JPH0728733A - Network traffic control method - Google Patents
Network traffic control methodInfo
- Publication number
- JPH0728733A JPH0728733A JP5175134A JP17513493A JPH0728733A JP H0728733 A JPH0728733 A JP H0728733A JP 5175134 A JP5175134 A JP 5175134A JP 17513493 A JP17513493 A JP 17513493A JP H0728733 A JPH0728733 A JP H0728733A
- Authority
- JP
- Japan
- Prior art keywords
- data
- transmission
- destination
- intermediate device
- network
- 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
Landscapes
- Computer And Data Communications (AREA)
- Small-Scale Networks (AREA)
- Communication Control (AREA)
Abstract
(57)【要約】
【構成】ネットワークに接続された中間装置１と、ネッ
トワークに接続されたデータの送信元である複数のクラ
イアント２，ネットワークに接続されたデータの送信先
である複数のサーバ３からなり、中間装置１はクライア
ント２からサーバ３への送信データを一旦待ち行列に入
れ、サーバ３のＣＰＵ負荷が低いときに中間装置１の待
ち行列内からデータサイズの小さい順にサーバ３に送信
する。
【効果】中間装置により送信先機器のＣＰＵ負荷状態の
把握及び送信データのスケジューリングを行い、送信先
のＣＰＵ負荷が小さいときにデータサイズの小さい順に
データを送信先に送信することにより、ネットワークの
スループットを向上させる。
(57) [Summary] [Configuration] An intermediate device 1 connected to a network, a plurality of clients 2 that are sources of data connected to the network 2, and a plurality of servers 3 that are destinations of data connected to the network The intermediate device 1 temporarily queues the transmission data from the client 2 to the server 3, and when the CPU load of the server 3 is low, the intermediate device 1 transmits the data from the queue of the intermediate device 1 to the server 3 in ascending order of data size. . [Effect] The throughput of the network is improved by grasping the CPU load state of the destination device by the intermediate device, scheduling the transmission data, and transmitting the data to the destination in ascending order of the data size when the CPU load of the destination is small. Improve.
Description
【０００１】[0001]
【産業上の利用分野】本発明はネットワーク上で複数の
機器が通信する場合の、送信元の機器と送信先の機器と
の間のトラフィック制御に関する。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to traffic control between a source device and a destination device when a plurality of devices communicate on a network.
【０００２】[0002]
【従来の技術】機器間で通信を行う場合、通信時のトラ
フィックを制御する方法の一つとして、送信先の機器か
ら何らかの合図があるまで送信データを待ち行列に入れ
ておく方法がある。例えば特開昭61−189748号公報であ
る。2. Description of the Related Art When communication is performed between devices, one method of controlling traffic at the time of communication is to put transmission data in a queue until there is a signal from the device at the transmission destination. For example, it is JP-A-61-189748.
【０００３】この際、送信元から送信先にデータを送信
する場合には、送信先からＡＣＫ（acknowledge)が返っ
てきてから送信を行う。At this time, when data is transmitted from the transmission source to the transmission destination, transmission is performed after ACK (acknowledge) is returned from the transmission destination.
【０００４】通常、送信データは送信先のＣＰＵ負荷や
送信データのサイズに関わらず、送信先からＡＣＫが返
ってくると送信元が順次データを送信していた。Usually, the transmission data is transmitted sequentially by the transmission source when an ACK is returned from the transmission destination regardless of the CPU load of the transmission destination and the size of the transmission data.
【０００５】[0005]
【発明が解決しようとする課題】上述した従来の送信制
御方法では、送信先のＣＰＵ負荷が高い場合でも送信先
から送信元にＡＣＫが返ってくるとデータを送信してし
まう。そうすると送信先ではＣＰＵ負荷が高い状態で更
に送信されたデータの処理を行わなければならないた
め、ＣＰＵが過負荷となってしまい、処理速度が低下し
てしまうということが起こりうる。送信先となる機器に
は複数の機器からデータが送信されているため、送信先
機器の処理速度が低下するとこれら複数の機器から送信
されたデータを送信先で処理しきれず、ネットワーク上
のトラフィック量が増大してしまいデータ通信時のスル
ープットが低下してしまうという問題がある。In the above-described conventional transmission control method, data is transmitted when an ACK is returned from the transmission destination to the transmission source even if the CPU load of the transmission destination is high. Then, the destination must process the further transmitted data while the CPU load is high, which may cause the CPU to become overloaded and reduce the processing speed. Since data is sent from multiple devices to the destination device, if the processing speed of the destination device slows down, the data sent from these multiple devices cannot be processed by the destination device and the amount of traffic on the network. Is increased and the throughput during data communication is reduced.
【０００６】また、サイズの巨大なデータが待ち行列の
最初にあり、その後ろにサイズの小さいデータがある場
合にも、送信先の処理に時間がかかりスループットの低
下が起こりうる。Also, when huge size data is at the beginning of the queue and there is small size data behind it, it takes a long time to process the transmission destination and the throughput may decrease.
【０００７】[0007]
【課題を解決するための手段】このような問題点を解決
するために、本発明は送信元機器と送信先機器の間に中
間装置を設け、送信先機器のＣＰＵ負荷を送信先機器で
周期的に測定し、この送信先機器のＣＰＵ負荷を中間装
置に送信し、中間装置で受信した送信先機器のＣＰＵ負
荷によりデータを送信するか送信を保留するか決定し、
送信データのサイズに基づいた待ち行列内スケジューリ
ングを行い、送信先のＣＰＵ負荷が小さいときにデータ
サイズの小さい順に送信を行うようにしたものである。In order to solve such a problem, the present invention provides an intermediate device between a transmission source device and a transmission destination device, and a CPU load of the transmission destination device is cycled by the transmission destination device. Measurement is performed, the CPU load of the destination device is transmitted to the intermediate device, and the CPU load of the destination device received by the intermediate device determines whether to transmit the data or suspend the transmission,
In-queue scheduling is performed based on the size of transmission data, and transmission is performed in ascending order of data size when the CPU load of the transmission destination is small.
【０００８】[0008]
【作用】本発明では、中間装置によって送信先の機器の
ＣＰＵ負荷を把握し、送信先機器のＣＰＵ負荷が小さい
時にデータ送信を行うため、送信先における処理遅延を
軽減し、ネットワーク上のトラフィック量増大を防ぐた
め、データ通信時のスループットが向上する。According to the present invention, the intermediate device grasps the CPU load of the transmission destination device and performs data transmission when the CPU load of the transmission destination device is small. Therefore, the processing delay at the transmission destination is reduced and the traffic volume on the network is reduced. Since the increase is prevented, the throughput during data communication is improved.
【０００９】[0009]
【実施例】図１に本発明のトラフィック制御方法の一実
施例を説明するためのネットワークシステムのブロック
図を示す。図１において、１は送信データを受診し送信
先機器の負荷を把握し、送信データのスケジューリング
を行う中間装置、２はデータの送信元機器であるクライ
アント、３はデータの送信先機器であるサーバである。1 is a block diagram of a network system for explaining an embodiment of a traffic control method of the present invention. In FIG. 1, 1 is an intermediate device that receives transmission data, grasps the load of the transmission destination device, and schedules the transmission data, 2 is a client that is the transmission source device of the data, and 3 is a server that is the transmission destination device of the data. Is.
【００１０】また、図２の１１で示している不等式の右
辺にある許容値とは、サーバが処理遅延を起こさずに動
作できるＣＰＵ負荷の上限である。この値は、ネットワ
ーク環境及びサーバとして用いるハ−ドウェアによって
異なるものである。The allowable value on the right side of the inequality shown by 11 in FIG. 2 is the upper limit of the CPU load at which the server can operate without causing a processing delay. This value depends on the network environment and the hardware used as the server.
【００１１】動作の詳細について、図１，図２，図３に
より説明する。Details of the operation will be described with reference to FIGS. 1, 2 and 3.
【００１２】まず、中間装置２の動作について、図１，
図２により説明する。図１のクライアント２からサーバ
３に対して出された送信データは図１の（ａ）で示すよ
うに送信要求として図１の中間装置１に送られる。送信
データは図２の４で示すように図１の中間装置１で受信
される。送信データを受信すると、図１の中間装置１内
では図２の５で示すように送信データの送信元アドレス
及び送信先アドレスを読み込む。次に図２の６で図１の
中間装置１内の待ち行列用バッファ容量を調べ、バッフ
ァに空きが無い場合は、図２の７に示すように、送信元
に再送信を要求する。バッファに空きがある場合は、図
２の８で示すように、送信データを待ち行列に入れ、図
２の９で示すように、待ち行列内でデータサイズの小さ
い順にスケジューリングする。その後、図２の１０及び
図１の（ｂ）で示すように送信先である図１のサーバ３
から周期的に送信された送信先ＣＰＵの負荷を読み込
み、読み込んだＣＰＵの負荷が許容値以下かどうかを図
２の１１で調べる。ＣＰＵ負荷が許容値より大きい場合
は図２の１２に示すように送信保留とする。ＣＰＵ負荷
が許容値以下であれば、図２の１３及び図１の（ｃ）で
示すように待ち行列の先頭にあるデータを送信する。First, the operation of the intermediate device 2 will be described with reference to FIG.
This will be described with reference to FIG. The transmission data sent from the client 2 in FIG. 1 to the server 3 is sent to the intermediate device 1 in FIG. 1 as a transmission request as shown in FIG. The transmission data is received by the intermediate device 1 of FIG. 1 as indicated by 4 in FIG. When the transmission data is received, the source address and the destination address of the transmission data are read in the intermediate device 1 of FIG. 1 as indicated by 5 in FIG. Next, at 6 in FIG. 2, the queue buffer capacity in the intermediate device 1 in FIG. 1 is checked, and if there is no free space in the buffer, the sender is requested to retransmit as shown in 7 in FIG. When there is a free space in the buffer, the transmission data is put in a queue as shown by 8 in FIG. 2, and as shown by 9 in FIG. 2, scheduling is performed in ascending order of data size in the queue. After that, as shown in 10 of FIG. 2 and FIG. 1B, the server 3 of FIG.
The load of the destination CPU, which is periodically transmitted from, is read, and it is checked at 11 in FIG. 2 whether the read load of the CPU is less than or equal to the allowable value. When the CPU load is larger than the allowable value, the transmission is suspended as indicated by 12 in FIG. If the CPU load is less than or equal to the allowable value, the data at the head of the queue is transmitted as shown in 13 of FIG. 2 and (c) of FIG.
【００１３】次に、キュー内のスケジューリングについ
て、図２，図３により説明する。図２の８でデータを待
ち行列内に入れる際、図３の１４で示すように、待ち行
列内に他のデータがあるかどうかを調べる。他のデータ
が無ければ送信データを待ち行列に入れて、図２の１０
に戻る。他のデータがある場合、図３の１５で示すよう
に送信データを待ち行列の最後尾に加える。その後、図
３の１６で示すように、待ち行列内の送信データのデー
タのサイズと送信データ一つ前のデータのサイズを比較
し、図３の１７で示すように送信データのサイズが送信
データの一つ前のデータのサイズ以上であれば、図２の
１０に戻る。送信データのサイズの方が送信データの一
つ前のデータのサイズより小さければ、図３の１８で示
すように一つ前のデータと送信データの順序を入れ替え
る。入れ替えた後、図３の１９で示すように待ち行列内
で送信データの前にデータがあるかどうかを調べる。デ
ータが無ければ図２の１０に戻る。データがある場合、
図３の１６に戻り、また待ち行列内の一つ前のデータと
のサイズ比較を行う。以上の操作により、待ち行列内の
データはサイズの小さい順に並ぶ。Next, scheduling in the queue will be described with reference to FIGS. When putting data in the queue at 8 in FIG. 2, it is checked whether there is any other data in the queue, as shown at 14 in FIG. If there is no other data, the transmission data is put in the queue and
Return to. If there is other data, the transmission data is added to the end of the queue as indicated by 15 in FIG. After that, as shown at 16 in FIG. 3, the size of the transmission data in the queue is compared with the size of the data immediately before the transmission data, and as shown at 17 in FIG. If the size is equal to or larger than the size of the immediately preceding data, the process returns to 10 in FIG. If the size of the transmission data is smaller than the size of the data immediately before the transmission data, the order of the data immediately before and the transmission data is exchanged as shown at 18 in FIG. After swapping, it checks to see if there is data in the queue before the data to be transmitted, as shown at 19 in FIG. If there is no data, the process returns to 10 in FIG. If you have data,
Returning to 16 in FIG. 3, the size of the previous data in the queue is compared. By the above operation, the data in the queue are arranged in ascending order of size.
【００１４】[0014]
【発明の効果】本発明は、中間装置において送信先の負
荷を把握し、この送信先の負荷に応じてデータの送信及
び送信の保留の選択を行い、データを送信する場合には
データサイズの小さい順に送信するようにしたので、デ
ータ送信時のスループットを向上させる。According to the present invention, in the intermediate device, the load of the transmission destination is grasped, and the transmission of the data and the selection of the suspension of the transmission are selected according to the load of the transmission destination. Since the data is transmitted in ascending order, the throughput during data transmission is improved.
【図１】本発明に関わるネットワークトラフィック制御
方法の一実施例を説明するブロック図。FIG. 1 is a block diagram illustrating an embodiment of a network traffic control method according to the present invention.
【図２】本実施例における中間装置の動作を示すフロー
チャート。FIG. 2 is a flowchart showing the operation of the intermediate device in this embodiment.
【図３】本実施例における中間装置の動作のうち、待ち
行列内のスケジューリング方法を示すフローチャート。FIG. 3 is a flowchart showing a scheduling method in a queue among the operations of the intermediate device according to the present embodiment.
１…中間装置、２…クライアント、３…サーバ。 1 ... Intermediate device, 2 ... Client, 3 ... Server.
Claims (1)
装置間でのデータ通信の際のトラフィック制御方法にお
いて、送信先の負荷を考慮し、送信先の負荷に応じて即
時送信と送信保留とを区別することを特徴とするネット
ワークトラフィック制御方法。1. In a traffic control method for data communication between a plurality of processing devices connected to a network, a load of a transmission destination is considered, and immediate transmission and transmission suspension are performed according to the load of the transmission destination. A method for controlling network traffic, which is characterized by distinction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5175134A JPH0728733A (en) | 1993-07-15 | 1993-07-15 | Network traffic control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5175134A JPH0728733A (en) | 1993-07-15 | 1993-07-15 | Network traffic control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH0728733A true JPH0728733A (en) | 1995-01-31 |
Family
ID=15990887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP5175134A Pending JPH0728733A (en) | 1993-07-15 | 1993-07-15 | Network traffic control method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH0728733A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09269925A (en) * | 1996-04-02 | 1997-10-14 | Nri & Ncc Co Ltd | Large-scale client / server system for load control |
| JP2002278873A (en) * | 2001-03-19 | 2002-09-27 | Nec Corp | Contents providing system in data broadcasting |
| JP2019191881A (en) * | 2018-04-24 | 2019-10-31 | ヤフー株式会社 | Information processing apparatus, information processing system, information processing method and information processing program |
| WO2025183459A1 (en) * | 2024-02-28 | 2025-09-04 | 주식회사 에스티씨랩 | Server, method, and system for controlling number of access allowances based on influence level calculation using metric information |
-
1993
- 1993-07-15 JP JP5175134A patent/JPH0728733A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09269925A (en) * | 1996-04-02 | 1997-10-14 | Nri & Ncc Co Ltd | Large-scale client / server system for load control |
| JP2002278873A (en) * | 2001-03-19 | 2002-09-27 | Nec Corp | Contents providing system in data broadcasting |
| JP2019191881A (en) * | 2018-04-24 | 2019-10-31 | ヤフー株式会社 | Information processing apparatus, information processing system, information processing method and information processing program |
| WO2025183459A1 (en) * | 2024-02-28 | 2025-09-04 | 주식회사 에스티씨랩 | Server, method, and system for controlling number of access allowances based on influence level calculation using metric information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5014221A (en) | Mechanism for arbitrating client access to a networked print server | |
| US5604866A (en) | Flow control system having a counter in transmitter for decrementing and incrementing based upon transmitting and received message size respectively for indicating free space in receiver | |
| EP1868093B1 (en) | Method and system for a user space TCP offload engine (TOE) | |
| EP1545042B1 (en) | Retransmission system and method for a transport offload engine | |
| US10430374B2 (en) | Selective acknowledgement of RDMA packets | |
| US5418912A (en) | System and method for controlling buffer transmission of data packets by limiting buffered data packets in a communication session | |
| US20030208600A1 (en) | System and method for managing persistent connections in HTTP | |
| US8131881B2 (en) | Completion coalescing by TCP receiver | |
| US20070208820A1 (en) | Apparatus and method for out-of-order placement and in-order completion reporting of remote direct memory access operations | |
| US20090268747A1 (en) | Communication apparatus | |
| KR101143172B1 (en) | Efficient transfer of messages using reliable messaging protocols for web services | |
| US6052737A (en) | Computer system, program product and method for dynamically optimizing a communication protocol for supporting more users | |
| WO2006046972A1 (en) | Apparatus and method for supporting memory management in an offload of network protocol processing | |
| JP2008086027A (en) | Method and apparatus for processing remote requests | |
| JPH0728733A (en) | Network traffic control method | |
| US20120331107A1 (en) | Systems and methods for negotiated accelerated block option for trivial file transfer protocol (tftp) | |
| EP1139228A2 (en) | An intelligent bus interconnect unit | |
| JPH05103016A (en) | Message transmission reception method | |
| JP2606128B2 (en) | Communication method between computers | |
| JP2000078187A (en) | Data communication device, method and storage medium | |
| JPH03114334A (en) | Window flow control system | |
| JP3230339B2 (en) | Communication control device | |
| JPH1196108A (en) | Computer system and bus controller | |
| JP2611739B2 (en) | Task communication system | |
| JP2026005543A (en) | Communication device, control method thereof, and program |