JPS6365987B2 - - Google Patents
Info
- Publication number
- JPS6365987B2 JPS6365987B2 JP13650583A JP13650583A JPS6365987B2 JP S6365987 B2 JPS6365987 B2 JP S6365987B2 JP 13650583 A JP13650583 A JP 13650583A JP 13650583 A JP13650583 A JP 13650583A JP S6365987 B2 JPS6365987 B2 JP S6365987B2
- Authority
- JP
- Japan
- Prior art keywords
- counter
- packet
- processing
- position information
- relative position
- 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
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/76—Architectures of general purpose stored program computers
- G06F15/80—Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors
- G06F15/8007—Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors single instruction multiple data [SIMD] multiprocessors
- G06F15/8023—Two dimensional arrays, e.g. mesh, torus
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Multi Processors (AREA)
Description
【発明の詳細な説明】
(1) 発明の技術分野
本発明は最近傍間結合（処理エレメントをメツ
シユ状に配置する結合形態）における処理エレメ
ント（PE）間通信における送信側PEを中心とし
た送り先PEの相対位置を送り先アドレスとする
通信方式に関する。[Detailed Description of the Invention] (1) Technical Field of the Invention The present invention relates to communication between processing elements (PEs) in nearest-neighbor coupling (a coupling form in which processing elements are arranged in a mesh shape), in which a destination centering on a transmitting PE is used. This relates to a communication method in which the relative position of a PE is used as a destination address.
(2) 従来技術と問題点
従来、並列計算機（PE）における通信方式と
しては送信側PEの処理としては
送信したい“行先PEアドレス”を求める。(2) Conventional technology and problems Conventionally, in a communication method in a parallel computer (PE), the processing on the sending side PE is to find the "destination PE address" to be sent.
どの隣接PEに送るか、送る方向の決定。 Deciding which neighboring PE to send to and in what direction.
自PEアドレスと行先アドレスを比較して、
どの隣接PEに送るかを判断しその送る方向を
決める。 Compare the own PE address and destination address,
Determine which neighboring PE to send to and decide the sending direction.
このとき、全てのPEのアドレスの情報をも
とにして方向を決定する。 At this time, the direction is determined based on the address information of all PEs.
第１回に示すように“行先PEアドレス”を
ヘツダとしてもつパケツトを生成し、で決め
た方向に送る。 As shown in Part 1, a packet with the "destination PE address" as a header is generated and sent in the direction determined by.
受信側PEの処理としては
受信したパケツトより“行先PEアドレス”
を取り出す。 The processing of the receiving PE is to determine the “destination PE address” from the received packet.
Take out.
自PEアドレスと行先PEアドレスを比較。 Compare the own PE address and destination PE address.
等しい場合は自PE内で処理。等しくない場
合は
ａ○ どの隣接PEに送るか、送る方向の決定。
自PEアドレスと行先アドレスを比較して、
どの隣接PEに送るか―その送る方向を決め
る。このとき、全てのPEのアドレスの情報
をもとにして方向を決定する。 If they are equal, process within the own PE. If they are not equal, a○ Decide which neighboring PE to send to and the sending direction.
Compare the own PE address and destination address,
Decide which neighboring PE to send to - the direction of the send. At this time, the direction is determined based on the address information of all PEs.
ｂ○ 受けとつたパケツトをａ○で決めた方向に送
る。 b○ Sends the received packet in the direction determined by a○.
方式であつた、該従来方式では
送信側PEで行先PEアドレスをもとにして送
る方向の決定をしたにも拘ず、受信側PEで再
び同じ決定のための処理を繰り返すため、行先
PEにパケツトが到達する時間の増加を招いて
いる。 In this conventional method, even though the sending PE decides on the sending direction based on the destination PE address, the receiving PE repeats the process for making the same decision.
This results in an increase in the time it takes for packets to reach the PE.
さらに、自PEアドレスと行先PEアドレスの
比較によつて送る方向を決定する処理は、PE
アドレスが規則的でない場合、非常に複雑であ
るため、処理時間が増大する。 Furthermore, the process of determining the sending direction by comparing the own PE address and the destination PE address is
If the addresses are not regular, they are very complex and therefore increase processing time.
また、PEアドレスを変えた場合、送る方向
を決定する処理を変更しなければならず、柔軟
性に欠ける。 Furthermore, if the PE address is changed, the process for determining the sending direction must be changed, resulting in a lack of flexibility.
という欠点があつた。There was a drawback.
(3) 発明の目的
本発明は前記欠点に鑑みて、PEアドレス変更
に柔軟に対応でき、かつ迅速に行先PEに到達で
きる通信方式を提供することを目的とする。(3) Purpose of the Invention In view of the above drawbacks, it is an object of the present invention to provide a communication method that can flexibly respond to PE address changes and quickly reach the destination PE.
(4) 発明の構成
該目的は最近傍結合された並列計算機における
処理エレメント（以下PEと称す。）間通信におい
て送信元PEは、該送信元PEを中心とした送り先
PEの相対位置情報〔カウンタ（Ｘ），カウンタ
（Ｙ）〕をデータに付けて送信部より送信し、受信
先PEは相対位置情報解析部により該相対位置情
報を解析し、当該相対位置情報が０〔カウンタ
（Ｘ）＝０、カウンタ（Ｙ）＝０〕の場合は自PE内
で該データを処理し、他方相対位置情報が０でな
い場合は、送信方向決定部により、該カウンタ
（Ｘ）、カウンタ（Ｙ）の値の正・零・負により送
信方向を決めるとともに、相対位置情報増減部に
より、当該カウンタ値を増減してデータとともに
次のPEに送信部より送信することを特徴とする
並列計算機における通信方式により達成される。(4) Structure of the Invention The purpose is that in communication between processing elements (hereinafter referred to as PEs) in parallel computers connected to nearest neighbors, a source PE communicates with a destination centering on the source PE.
The relative position information of the PE [counter (X), counter (Y)] is attached to the data and transmitted from the transmitting unit, and the receiving PE analyzes the relative position information using the relative position information analysis unit, and the relative position information is 0 [counter (X) = 0, counter (Y) = 0], the data is processed within the own PE, and on the other hand, if the relative position information is not 0, the transmission direction determining unit processes the data in the counter (X). , the transmission direction is determined by the positive, zero, or negative value of the counter (Y), and the relative position information increase/decrease unit increases or decreases the counter value and transmits it together with the data to the next PE from the transmitter. This is achieved by a communication method in parallel computers.
(5) 発明の実施例
以下図面を参照しつゝ本発明を詳細に説明す
る。第１図は従来のパケツトの構成の詳細を示す
図である。(5) Embodiments of the invention The present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram showing details of a conventional packet structure.
図において、１１は行先PEアドレス、１２は
データである。 In the figure, 11 is a destination PE address, and 12 is data.
第２図は本発明の一実施例を示すパケツトの構
成の詳細を示す図である。 FIG. 2 is a diagram showing details of the structure of a packet showing an embodiment of the present invention.
図において、２１はカウンタ（Ｘ）、２２はカ
ウンタ（Ｙ）、２３はデータである。 In the figure, 21 is a counter (X), 22 is a counter (Y), and 23 is data.
第３図は本発明の一実施例を示すPEの結合形
態と相対位置関係を示す図である。 FIG. 3 is a diagram showing the bonding form and relative positional relationship of PEs showing one embodiment of the present invention.
図において、３１はＹ軸の正方向、３２はＸ軸
の正方向、３３はＹ軸の負方向、３４はＸ軸の負
方向である。ここでPE0を原点として３２，３４
はＸ軸を３１，３３はＹ軸を示しており、各PE
の位置を座標点として考える。 In the figure, 31 is the positive direction of the Y axis, 32 is the positive direction of the X axis, 33 is the negative direction of the Y axis, and 34 is the negative direction of the X axis. Here, with PE 0 as the origin, 32, 34
indicates the X axis, 31 and 33 indicate the Y axis, and each PE
Consider the position of as a coordinate point.
第４図は本発明の一実施例を示すカウンタ
（Ｘ）とカウンタ（Ｙ）をヘツダとしてもつパケ
ツトを受け取つたPEの処理を示す図である。 FIG. 4 is a diagram showing the processing of a PE that receives a packet having a counter (X) and a counter (Y) as headers, showing an embodiment of the present invention.
第５図は本発明の一実施例を示す処理エレメン
ト（PE）の機能ブロツク図である。 FIG. 5 is a functional block diagram of a processing element (PE) showing one embodiment of the present invention.
図において、５１は受信部、５２は相対位置情
報解析部、５３は送信方向決定部、５４は相対位
置情報増減部、５５は送信部、５６は処理部であ
る。 In the figure, 51 is a receiving section, 52 is a relative position information analysis section, 53 is a transmission direction determining section, 54 is a relative position information increasing/decreasing section, 55 is a transmitting section, and 56 is a processing section.
第３図のPE0から他のPEにデータを送りたい
場合を考える。この場合各PEは最近傍結合して
いるとする。このとき、PE0を中心にして他の
PEの相対位置を求め、それを相対位置情報とす
る。 Consider the case where we want to send data from PE 0 in Figure 3 to other PEs. In this case, each PE is assumed to be connected to its nearest neighbors. At this time, centering on PE 0 , other
Find the relative position of PE and use it as relative position information.
例えば、PE1，PE2のPE0を中心にした相対位
置情報は、それぞれPE1はＸ方向に＋１すなわち
カウンタ（Ｘ）＝＋１、Ｙ方向に＋１すなわちカ
ウンタ（Ｙ）＝＋１となり、PE2はＸ方向に−２
すなわちカウンタ（Ｘ）＝−２，Ｙ方向に０すな
わちカウンタ（Ｙ）＝０となると考え、Ｘ方向の
相対位置情報をカウンタ（Ｘ）２１で、Ｙ方向の
相対位置情報をカウンタ（Ｙ）２２で与える。 For example, the relative position information of PE 1 and PE 2 with PE 0 as the center is that PE 1 is +1 in the X direction, that is, counter (X) = +1, +1 in the Y direction, that is, counter (Y) = +1, and PE 2 is is -2 in the X direction
In other words, considering that counter (X) = -2 and 0 in the Y direction, that is, counter (Y) = 0, the relative position information in the X direction is calculated by the counter (X) 21, and the relative position information in the Y direction is calculated by the counter (Y) 22. Give with.
このカウンタ（Ｘ）２１とカウンタ（Ｙ）２２
を送るパケツトのヘツダとしてもち、行先PEの
アドレスの代りとする。 This counter (X) 21 and counter (Y) 22
as the header of the packet to be sent, in place of the address of the destination PE.
このヘツダをもつパケツトを受けとつたPEの
処理を第４図に示す。第４図において、“他の隣
接PEに送る処理”では、カウンタ（Ｘ）２１、
カウンタ（Ｙ）２２の正・零・負によつて、送る
方向を知ることができる。 Figure 4 shows the processing of a PE that receives a packet with this header. In FIG. 4, in "processing to send to other adjacent PEs", counter (X) 21,
The sending direction can be determined by the positive, zero, and negative values of the counter (Y) 22.
また、パケツトをある方向に送る場合、カウン
タ（Ｘ）２１とカウンタ（Ｙ）２２をその正・負
によつて増減すなわち１だけカウントダウンある
いはカウントアツプを行い、送り出すパケツトの
ヘツダとする。 When a packet is sent in a certain direction, the counter (X) 21 and counter (Y) 22 are increased or decreased depending on their positive or negative values, that is, they are counted down or counted up by 1, and are used as the header of the packet to be sent.
こうして、カウンタ（Ｘ）２１とカウンタ
（Ｙ）２２がともに零となつたパケツトを受けと
つたPEに、自PE内でそのパケツトを処理させる
ことができる。 In this way, a PE that receives a packet whose counter (X) 21 and counter (Y) 22 are both zero can be made to process that packet within its own PE.
次に本発明の処理動作について説明すると、送
信側PEの処理としては
送信したい行先PEアドレスの代りに、自PE
を中心にした行先PEの相対位置を求めて“カ
ウンタ（Ｘ）と“カウンタ（Ｙ）”に人がセツ
トする。 Next, to explain the processing operation of the present invention, in the processing of the sending PE, instead of the destination PE address to be sent,
A person determines the relative position of the destination PE with ``counter (X)'' and ``counter (Y)'' set at the center.
どの隣接PEに送るか、送信方向決定部５３
により送る方向の決定。 The transmission direction determining unit 53 determines which neighboring PE to send to.
Determine the direction of sending.
このとき、カウンタ（Ｘ）とカウンタ（Ｙ）
の正・零・負によつて、どの隣接PEに送るか
が即わかる。 At this time, counter (X) and counter (Y)
Depending on the positive, zero, or negative value of , you can immediately know which neighboring PE to send to.
送る方向を決めたら、カウンタ（Ｘ）、カウ
ンタ（Ｙ）をその正・負によつて、相対位置情
報増減部５４で１だけカウントダウンあるいは
カウントアツプする。 Once the sending direction is determined, the relative position information increase/decrease unit 54 counts down or counts up by 1 depending on whether the counter (X) or the counter (Y) is positive or negative.
第２図に示すように、“カウンタ（Ｘ）２１”
と“カウンタ（Ｙ）”をヘツダとしてもつパケ
ツトを生成し、で決めた方向に送る。 As shown in Figure 2, "Counter (X) 21"
A packet with ``counter (Y)'' as a header is generated and sent in the direction determined by .
他方受信側PEの処理としては
′ 受信したパケツトより“カウンタ（Ｘ）”と
“カウンタ（Ｙ）”を取り出す。 On the other hand, the processing on the receiving side PE is as follows: ``Counter (X)'' and ``Counter (Y)'' are extracted from the received packet.
′ カウンタ（Ｘ）とカウンタ（Ｙ）がともに
０に等しいか（カウンタ（Ｘ）＝０かつカウン
タ（Ｙ）＝０）を相対位置情報解析部５２で調
べ、ともに０に等しい場合は自PE内の処理部
５６で処理し、他方どちらかが０に等しくない
場合は、ａ○どの隣接PEに送るか、送る方向を
送信方向決定部５３で決定する。この時カウン
タ（Ｘ）とカウンタ（Ｙ）の正・零・負によつ
て、どの隣接PEに送るかが即わかる。送る方
向を決めたら、相対位置増減部５４によりカウ
ンタ（Ｘ）、カウンタ（Ｙ）をそれぞれ正・負
によつて１だけカウントダウンあるいはカウン
トアツプする。ｂ○受けとつたパケツトのヘツダ
を、カウントダウンあるいはカウントアツプさ
れたカウンタ（Ｘ）２１、カウンタ（Ｙ）２２
にしたパケツトをａ○で決めた方向に送信部５５
から送る。' The relative position information analysis unit 52 checks whether the counter (X) and counter (Y) are both equal to 0 (counter (X) = 0 and counter (Y) = 0), and if both are equal to 0, the If either of the two values is not equal to 0, the transmission direction determining unit 53 determines which neighboring PE to send the data to and the direction in which to send it. At this time, depending on whether the counter (X) and counter (Y) are positive, zero, or negative, it can be immediately determined which neighboring PE the data should be sent to. Once the sending direction is determined, the relative position increase/decrease unit 54 counts down or counts up the counter (X) and counter (Y) by 1 depending on whether they are positive or negative, respectively. b○ Counter (X) 21 and counter (Y) 22 count down or count up the header of the received packet.
The transmitter 55 sends the packet in the direction determined by a○.
Send from.
(6) 発明の効果
以上説明した様に本発明によれば、送信元PE
と受信先PEの相対位置情報により、送信制御す
るので、PEアドレス変更に柔軟に対処できると
ともに速やかに送信できるという効果がある。(6) Effects of the invention As explained above, according to the present invention, the source PE
Since the transmission is controlled based on the relative position information of the receiving PE and the destination PE, it is possible to flexibly deal with changes in the PE address and also enable prompt transmission.
第１図は従来のパケツトの構成の詳細を示す図
である。第２図は本発明の一実施例を示すパケツ
トの構成の詳細を示す図である。第３図は本発明
の一実施例を示すPEの結合形態と相対位置関係
を示す図である。第４図は本発明の一実施例を示
すカウンタ（Ｘ）とカウンタ（Ｙ）をヘツダとし
てもつパケツトを受け取つたPEの処理を示す図
である。第５図は本発明の一実施例を示す処理エ
レメント（PE）の機能ブロツク図である。
２１はカウンタ（Ｘ）、２２はカウンタ（Ｙ）、
２３はデータ、５１は受信部、５２は相対位置情
報解析部、５３は送信方向決定部、５４は相対位
置情報増減部、５５は送信部、５６は処理部。
FIG. 1 is a diagram showing details of a conventional packet structure. FIG. 2 is a diagram showing details of the structure of a packet showing an embodiment of the present invention. FIG. 3 is a diagram showing the bonding form and relative positional relationship of PEs showing one embodiment of the present invention. FIG. 4 is a diagram showing the processing of a PE that receives a packet having a counter (X) and a counter (Y) as headers, showing an embodiment of the present invention. FIG. 5 is a functional block diagram of a processing element (PE) showing one embodiment of the present invention. 21 is a counter (X), 22 is a counter (Y),
23 is data, 51 is a receiving section, 52 is a relative position information analysis section, 53 is a transmission direction determining section, 54 is a relative position information increasing/decreasing section, 55 is a transmitting section, and 56 is a processing section.
Claims (1)
メント間でパケツトと称する１単位情報を順次転
送処理を行なうデータ処理システムにおいて、 パケツトに少なくとも位置情報部を備え、 前段の処理エレメントから受信したパケツト中
の位置情報部の値が、予め定められた値である時
は当該処理エレメントで該パケツトの処理を行な
い、該位置情報部の値が、予め定められた値でな
い時は該パケツト中の位置情報部の値を所定値分
更新して後段の処理エレメントへ送出するよう処
理エレメントを構成し、 相対位置情報を用いて処理エレメント間の通信
を可能としたことを特徴とするデータ処理システ
ム。[Scope of Claims] 1. In a data processing system in which a plurality of processing elements are connected and one unit of information called a packet is sequentially transferred between the processing elements, the packet is provided with at least a position information part, and the data is transmitted from the preceding processing element. When the value of the location information part in the received packet is a predetermined value, the processing element processes the packet, and when the value of the location information part is not the predetermined value, the packet is processed. Data processing characterized in that the processing elements are configured to update the value of the position information section inside by a predetermined value and send it to the subsequent processing element, and communication between the processing elements is enabled using relative position information. system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58136505A JPS6028345A (en) | 1983-07-26 | 1983-07-26 | Communication system in parallel computer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP58136505A JPS6028345A (en) | 1983-07-26 | 1983-07-26 | Communication system in parallel computer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JPS6028345A JPS6028345A (en) | 1985-02-13 |
| JPS6365987B2 true JPS6365987B2 (en) | 1988-12-19 |
Family
ID=15176727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP58136505A Granted JPS6028345A (en) | 1983-07-26 | 1983-07-26 | Communication system in parallel computer |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPS6028345A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH042492U (en) * | 1990-04-19 | 1992-01-10 |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS6395559A (en) * | 1986-10-13 | 1988-04-26 | Mitsubishi Electric Corp | Processor for array computer |
| JP2543878B2 (en) * | 1987-03-31 | 1996-10-16 | 株式会社東芝 | Parallel processing system |
| JPS63293668A (en) * | 1987-05-27 | 1988-11-30 | Matsushita Electric Ind Co Ltd | Communication method for parallel computers |
| JP2657090B2 (en) * | 1989-03-14 | 1997-09-24 | 三洋電機株式会社 | Computer data communication system |
| JPH0338132A (en) * | 1989-07-04 | 1991-02-19 | Hitachi Ltd | Relay line selection system |
| JP3087900B2 (en) * | 1989-09-18 | 2000-09-11 | 富士通株式会社 | Computer nodes that constitute a parallel computer |
| GB2393285B (en) * | 2002-09-17 | 2007-01-03 | Micron Europe Ltd | Method for finding global extrema of a set of bytes distributed across an array of parallel processing elements |
| US7913062B2 (en) | 2003-04-23 | 2011-03-22 | Micron Technology, Inc. | Method of rotating data in a plurality of processing elements |
| US7676648B2 (en) | 2003-04-23 | 2010-03-09 | Micron Technology, Inc. | Method for manipulating data in a group of processing elements to perform a reflection of the data |
| US7596678B2 (en) | 2003-04-23 | 2009-09-29 | Micron Technology, Inc. | Method of shifting data along diagonals in a group of processing elements to transpose the data |
| US7581080B2 (en) | 2003-04-23 | 2009-08-25 | Micron Technology, Inc. | Method for manipulating data in a group of processing elements according to locally maintained counts |
| JP2005295236A (en) * | 2004-03-31 | 2005-10-20 | Serukurosu:Kk | Communication device |
| US10049322B2 (en) * | 2015-05-21 | 2018-08-14 | Google Llc | Prefetching weights for use in a neural network processor |
-
1983
- 1983-07-26 JP JP58136505A patent/JPS6028345A/en active Granted
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH042492U (en) * | 1990-04-19 | 1992-01-10 |
Also Published As
| Publication number | Publication date |
|---|---|
| JPS6028345A (en) | 1985-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JPS6365987B2 (en) | ||
| US5953507A (en) | Method and apparatus for providing a 3-way connection between a mobile computing device, a stationary computing device and a computer network | |
| US4887075A (en) | Local area network system with a multi-computer system coupled thereto and method for controlling the same | |
| EP1482657B1 (en) | Using directional antennas to enhance wireless mesh networks | |
| US6345301B1 (en) | Split data path distributed network protocol | |
| US7769839B2 (en) | Method and structure for autoconfiguration of overlay networks by automatic selection of a network designated router | |
| US7706275B2 (en) | Method and apparatus for routing data in an inter-nodal communications lattice of a massively parallel computer system by employing bandwidth shells at areas of overutilization | |
| EP0344722B1 (en) | Operation mode settable lan interconnecting apparatus | |
| EP0421779A2 (en) | Communication device | |
| JPH06506571A (en) | Methods and apparatus for managing and facilitating distributed heterogeneous network communications | |
| US6657973B1 (en) | Communications node, network system and method of controlling network system | |
| EP0378400B1 (en) | Two directional data communications system | |
| US4977499A (en) | Method and apparatus for commanding operations on a computer network | |
| US6718396B1 (en) | Network structure method and route determination equipment | |
| JPS61273633A (en) | Joint processing system for decentralized data base | |
| JPS6292642A (en) | Error response control system for transfer data | |
| JP2916185B2 (en) | Dynamic selection method of incoming communication adapter | |
| KR100832545B1 (en) | ＩＥＥＥ １３９４ Bridge implementation method and apparatus therefor | |
| JPH0821946B2 (en) | LAN bridge device | |
| JP2531126Y2 (en) | MAC Bridge | |
| JPH02219334A (en) | Exclusive control processing system | |
| JPH02224443A (en) | Communication system | |
| JPH0583267A (en) | Representative answer processing system for multi-cast communication | |
| JPH04213140A (en) | Control system for communication between host computers | |
| JPH0443458B2 (en) |