JPS6130832A - Information transmission system between processors - Google Patents

Information transmission system between processors

Info

Publication number
JPS6130832A
JPS6130832A JP15364684A JP15364684A JPS6130832A JP S6130832 A JPS6130832 A JP S6130832A JP 15364684 A JP15364684 A JP 15364684A JP 15364684 A JP15364684 A JP 15364684A JP S6130832 A JPS6130832 A JP S6130832A
Authority
JP
Japan
Prior art keywords
processor
information
processors
child
signal
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
JP15364684A
Other languages
Japanese (ja)
Inventor
Wataru Fujikawa
藤川 亘
Hiroyuki Kanno
管野 博行
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.)
NEC Corp
NEC Engineering Ltd
Original Assignee
NEC Corp
NEC Engineering 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 NEC Corp, NEC Engineering Ltd filed Critical NEC Corp
Priority to JP15364684A priority Critical patent/JPS6130832A/en
Publication of JPS6130832A publication Critical patent/JPS6130832A/en
Pending legal-status Critical Current

Links

Landscapes

  • Multi Processors (AREA)
  • Small-Scale Networks (AREA)

Abstract

PURPOSE:To decrease number of total signals by allowing communication between a master and a slave processor only connected directly in transmitting the information from one processor to all other processors. CONSTITUTION:A root processor P0 forms all information as to a processor P1, and its descendant processors P4, P5, P10-P12 into a signal and transmits the signal to the processor P1 through a transmission line l1. The processor P1 eliminates the information relating to the processor P1, forms the information relating to the processor P4 into one signal and transmits it to the processor P4 through a transmission line l4. Further, the processor P1 forms all the information relating to the processor P5 and its descendant processors P10-P12 into one signal and transmits it to the processor P5 through a transmission line l5. The processor P5 transmits the information to the processors P10-P12 through transmission lines l10-l12 with the similar processing as the processor P1.

Description

【発明の詳細な説明】 〔技術分野〕 本発明は、プロセッサ間の主従関係が親子関係で成り立
つツリー構造のマルチプロセッサシステムにおけるプロ
セッサ間情報伝達方式に関する。
DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an inter-processor information transmission system in a tree-structured multiprocessor system in which the master-slave relationship between processors is a parent-child relationship.

〔従来技術〕[Prior art]

第1図は上記のマルチプロセッサシステムの一例を示す
構成図で、根プロセッサPoと、これ(=伝送路1h 
、 J! 、 Isでそれぞれ接続されたプロセッサp
、 + PI * pmの間、プロセッサP1と、これ
に伝送路14’、1mでそれぞれ接続されたプロセラf
 P4 + Pgの間、プロセラ−9−P1と、これに
伝送路A!a 、 lγでそれぞれ接続されたプロセラ
fP・。
FIG. 1 is a configuration diagram showing an example of the above multiprocessor system, in which the root processor Po and this (=transmission line 1h
, J! , Is respectively connected processors p
, + PI * pm, the processor P1 and the processor f connected to it by transmission paths 14' and 1m, respectively.
Between P4 + Pg, Procera-9-P1 and transmission line A! a, procera fP· connected by lγ, respectively.

pyの間、プロセッサP3と、これに伝送路la rl
mでそれぞれ接続されたプロセッサPg 、 p−の間
、プロセッサP1と、これに伝送路lIo、111,1
11でそれぞれ接続されたプロセッサPro * Pt
t + Pttの間、そしてプロセッサP@と、これ(
=伝送踏石、。
During the py, the processor P3 and the transmission line la rl are connected to it.
Between the processors Pg and p-, which are connected at
11 respectively connected processors Pro*Pt
t + Ptt, and processor P@ and this (
= Transmission stepping stone.

lu r lIsでそれぞれ接続されたプロセラfP+
s+PH4、Puの間に親子関係が成立している。
Procera fP+ connected with lu r lIs respectively
A parent-child relationship is established between s+PH4 and Pu.

へ情報を伝達する場合について第2図を参照しながら説
明する。
The case of transmitting information to will be explained with reference to FIG.

先ず、根プロセッサP(1から子プロセッサPLへ情報
を伝達する場合、プロセラ?Ptに関する情報だけをプ
ロセラfPsへ伝送路11を通して伝達する。プロセラ
fPaあるいはPMへ情報を伝達する場合、それぞれプ
ロセラfP4 + Piに関する情報だけをプロセッサ
PIを経由してそれぞれ伝送路A’4 、 Alsを通
して伝達する。プロセラf Pro rpHあるいはP
INへ伝達する場合、それぞれプロセラ’t Pro 
r Fil * Puに関する情報だけをプロセッサP
1およびp、を経由してそれぞれ伝送路1Ior11s
 + Inを通して伝達する。この時、経由されるプロ
セッサでは情報の内容(二は関知しない。この場合、信
号数は、プロセラpP+と子プロセッサ間1間で子プロ
セッサPiとその子および孫プロセッサの数である。従
って、根プロセッサP0とそのすべての子、孫プロセッ
f P4 r Ps r Pro + Pn rP12
の間の伝送路に関する信号数について表−1の結果が得
られる。
First, when transmitting information from the root processor P (1 to the child processor PL), only information regarding the processor Pt is transmitted to the processor fPs through the transmission path 11. When transmitting information to the processor fPa or PM, the processor fP4 + Only the information regarding Pi is transmitted via the processor PI through the transmission lines A'4 and Als, respectively.Processor f Pro rpH or P
When transmitting to IN, each processor't Pro
r Fil * Only information about Pu is sent to processor P
1 and p, respectively, the transmission line 1Ior11s
+ Transmits through In. At this time, the transmitted processor is not concerned with the content of the information (2 is not concerned. In this case, the number of signals is the number of child processor Pi and its child and grandchild processors between processor pP+ and child processor 1. Therefore, the root processor P0 and all its children and grandchildren f P4 r Ps r Pro + Pn rP12
The results shown in Table 1 are obtained regarding the number of signals related to the transmission path between.

表−1 次に、プロセラfPsから他の全プロセッサへプロセラ
fPsの情報を伝達する場合の信号総数は表−2に示し
たプロセッサP+ (0(、I< 15 )へ伝達する
時の信号数の合針により51である。
Table 1 Next, the total number of signals when transmitting processor fPs information from processor fPs to all other processors is equal to the number of signals when transmitting to processor P+ (0 (, I < 15) shown in Table 2. It is 51 due to the matching needles.

表−2 次に、根プロセッサPGが全プロセッサPI+P1+・
・・+PI8から情報を収集する場合、各プロセラf 
P s rP2+・・・* pHへ要求信号を個別に送
信し、その応答信号により情報の収集を行なう。この場
合の総信号数は、51×2=102である。
Table 2 Next, the root processor PG
...+When collecting information from PI8, each processor f
A request signal is individually transmitted to P s rP2+...* pH, and information is collected based on the response signal. The total number of signals in this case is 51×2=102.

このように、従来のマルチプロセッサシステムにおいて
は、ひとつの主要プロセッサから他の全プロセッサへ情
報を伝達する場合、各プロセッサに直接、信号を送出し
ていたため、プロセッサ間の伝送路に多量の信号が流れ
るという欠点があつた。
In this way, in conventional multiprocessor systems, when transmitting information from one main processor to all other processors, the signal was sent directly to each processor, resulting in a large amount of signals being transmitted through the transmission path between the processors. It had the drawback of flowing.

〔発明の目的〕[Purpose of the invention]

本発明の目的は、総信号数を低減させたプロセッサ間情
報伝達方式を提供すること(=ある。
An object of the present invention is to provide an inter-processor information transmission system that reduces the total number of signals.

〔発明の構成〕[Structure of the invention]

本発明は、プロセッサ主従関係のツリー構造を利用し、
各プロセッサは親プロセッサおよび子プロセッサとだけ
情報の伝達を行ない、他のプロセッサとの情報伝達は親
プロセッサおよび子プロセッサ(1行なわせるようにし
たものである。
The present invention utilizes a tree structure of processor master-slave relationships,
Each processor communicates information only with its parent processor and child processors, and information is communicated with other processors only between the parent processor and child processors (in one line).

すなわち、本発明のプロセッサ間情報伝達方式は、各プ
ロセッサが、情報の伝達と収集および送信元を識別する
機能を有する情報識別部と、伝達信号の場合、自情報部
の処理を行ない送信元を除く親プロセッサとすべての子
プロセッサへ伝達し、収集信号の場合、すべての子プロ
セッサへ伝達しあるいはすべての子プロセッサからの情
報収集完了時、その情報を親プロセッサへ伝達する機能
を有する情報処理部とを有し、ひとつのプロセッサから
システム内全プロセッサへの情報の伝達あるいはすべの
子及び孫プロセッサの情報収集を行なう場合、各プロセ
ッサと直接通信することなしに、直接接続された親と子
プロセッサのみと通信を行なうことを特徴とする。
In other words, in the inter-processor information transmission system of the present invention, each processor has an information identification section that has the function of transmitting and collecting information and identifying the transmission source, and in the case of a transmission signal, processes its own information section and identifies the transmission source. an information processing unit that has a function of transmitting the information to the parent processor and all child processors excluding the parent processor, and in the case of a collection signal, transmitting the information to all child processors, or transmitting the information to the parent processor when information collection from all child processors is completed; When transmitting information from one processor to all processors in the system or collecting information from all child and grandchild processors, the directly connected parent and child processors do not need to communicate directly with each processor. It is characterized by communicating only with people.

〔実施例〕〔Example〕

以下、図面を参照して本発明の詳細な説明する。 Hereinafter, the present invention will be described in detail with reference to the drawings.

第6図は、本発明のプロセッサ間情報伝達方式を第1図
のマルチプロセッサシステムに適用したときの各プロセ
ッサPIの一実施例を示すブロック図である。
FIG. 6 is a block diagram showing an embodiment of each processor PI when the inter-processor information transmission method of the present invention is applied to the multiprocessor system of FIG. 1.

各プロセッサPIは、情報識別部et+と情報処理部e
2を有し、情報識別部elは、伝達・収集信号判定部e
11と、親および子プロセッサからの信号元を判定する
信号元判定部allを有し、情報処理部e2は、伝達信
号を処理する会違信号処理部e1mlと、収集信号と親
プロセッサからの要求信号を処理するイ饗件キエ妾球信
号処理部e22と、収集信号と子プロセッサからの信号
を処理する信号処理部elを有する。bh(k=1,2
+”・、10)は内部パスである。
Each processor PI includes an information identification section et+ and an information processing section e.
2, the information identification unit el is a transmission/collection signal determination unit e
11, and a signal source determination unit all that determines the signal sources from the parent and child processors, and the information processing unit e2 includes a signal source determination unit all that determines the source of signals from the parent and child processors, and an error signal processing unit e1ml that processes the transmission signal, and a signal source determination unit that determines the source of signals from the parent and child processors. It has a signal processing unit e22 that processes signals, and a signal processing unit el that processes collected signals and signals from child processors. bh(k=1,2
+”・, 10) is an internal path.

次に、第1図のマルチプロセッサシステムにおいて、根
プロセッサPaから他のプロセッサPI+P2.・・・
、Pi6へ情報を伝達する場合について第4図を参照し
ながら説明する。
Next, in the multiprocessor system of FIG. 1, from the root processor Pa to other processors PI+P2. ...
, Pi6 will be described with reference to FIG.

先ス、根プロセッサpoから子プロセラfPIへ情報を
伝達する場合、根プロセッサPoの信号処理部et+に
おいてプロセッサPIとその子および孫プロセッサP4
 + pm + plo−P11100情鞄なひとつの
信号に乗せて伝送路lIを通してプロセッサp、へ伝達
する。プロセッサP1では、伝達・収集信号判定部e1
1と信号処理部e!1において、伝達された情報の内容
を参照してプロセッサP、に関する情報を取り除く。次
いでプロセッサP+は、一方では子プロセッサP4に対
してプロセッサP4に関する情報をひとつの信号に乗せ
て伝送路14 を通してプロセッサP4へ伝達し、他方
では子プロセッサpmに対してプロセッサPIとその子
プロセラf Pro−P12100情報をひとつの信号
(=乗せて伝送路4を通してプロセッサP、へ伝達する
。プロセッサPδではプロセッサP1と同様の処理でプ
ロセッサPIG + pH、Patに対して情報をそれ
ぞれ伝送路1IO1l口、11!を通して伝達する。こ
の時、プロセッサP1と子プロセッサP」の間の信号数
はひとつである。
In the first step, when transmitting information from the root processor po to the child processor fPI, in the signal processing unit et+ of the root processor Po, the processor PI and its child and grandchild processors P4
+ pm + plo-P11100 is transmitted to the processor p through the transmission line II on one sensitive signal. In the processor P1, a transmission/collection signal determination unit e1
1 and signal processing section e! 1, information regarding the processor P is removed by referring to the contents of the transmitted information. Next, the processor P+ transmits information about the processor P4 to the child processor P4 on one signal through the transmission path 14, and on the other hand transmits the information regarding the processor P4 to the child processor pm through the processor PI and its child processor f Pro-. The P12100 information is transferred to the processor P through the transmission line 4 as a single signal.The processor Pδ uses the same processing as the processor P1 to send information to the processors PIG + pH and Pat, respectively, on the transmission lines 1IO1l and 11! At this time, the number of signals between the processor P1 and the child processor P is one.

上記以外の他のプロセッサに対しても上記と同様に情報
伝達が行なわれる。
Information is transmitted to other processors other than those mentioned above in the same manner as above.

プロセッサpmから他の全プロセッサへプロセラ+ps
の情報を伝達する場合の信号総数は、送信すべきプロセ
ッサ数(二等しく15であり、従来例(第2図)の場合
の信号数51の約1/3となる。
Processor pm to all other processors Procera+ps
The total number of signals when transmitting information is 15, which is equal to the number of processors to be transmitted (2), and is about 1/3 of the number of signals, 51, in the conventional example (FIG. 2).

なお、各プロセッサPIの制御は、信号処理部C!1(
二おいて次のとおり実現される。すなわち、親プロセッ
サからの信号受信時にはすべての子プロセッサへ同様の
信号を送信し、子プロセッサからのイ(号受信時には親
プロセッサと送信元プロセッサを除くすべての子プロセ
ッサへ同様の信号を送信する。
Note that each processor PI is controlled by the signal processing unit C! 1(
2, the following will be realized. That is, when receiving a signal from a parent processor, a similar signal is sent to all child processors, and when receiving a signal from a child processor, a similar signal is sent to all child processors except the parent processor and the sending processor.

次(=、根プロセッサPaが他の全プロセッサから情報
を収集する場合について、第5図を参照して説明する。
Next (=, the case where the root processor Pa collects information from all other processors will be described with reference to FIG. 5.

根プロセッサpoでは、子プロセッサPH、P2 r 
Paへ要求信号を送信し、プロセッサ子プロセッサP4
 、 Psへ要求信号を送信する。プセツサp、へ自情
報を伝達する。プロセッサpsでは子プロセッサP1a
 + Po + Patへそれぞれ要求信号を送信する
。各プロセッサPIG r pH+ patではプロセ
ッサP4と同様に親プロセッサPIへ自情報を伝達する
。さらに、プロセッサPIではすべての子プロセッサP
so + Po + PI3から情報を受信しプロセッ
サPro + Pat + Pa意の情報を親プロセッ
サP、へ伝達する。プロセッサPlではプロセッサP。
In the root processor po, the child processors PH, P2 r
Sends a request signal to Pa, and sends a request signal to child processor P4.
, sends a request signal to Ps. Transmits own information to Pusetsa p. In processor ps, child processor P1a
A request signal is sent to each of +Po and Pat. Each processor PIG r pH+ pat transmits its own information to the parent processor PI in the same way as the processor P4. Furthermore, in processor PI all child processors P
It receives information from so + Po + PI3 and transmits information about processor Pro + Pat + Pa to parent processor P. Processor P in processor Pl.

と同様に親プロセッサpoへ自情報とすべての子および
孫プロセッサP4 + Pg + Pro + Pi 
+ pHの情報を伝達する。上記以外の他のプロセラf
(二対しても上記と同様に処理される。この時の線信号
数は15X2=30であり、従来例(第2図)の場合の
信号数102の約%となる。
Similarly, the self information to the parent processor po and all child and grandchild processors P4 + Pg + Pro + Pi
+ Conveys pH information. Other processors other than the above
(The same processing as above is performed for 2 lines. The number of line signals at this time is 15×2=30, which is about % of the number of signals 102 in the conventional example (FIG. 2).

〔発明の効果〕〔Effect of the invention〕

本発明は、以上説明したようζ二、プロセツチ主従関係
のツリー構造を利用し、各プロセッサは親プロセッサお
よび子プロセッサとだけ情報の伝達を行ない、他のプロ
セッサとの情味の伝達は親プロセッサおよび子プロセッ
サ(=任せるようセしたのでプロセッサ間の信号の総数
が低減され、伝送路の有効利用を図ることができる。
As explained above, the present invention utilizes the tree structure of the processor master-slave relationship, in which each processor communicates information only with its parent processor and child processors, and the information is communicated with other processors only with the parent processor and child processors. Since the child processor is set to be left to the child processor, the total number of signals between processors is reduced, and the transmission path can be used effectively.

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

第1図はツリー構造のマルチプロセッサシステムの一例
を示す構成図、第2図は従来方式ζ二よる情報伝達時の
信号シーケンスと信号内容を示す説明図、第6図は本発
明のプロセッサ間情報伝達方式を適用したマルチプロセ
ッサシステムを構成する各プロセッサの一実施例のブロ
ック図、第4図は第6図における情報伝達時の信号シー
ケンスと信号内容を示す説明図、′s5図は第6図4−
おける情報収集時の信号シーダンスと信号内容を示す説
明図である。 Po・・・根プロセッチ PI + p、 +・・・+P11・・・プロセッサl
h + li +・・・+lti・・・伝送路eI・・
・情報識別部 8口・・・伝達・収集信号判定部 e+x・・・信号光判定部 e!・・・情報処理部 en・・・H信号処理部 effiト・・信号処理部 e+n・・・信号処理部 b+ 、 bi 、・・・、blO・・・内部バス四・
・・プロセッサP1に関する情報を持った信号 特許出願人  日本電気株式会社 ゛ 、 代理人  弁理士 内 原  (晋 イ第1図 第2図 l文集イ也号 第3図 第4図
FIG. 1 is a block diagram showing an example of a tree-structured multiprocessor system, FIG. 2 is an explanatory diagram showing the signal sequence and signal content during information transmission using the conventional method ζ2, and FIG. 6 is an illustration of inter-processor information according to the present invention. A block diagram of an embodiment of each processor constituting a multiprocessor system to which the transmission method is applied, FIG. 4 is an explanatory diagram showing the signal sequence and signal content during information transmission in FIG. 4-
FIG. 2 is an explanatory diagram showing signal seedance and signal content at the time of information collection. Po...root processor PI + p, +...+P11...processor l
h + li +...+lti...transmission line eI...
・Information identification section 8 ports...Transmission/collection signal judgment section e+x...Signal light judgment section e! ...Information processing section en...H signal processing section effit...Signal processing section e+n...Signal processing section b+, bi,..., blO...Internal bus 4.
...Signal patent applicant with information regarding processor P1 NEC Corporation, agent: patent attorney Susumu Uchihara (Figure 1, Figure 2, Collection of articles No. 3, Figure 4)

Claims (1)

【特許請求の範囲】[Claims] プロセッサ間の主従関係が親子関係で成り立つツリー構
造のマルチプロセッサシステムにおいて、各プロセッサ
は、情報の伝達と収集および送信元を識別する機能を有
する情報識別部と、伝達信号の場合、自情報部の処理を
行ない送信元を除く親プロセッサとすべての子プロセッ
サへ伝達し、収集信号の場合、すべての子プロセッサへ
伝達し、あるいはすべての子プロセッサからの情報収集
完了時、その情報を親プロセッサへ伝達する機能を有す
る情報処理部とを有し、ひとつのプロセッサからシステ
ム内の全プロセッサへの情報の伝達あるいはすべての子
および孫プロセッサの情報収集を行なう場合、各プロセ
ッサと直接通信することなしに、直接接続された親と子
プロセッサのみと通信を行なうことを特徴とするプロセ
ッサ間情報伝達方式。
In a tree-structured multiprocessor system in which the master-slave relationship between processors is a parent-child relationship, each processor has an information identification unit that has the functions of transmitting and collecting information and identifying the source, and in the case of a transmission signal, an information identification unit of its own information unit. Performs processing and transmits the information to the parent processor and all child processors except for the source, and in the case of a collection signal, transmits it to all child processors, or upon completion of information collection from all child processors, transmits the information to the parent processor. When transmitting information from one processor to all processors in the system or collecting information from all child and grandchild processors, without directly communicating with each processor, An inter-processor information transmission method characterized by communicating only with directly connected parent and child processors.
JP15364684A 1984-07-24 1984-07-24 Information transmission system between processors Pending JPS6130832A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP15364684A JPS6130832A (en) 1984-07-24 1984-07-24 Information transmission system between processors

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP15364684A JPS6130832A (en) 1984-07-24 1984-07-24 Information transmission system between processors

Publications (1)

Publication Number Publication Date
JPS6130832A true JPS6130832A (en) 1986-02-13

Family

ID=15567080

Family Applications (1)

Application Number Title Priority Date Filing Date
JP15364684A Pending JPS6130832A (en) 1984-07-24 1984-07-24 Information transmission system between processors

Country Status (1)

Country Link
JP (1) JPS6130832A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH022262A (en) * 1988-06-15 1990-01-08 Fuji Facom Corp Polling method for layer type data transmission system
JPH09215062A (en) * 1996-01-30 1997-08-15 Hitachi Building Syst Co Ltd Building remote monitoring device
JPH117595A (en) * 1997-06-18 1999-01-12 Yaskawa Electric Corp Sensor data collection device

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH022262A (en) * 1988-06-15 1990-01-08 Fuji Facom Corp Polling method for layer type data transmission system
JPH09215062A (en) * 1996-01-30 1997-08-15 Hitachi Building Syst Co Ltd Building remote monitoring device
JPH117595A (en) * 1997-06-18 1999-01-12 Yaskawa Electric Corp Sensor data collection device

Similar Documents

Publication Publication Date Title
KR940003698B1 (en) Method of program management in multi-computer system
EP0291907A3 (en) Interprocessor communication system in information processing system enabling communication between execution processor units during communication between other processor units
JPS61255461A (en) Multi cpu communicating device
JP2513029B2 (en) Task management method
JPH05128079A (en) Tracing system for multiprocessor system
JP2573617B2 (en) Online transaction processing method
JPS63244165A (en) Inter-computer file transfer method
JPH022262A (en) Polling method for layer type data transmission system
JPH01123351A (en) System for sharing remote job entry terminal
JPS63280332A (en) System for adjusting difference of function generation between programs
JPH0511824B2 (en)
JPH0239244A (en) Collation system for processing result of multi-computer system
JPH03288205A (en) Programmable controller system
JPH0333967A (en) Transfer control system for network constitutional information
JP2505792B2 (en) Micro computer system
CN114254921A (en) Transaction processing rule matching method and device
SU926662A1 (en) Adaptive data processing system
JPS6225006Y2 (en)
JPH0375808A (en) Software timer control system
JPS58132448U (en) Loop data transmission device
JPH0831871B2 (en) Network processing system
JPS6444475U (en)
JPS59151239A (en) Multiport console
JPH0497469A (en) Constitution check control circuit
JPS61221900A (en) Data transmission controller having simulation function