.
またタスクの総数が(N−1)のときに優先f1のタス
クの実行要求が発生するとこのタスクは登録され代りに
優先度2のタスクが1つ消去される。Further, when the total number of tasks is (N-1), when a request for execution of a task with priority f1 occurs, this task is registered and one task with priority 2 is deleted in its place.
優先度2のタスクは実行を省略されてもシステムの重故
障につながらないタスクであるので、この実行を省略し
計算機の縮退を行って計算機の異常な高負荷を避けるこ
とができる。Since priority 2 tasks are tasks that will not cause a major system failure even if their execution is omitted, abnormally high loads on the computer can be avoided by omitting their execution and degenerating the computer.
以上のようkこの発明和よれば、計算機の負荷が異常に
大きくなったときは、あらかじめ計画したとおりの計算
機の縮退が自動的に行われてシステムの重故障を避ける
ことができる。したがって正常な運転状態下において処
理すべき事項が同一である場合、従来のシステムよりも
少ないハードウェアの量でこの発明のシステムを設計す
ることができる。As described above, according to this invention, when the load on a computer becomes abnormally large, the computer is automatically degenerated as planned in advance, thereby avoiding a serious system failure. Therefore, under normal operating conditions, the system of the present invention can be designed with less hardware than conventional systems when the tasks to be handled are the same.
図面はこの発明の一実施例を示すブロック図である。
(1)・・・タスクの待行列、(2)・・・タスク応答
スケジュラ、(3)・・・タスク実行スケジュラ。
代理人 葛 野 信 −The drawing is a block diagram showing an embodiment of the present invention. (1)...Task queue, (2)...Task response scheduler, (3)...Task execution scheduler. Agent Shin Kuzuno −
Claims (1)
の実行要求が発生するととに実行要求発生順を付しかつ
当該タスクに対しあらかじめ定めた実行優先度を付して
タスクの待行列中に登録する段階と、 タスク応答スケジュラにより上記タスクの待行列を調べ
待行列中和記憶されているタスクのうち実行要求発生順
が最先のタスクをタスク実行スケジュラにより実行しこ
の実行が終了したタスクを上記タスクの待行列から消去
する段階と、上記タスクの待行列に登録されているタス
クの総数が所定値に達した場合上記タスク応答スケシェ
ラは上記タスクの待行列に登録されているタス(1) りの中から最も実行優先度が低くかつ実行要求発生順が
最後のタスクを消去する段階とを備えた計算機の縮退方
法。[Claims] One unit of work to be executed by a computer is defined as a task, and when a task execution request occurs, the order of occurrence of the execution request is attached to the request, and a predetermined execution priority is attached to the task. There is a step of registering a task in the queue, and a task response scheduler checks the queue of the above tasks, neutralizes the queue, executes the task whose execution request is first in the order of occurrence among the stored tasks, and executes this task with the task execution scheduler. a step of deleting the task whose execution has been completed from the task queue; and when the total number of tasks registered in the task queue reaches a predetermined value, the task response scheduler is registered in the task queue. (1) A method for degenerating a computer, which includes the step of deleting the task with the lowest execution priority and the last task in the order of execution requests.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56118849A JPS5819958A (en) | 1981-07-29 | 1981-07-29 | Degenerating method for computer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56118849A JPS5819958A (en) | 1981-07-29 | 1981-07-29 | Degenerating method for computer |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS5819958A true JPS5819958A (en) | 1983-02-05 |
Family
ID=14746657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP56118849A Pending JPS5819958A (en) | 1981-07-29 | 1981-07-29 | Degenerating method for computer |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS5819958A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6190253A (en) * | 1984-10-09 | 1986-05-08 | Fujitsu Ltd | Information processing method and its device using system with queue |
US5511214A (en) * | 1992-08-25 | 1996-04-23 | Fujitsu Limited | On-line processing system and overload suppressing method |
WO2000022498A1 (en) * | 1998-10-14 | 2000-04-20 | Hitachi, Ltd. | Method and device for stopping job during system freeze |
JP2008278009A (en) * | 2007-04-26 | 2008-11-13 | Nippon Telegr & Teleph Corp <Ntt> | Resource reservation control method, device and program |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5197950A (en) * | 1975-02-26 | 1976-08-28 | Kanripuroguramuno seigyohoshiki | |
JPS5475244A (en) * | 1977-11-29 | 1979-06-15 | Fujitsu Ltd | Inter-subsystem communication system equipped with dequeue promoting function |
-
1981
- 1981-07-29 JP JP56118849A patent/JPS5819958A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5197950A (en) * | 1975-02-26 | 1976-08-28 | Kanripuroguramuno seigyohoshiki | |
JPS5475244A (en) * | 1977-11-29 | 1979-06-15 | Fujitsu Ltd | Inter-subsystem communication system equipped with dequeue promoting function |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6190253A (en) * | 1984-10-09 | 1986-05-08 | Fujitsu Ltd | Information processing method and its device using system with queue |
US5511214A (en) * | 1992-08-25 | 1996-04-23 | Fujitsu Limited | On-line processing system and overload suppressing method |
WO2000022498A1 (en) * | 1998-10-14 | 2000-04-20 | Hitachi, Ltd. | Method and device for stopping job during system freeze |
JP2008278009A (en) * | 2007-04-26 | 2008-11-13 | Nippon Telegr & Teleph Corp <Ntt> | Resource reservation control method, device and program |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6662204B2 (en) | Thread control system and method in a computer system | |
US5748959A (en) | Method of conducting asynchronous distributed collective operations | |
JPS5819958A (en) | Degenerating method for computer | |
US4638432A (en) | Apparatus for controlling the transfer of interrupt signals in data processors | |
US6209019B1 (en) | Data processing system, computer network, and data processing method | |
JPH07306795A (en) | Data base equivalent processor of duplex system computer | |
JPH05314085A (en) | System for waiting operation mutually among plural computers | |
JPS60238938A (en) | Operating method of computer system | |
JP2666732B2 (en) | Load control method for computer system | |
JP2579946B2 (en) | Congestion control method | |
JP2704137B2 (en) | Current / spare switching method | |
GB2030331A (en) | Real-time Data Processing System for Processing Time Period Commands | |
JPH07121490A (en) | Multiple processing system and program execution control method | |
JP2618126B2 (en) | System audit monitoring method and apparatus | |
JP2538090B2 (en) | Failure simulation method | |
JPH02139630A (en) | Online task exchange device | |
JP2571003B2 (en) | Data concentrator | |
JPH0784786A (en) | Program execution control method | |
JPS648378B2 (en) | ||
JP3429559B2 (en) | System version switching method and system version switching method | |
JPS62245456A (en) | Time monitor and control system | |
JPS62114038A (en) | Test diagnostic system | |
JPS63132360A (en) | Input and output processing system | |
JPS6095669A (en) | Common resources managing circuit | |
JPH03253951A (en) | Memory garbage processing system |