JPS62247653A - Constitution of packet exchange - Google Patents

Constitution of packet exchange

Info

Publication number
JPS62247653A
JPS62247653A JP61089929A JP8992986A JPS62247653A JP S62247653 A JPS62247653 A JP S62247653A JP 61089929 A JP61089929 A JP 61089929A JP 8992986 A JP8992986 A JP 8992986A JP S62247653 A JPS62247653 A JP S62247653A
Authority
JP
Japan
Prior art keywords
circuit
line
switch
packet
outgoing
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
JP61089929A
Other languages
Japanese (ja)
Inventor
Minoru Sugano
実 菅野
Kazuyuki Hayashi
和行 林
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.)
Hitachi Ltd
Original Assignee
Hitachi 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 Hitachi Ltd filed Critical Hitachi Ltd
Priority to JP61089929A priority Critical patent/JPS62247653A/en
Publication of JPS62247653A publication Critical patent/JPS62247653A/en
Pending legal-status Critical Current

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

PURPOSE:To improve the circuit efficiency by storing a header part for analysis after delaying a packet for a fixed period of time by a buffer, selecting an outgoing circuit out of the idle circuits for allocation and transmissing immediately the packet under reception to a terminating station. CONSTITUTION:The packet input received from an incoming circuit 1 undergoes the protocol processing 2 and is delayed by a buffer part 3 for a fixed short time. At the same time, a header analyzing part 4 stores the header part for analysis. A transmitting destination is detected and an outgoing circuit request signal 5 is transmitted to a conflict circuit 11. The circuit 11 chooses an idle circuit to transmit an outgoing circuit allocating signal 6 to the part 4 of a transmission requester. At the same time, the circuit 11 controls 14 the open/ close of a switch 15 by a switch control signal 12 to close a switch set within the switch 15. Thus the part 4 transmits immediately the packet under reception to a terminating station. As a result, the traffic concentration never occurs at a specific circuit. Then the delay time is shortened and the circuit availability is improved.

Description

【発明の詳細な説明】 〔産業上の利用分野〕 本発明は、パケット交換機に係り、特に高速回線を多数
収容する高速パケット交換機に好適な、交換機内にパケ
ットを蓄積せずに交換するパケット交換機の構成方式に
関する。
[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a packet switch, and is particularly suitable for a high-speed packet switch that accommodates a large number of high-speed lines, and which exchanges packets without storing them in the switch. Regarding the configuration method.

〔従来の技術〕[Conventional technology]

高速回線を多数収容する高速パケット交換機の構成方式
については、電子通信学会技術研究報告5E85−91
 r高速パケット交換方式」及び5E85−123 r
同報機能を持つ高速多元パケット交換システム構成力の
検討」において論じられているが、パケットの処理を回
線個別に行い、さらに回線毎にパケットを蓄積する構成
としている。
Regarding the configuration method of high-speed packet switching equipment that accommodates a large number of high-speed lines, please refer to IEICE technical research report 5E85-91.
r high-speed packet switching system” and 5E85-123 r
As discussed in ``Study of configuration capability of high-speed multi-packet switching system with broadcast function'', the configuration is such that packet processing is performed individually on each line, and furthermore, packets are stored on each line.

〔発明が解決しようとする問題点〕[Problem that the invention seeks to solve]

上記従来技術では、パケットのルーティング処理は、呼
毎の固定ルーティングとなり、回線へのパケットトラヒ
ックが偏り回線を効率的に使用できずさらに、パケット
を蓄積するための時間だけ交換機の遅延時間が増加する
。この結果、トラヒツトの増加に対しては、トラヒック
の偏りを考慮して、トラヒック増加分以上の回線を増加
せざるを得す、さらに交換機での遅延時間が大きいため
、高速の処理を要するコンピュータ間の通信や高速の通
信を必要とするローカルエリアネットワーク同士の接続
に関して制限を加えざるを得ない。
In the above conventional technology, the packet routing process is a fixed routing for each call, and the packet traffic to the line is uneven, making it impossible to use the line efficiently, and furthermore, the delay time of the switch increases by the time it takes to accumulate packets. . As a result, in response to an increase in traffic, it is necessary to take into account the imbalance in traffic and increase the number of lines by more than the amount of traffic increase.Furthermore, because the delay time at the exchange is large, it is necessary to increase the number of lines between computers that require high-speed processing. There is no choice but to impose restrictions on connections between local area networks that require high-speed communications and high-speed communications.

〔問題点を解決するための手段〕[Means for solving problems]

上記目的は、パケットのルーティング処理において、着
局対応の複数の回線から空きの1回線を予め定められた
優先順位に従って選ぶ機能を持った競合回路と1回線毎
に受信したパケットのヘッダ部を分析し、着局を定め、
前記競合回路へ要求を出すヘッダ分析部と、人出回線間
のスイッチを制御するスイッチ制御部と、前記処理に必
要な時間だけパケットを遅延させる遅延用バッファを用
いて、人出回線間のスイッチを制御し、パケットを交換
することにより、達成される。
The above purpose is to analyze the header part of the packet received for each line with a competing circuit that has the function of selecting one free line from multiple destination lines according to a predetermined priority order in packet routing processing. and determine the landing point,
A header analysis unit that issues a request to the competing circuit, a switch control unit that controls the switch between the traffic lines, and a delay buffer that delays the packet by the time necessary for the processing are used to switch between the traffic lines. This is achieved by controlling and exchanging packets.

〔作用〕[Effect]

上記した競合回路は、入回線のパケット毎に発生する出
回線割当て要求に対応して空回線を割当てるので、パケ
ット毎のダイナミックルーティングを可能とする。また
、上記したヘッダ分析部は。
The contention circuit described above allocates an idle line in response to an outgoing line allocation request generated for each incoming line packet, thus enabling dynamic routing for each packet. Also, the header analysis section mentioned above.

ヘッダを蓄積すると着局分析及び出回線要求を上記した
競合回路へ送出し、上記した競合回路で選択された空回
線と入回線間の接続を上記したスイッチ制御部で行い、
前記処理の間上記した遅延用バッファでパケットを蓄積
し、送出することにより短い遅延時間でパケットを交換
できる。
Once the headers are accumulated, the incoming station analysis and outgoing line request are sent to the above-mentioned competing circuit, and the above-mentioned switch control unit connects the idle line selected by the above-mentioned competing circuit to the incoming line,
By storing packets in the delay buffer described above during the processing and sending them out, packets can be exchanged with a short delay time.

〔実施例〕〔Example〕

以下、本発明の一実施例を第1図により説明する。第1
図において、1は入回線、2は入回線プロトコル処理部
、3は遅延用バッファ、4はヘッダ分析部、5は競合回
路対応の出回線要求信号線、6は出回線割付信号線、7
はプロトコル処理部と遅延用バッファ間の制御信号線、
8は遅延バッファ入力線、9はプロトコル処理部とヘッ
ダ分析部間の制御信号線、10はスイッチ入力線、11
は着局対応に設けられた出回線選択競合回路、12はス
イッチ制御信号線、13は出回線使用信号線、14はス
イッチ開閉制御部、15はスイッチ、16はスイッチ出
力線、17は出回線プロトコル処理部、18は出回線で
ある。
An embodiment of the present invention will be described below with reference to FIG. 1st
In the figure, 1 is an incoming line, 2 is an incoming line protocol processing unit, 3 is a delay buffer, 4 is a header analysis unit, 5 is an outgoing line request signal line for competing circuits, 6 is an outgoing line allocation signal line, and 7
is the control signal line between the protocol processing section and the delay buffer,
8 is a delay buffer input line, 9 is a control signal line between the protocol processing section and the header analysis section, 10 is a switch input line, 11
12 is a switch control signal line, 13 is an outgoing line use signal line, 14 is a switch opening/closing control unit, 15 is a switch, 16 is a switch output line, and 17 is an outgoing line. The protocol processing unit 18 is an outgoing line.

なお、遅延用バッファ3はファーストインファーストア
ウトのメモリ又は通常のランダムメモリを用いて構成で
き、ヘッダ処理部4や競合回路11やスイッチ開閉制御
部14は動作をパターン化し易いので、読み出し専用メ
モリを用いた布線論理又はマイクロプロセッサを用いた
蓄積プログラム制御等により構成できる。
Note that the delay buffer 3 can be configured using a first-in-first-out memory or a normal random memory, and since the operations of the header processing section 4, contention circuit 11, and switch opening/closing control section 14 can be easily patterned, read-only memory can be used. It can be configured by using wiring logic or storage program control using a microprocessor.

第1図において、入回線プロトコル処理部2はHDLC
手順に基づくフレーム の0削除や32ビット未満フレ
ームの検出やフレーム誤りの検出を行い、検出した異常
を遅延用バッファ3やヘッダ分析部4へ制御信線7又は
9により通知する。
In FIG. 1, the incoming line protocol processing unit 2 is an HDLC
It deletes zeros from frames, detects frames with less than 32 bits, and detects frame errors based on the procedure, and notifies the delay buffer 3 and header analysis unit 4 of the detected abnormalities through the control signal line 7 or 9.

ヘッダ分析部4は遅延バッファ入力線8よりパケットの
ヘッダ部を取り込み宛先を分析し着局を決め5着局対応
の出回線要求信号線5により競合回路11へ出回線割当
て要求を出す。競合回路11は各入回線1対応のヘッダ
分析部4からの出回線要求から1つを選び、スイッチ開
閉制御部14からの出回線使用信号線13により空回線
を捜し、選ばれた出回線要求に空回線を割当て、スィッ
チ開閉制御部14ヘスイツチ制御信号線12でスイッチ
15の対応接点閉接指示を出すと同時に選ばれた出回線
要求対応のヘッダ処理部4へ出回線割付信号線6により
応答を返す。競合回路11は前記処理を出回線割当て要
求が無くなるまで繰り返し続け、途中で空回線が無くな
った場合は空回線ができるまで処理を中断する。ヘッダ
分析部4は、競合回路11により出回線割付信号線6に
より応答を受は取ると。
The header analyzer 4 takes in the header part of the packet from the delay buffer input line 8, analyzes the destination, determines the destination station, and issues an outgoing line allocation request to the competition circuit 11 via the outgoing line request signal line 5 corresponding to the destination station. The competition circuit 11 selects one outgoing line request from the header analysis unit 4 corresponding to each incoming line 1, searches for an empty line using the outgoing line use signal line 13 from the switch opening/closing control unit 14, and selects the selected outgoing line request. Allocates a vacant line to the switch opening/closing control unit 14 and issues an instruction to close the corresponding contact of the switch 15 via the switch control signal line 12.At the same time, a response is sent to the header processing unit 4 corresponding to the selected outgoing line request via the outgoing line allocation signal line 6. return it. The competition circuit 11 continues to repeat the above process until there are no outgoing line allocation requests, and if there are no free lines during the process, the process is interrupted until a free line is available. The header analysis unit 4 receives a response from the competition circuit 11 via the output line allocation signal line 6.

遅延用バッファ3へ送出指示を出し、パケットが遅延用
バッファ3より送出完了するまで出回線要求信号線5に
より要求を出し続ける。遅延用バッファ3は、遅延バッ
ファ入力線8より受信したパケットをバッファに格納し
、ヘッダ分析部4より送信指示を受は取るまで遅延させ
た後スイッチ入力線10経由してスイッチ15ヘパケツ
トを送る。スイッチ15で交換されたパケットは、スイ
ッチ出力線16を経由して出回線プロトコル処理部17
へ送られる。出回線プロトコル処理部17は、HDLC
手順に基づくO挿入やFe2の付加等を行い、出回線1
8ヘパケツトを送信する。
A sending instruction is issued to the delay buffer 3, and the request continues to be issued via the outgoing line request signal line 5 until the packet is completely sent out from the delay buffer 3. The delay buffer 3 stores the packet received from the delay buffer input line 8, delays it until it receives a transmission instruction from the header analyzer 4, and then sends the packet to the switch 15 via the switch input line 10. The packets exchanged by the switch 15 are sent to the outgoing line protocol processing unit 17 via the switch output line 16.
sent to. The outgoing line protocol processing unit 17 is an HDLC
Perform O insertion and Fe2 addition based on the procedure, and connect the output line 1.
Send packet to 8.

以上説明した処理により入回線1より入ったバケラトは
、ヘッダ処理部4におけるヘッダ蓄積時間と分析時間及
び競合回路11の処理時間とスイッチ開閉制御部14の
処理時間の合計時間だけ遅延して出回線より次の局へ送
られる。なお、ヘッダ処理部4や競合回路11やスイッ
チ開閉制御部14は、前記したように布線論理化容易で
あり、現在の技術水準でも十分高速化可能であるため、
パケットの交換遅延時間をパケットを全て蓄積する場合
に比べて十分小さくできる。
Through the processing described above, the baquerato that entered from the incoming line 1 is delayed by the total time of the header accumulation time and analysis time in the header processing unit 4, the processing time of the competition circuit 11, and the processing time of the switch opening/closing control unit 14, and then is transmitted to the outgoing line. is sent to the next station. Note that the header processing unit 4, competition circuit 11, and switch opening/closing control unit 14 can be easily wired and logicalized as described above, and can be sufficiently increased in speed even with the current technological level.
The packet exchange delay time can be made sufficiently smaller than when all packets are stored.

また、第1図における入回線プロトコル処理部2、制御
信号線7や9及び出回線プロトコル処理部17を省いた
構成方式も可能であり、動作も第1図を用いて説明した
のと同様であるので、ここでは説明を省略する。さらに
、スイッチ15を開閉スイッチでなく、バス又はループ
等を用いた構成にする事も可能である。
It is also possible to configure a configuration in which the incoming line protocol processing unit 2, control signal lines 7 and 9, and outgoing line protocol processing unit 17 in FIG. 1 are omitted, and the operation is the same as that described using FIG. Therefore, the explanation will be omitted here. Further, the switch 15 may be configured using a bus, a loop, or the like instead of an open/close switch.

以上説明した様に本発明によれば、パケット毎に複数の
出回線より空回線を選択する事により、特定の回線にト
ラヒックが集中する事を防止でき、全回線を均等に使用
できるため、回線の使用効率を向1−できる。さらに、
パケットの遅延時間をヘッダの上積時間とルーティング
処理時間の和にできるため、パケット全てを蓄積してか
ら処理する場合に比入で遅延時間を短縮することができ
る。
As explained above, according to the present invention, by selecting an empty line from a plurality of outgoing lines for each packet, it is possible to prevent traffic from concentrating on a specific line, and all lines can be used equally. It is possible to improve the usage efficiency of moreover,
Since the packet delay time can be made the sum of the header accumulation time and the routing processing time, the delay time can be reduced by a proportional amount when all packets are accumulated and then processed.

〔発明の効果〕〔Effect of the invention〕

本発明によれば、パケット毎のダイナミックルーティン
グを実現できるので、特定回線にトラヒックが集中する
のを防止でき、回線の使用効率を向上する事ができ、さ
らに、パケットのヘッダを蓄積するだけで交換動作を開
始できるので、交換機の遅延時間減少に効果がある。
According to the present invention, it is possible to realize dynamic routing for each packet, so it is possible to prevent traffic from concentrating on a specific line, improve the efficiency of line usage, and furthermore, it is possible to perform exchange by simply accumulating packet headers. Since the operation can be started, it is effective in reducing the delay time of the exchange.

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

第1図は本発明によるパケット交換機の構成図である。 1・・・入回線 2・・・入目線プロトコル処理部 3・・・遅延用バッファ 4・・・ヘッダ分析部 5・・・競合回路対応の出回線要求信号6・・・出回線
割付信号 7・・・プロトコル処理部と遅延用バッファ間の制御信
号 8・・・遅延バッファ入力線 9・・・プロトコル処理部とヘッダ分析部間の制御信号 10・・・スイッチ入力線 11・・・出回線選択競合回路 12・・・スイッチ制御信号 13・・・出回線使用信号 14・・・スイッチ開閉制御部 15・・・スイッチ 16・・・スイッチ出力線 17・・・出回線プロトコル処理部 18・・・出回線 第1図
FIG. 1 is a block diagram of a packet switch according to the present invention. 1... Incoming line 2... Incoming line protocol processing unit 3... Delay buffer 4... Header analysis unit 5... Outgoing line request signal compatible with competing circuits 6... Outgoing line allocation signal 7 ... Control signal between the protocol processing section and the delay buffer 8 ... Delay buffer input line 9 ... Control signal between the protocol processing section and the header analysis section 10 ... Switch input line 11 ... Output line Selection competition circuit 12... Switch control signal 13... Outgoing line use signal 14... Switch opening/closing control unit 15... Switch 16... Switch output line 17... Outgoing line protocol processing unit 18...・Outgoing line diagram 1

Claims (1)

【特許請求の範囲】[Claims] パッケト交換機において、あらかじめ定められた時間だ
け遅延を付加できるバッファメモリとパケットのヘッダ
部を蓄積、分析し着局を定める機能を有するブロックと
前記ブロックらかの要求に従って着局対応に管理してい
る回線の空きを監視、抽出する機能を有するブロックと
前記ブロックで選択された空回線とこの空回線を割当て
られた要求元の回線の間のスイッチを閉じる機能を有す
るスイッチを有し、着局対応に管理する回線を複数とし
、パケット毎に空回線を割当て、ヘッダ部を蓄積すると
直ちに空回線を抽出し、受信中のパケットを着局へ向け
て送信開始することを特徴とするパケット交換機の構成
方式。
In a packet switch, there is a buffer memory that can add a delay by a predetermined amount of time, a block that has the function of storing and analyzing the packet header part, and determining the destination, and a block that manages the arrival according to the requests of the said block. It has a block that has the function of monitoring and extracting the free line, and a switch that has the function of closing the switch between the free line selected in the block and the line of the request source to which this free line is assigned, and supports incoming calls. A configuration of a packet switching device characterized in that it manages a plurality of lines, allocates an idle line for each packet, immediately extracts the idle line after storing the header part, and starts transmitting the packet being received to the destination station. method.
JP61089929A 1986-04-21 1986-04-21 Constitution of packet exchange Pending JPS62247653A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP61089929A JPS62247653A (en) 1986-04-21 1986-04-21 Constitution of packet exchange

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP61089929A JPS62247653A (en) 1986-04-21 1986-04-21 Constitution of packet exchange

Publications (1)

Publication Number Publication Date
JPS62247653A true JPS62247653A (en) 1987-10-28

Family

ID=13984387

Family Applications (1)

Application Number Title Priority Date Filing Date
JP61089929A Pending JPS62247653A (en) 1986-04-21 1986-04-21 Constitution of packet exchange

Country Status (1)

Country Link
JP (1) JPS62247653A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH024073A (en) * 1987-12-23 1990-01-09 Philips Gloeilampenfab:Nv Transmission system exchange network and exchange network controller
JPH02117241A (en) * 1988-10-27 1990-05-01 Mitsubishi Electric Corp Data switching device
JPH04145744A (en) * 1990-10-08 1992-05-19 Toshiba Corp Cell switch
JPH06188918A (en) * 1992-12-22 1994-07-08 Nec Corp Packet switchboard
JPH09102800A (en) * 1995-10-06 1997-04-15 Chokosoku Network Computer Gijutsu Kenkyusho:Kk Data exchange switch
US5881065A (en) * 1995-10-04 1999-03-09 Ultra-High Speed Network And Computer Technology Laboratories Data transfer switch for transferring data of an arbitrary length on the basis of transfer destination

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH024073A (en) * 1987-12-23 1990-01-09 Philips Gloeilampenfab:Nv Transmission system exchange network and exchange network controller
JPH02117241A (en) * 1988-10-27 1990-05-01 Mitsubishi Electric Corp Data switching device
JPH04145744A (en) * 1990-10-08 1992-05-19 Toshiba Corp Cell switch
JPH06188918A (en) * 1992-12-22 1994-07-08 Nec Corp Packet switchboard
US5881065A (en) * 1995-10-04 1999-03-09 Ultra-High Speed Network And Computer Technology Laboratories Data transfer switch for transferring data of an arbitrary length on the basis of transfer destination
JPH09102800A (en) * 1995-10-06 1997-04-15 Chokosoku Network Computer Gijutsu Kenkyusho:Kk Data exchange switch

Similar Documents

Publication Publication Date Title
EP0535860B1 (en) Congestion control for high speed packet networks
US5473604A (en) Method for avoiding node overload in a packet switching network
US4677616A (en) Flow control scheme for a switching network
JP3184817B2 (en) Flow control method, suspension control system and node
JP2540624B2 (en) Exchange device
EP1152574A2 (en) Packet switching system and method
JPH021671A (en) Load control system for packet switchboard
JPH08274793A (en) Delay minimization system provided with guaranteed bandwidthdelivery for real time traffic
WO2001086886A2 (en) Three-dimensional switch providing packet routing between multiple multimedia buses
JPH0657014B2 (en) Flow control in adaptive selection type packet switching system.
JPH03220837A (en) Resource allocation method for transmitting data in data communication network and wide-band isdn communication network and terminal adapter for connecting to data communication network
US7346065B2 (en) Bridge apparatus and bridge method
US6996119B2 (en) Adaptive polling method for router
JP3807614B2 (en) Packet order control method in multilink communication system
US5315587A (en) Data flow control
US20060018327A1 (en) Method for the transmission of data telegrams in a switched cyclical communication system
JPS62247653A (en) Constitution of packet exchange
US5642352A (en) LAN connecting device and LAN connecting system
US6954433B2 (en) IP processor
US6694373B1 (en) Method and apparatus for hitless switchover of a voice connection in a voice processing module
JP2510875B2 (en) Traffic monitoring system
KR100250485B1 (en) High speed packet switching device of frames between nodes
JP2715137B2 (en) Communication network control method
JPS6285533A (en) Decentralized packet exchange system
JPH06244867A (en) Backup route connection standby routing system