JP2006287598A - High-speed transmission control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a high-speed transmission system for performing optimum transfer control of a large amount of data by taking into consideration a characteristic of data to be a transmission target and a characteristic of a network to which the data is transmitted. <P>SOLUTION: A high-speed transmission controller 1 for transmitting data through the network 50 determines whether to compress and transmit the data of a transmission target or transmit the data without compressing the data on the basis of a file format of data stored in compression acceptance/rejection determination tables 31 and 32 and/or data capacity and a compression/decompression processing speed. In addition, the high-speed transmission controller 1 measures the state of a line for transmitting the data of a transmission target from a line error rate, an execution speed and bandwidth stored in network characteristic tables 33 and 34 in advance and determines a frame size in a transmission unit in accordance with the measured line state. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a high-speed transmission control method for transferring a large amount of data via a network.

In recent years, in the manufacturing industry such as the automobile, electrical / precision, and construction industries, the three-dimensional design / manufacturing data is increasing. Also, for each life cycle of the development business process, for example, in each phase such as 3D design, static / dynamic verification, manufacturability / assembling evaluation and environmental evaluation, 3D data is networked between departments and companies. To improve the efficiency and efficiency (development of time, improvement of quality, development of low cost). However, in order to exchange such large volumes of data between corporate bases (including overseas), and between suppliers and set manufacturers (companies that manufacture and sell finished products corresponding to suppliers), the existing network environment has a transfer rate. There are many difficulties in terms of (high speed, low speed) and line quality (especially in Southeast Asia: China, Thailand, Vietnam, etc.). On the other hand, in the game industry, development is carried out between global bases, and large-capacity sources and farm data are exchanged. High-speed transmission of large-capacity data via a network is desired.
Specifically, since data such as CAD drawings has a large capacity, it takes a long communication time when data is transmitted and received between companies via a network. In particular, when it is necessary to perform transmission while maintaining data quality, such as CAD drawings, data transmission that occurs at the time of communication in addition to slow line speed when sending to countries with poor line quality, etc. Since the error retransmission for recovering frequently occurs, the communication time is considerably long.

  The target factors of optimal control for high-speed transmission are data characteristics (file format, data capacity, sensitivity) and network characteristics (line type (Internet, extranet, WAN, LAN), transmission speed, packet loss rate, traffic, Protocol), but at present, optimal control is not necessarily performed in accordance with these factors. For example, even when 1 MB of data created by an application is transmitted, it is compressed and transmitted using the TCP / IP (FTP: File Transfer Protocol) protocol, even though it is data with poor compression efficiency, or 3 MB with a large capacity of 300 MB. Dimensional data is transmitted all at once, and there is a problem that error retransmission occurs when the line is cut off during transmission and the data is retransmitted from the beginning. Control is not taken into consideration.

  The invention described in Patent Document 1 can determine whether or not to compress based on the data capacity, and can suppress the load on the network when searching and transmitting the original electronic document registered in the document management server. If the file size of the original electronic document is larger than the specified value when the original electronic document is registered in the document management server 63, the original electronic document and the original electronic data after compression or compression division are registered. Documents are registered, and a search sheet 65 for searching the original electronic document is output, so that the original electronic document is searched from the reading of the search sheet 65, and further, the compressed original having a different file extension only. The electronic document file is searched and the above-mentioned pressure is sent from the document management server 63 to the copying apparatus 64 It is configured to transmit the document file data (see FIG. 6).

Patent Document 2 discloses a technique for determining a data frame size in accordance with a line error rate.
In Patent Document 3, it is determined whether or not the first narrow area / high speed wireless communication path is available, and if the narrow area / high speed wireless communication path is not available, the wide area / low speed wireless communication path is used. A technique for connecting to a LAN is disclosed.

JP 2004-362354 A JP 63-203062 A Special Table 2003-510897

  However, the transmission control method described in Patent Document 1 does not consider the time required for compression or the time required for decompression. In other words, if a large amount of data is used, it takes a long time for compression + long time for decompression. Therefore, there is a problem that the transmission efficiency is insufficient.

  Further, the transmission control systems described in Patent Documents 2 and 3 have a problem that transmission efficiency is insufficient because communication speed and bandwidth are not taken into consideration.

  Therefore, according to the present invention, in order to transmit large-capacity data at high speed over various transmission lines, optimal high-speed transmission control of data is performed in consideration of characteristics of data to be transmitted and characteristics of a network through which the data is transmitted. The purpose is to do.

The high-speed transmission control system according to the first embodiment of the present invention that achieves the above object is a high-speed transmission control system that transmits data via a network, and transmits or compresses data to be transmitted without being compressed. This is determined based on the file format of the data and / or the capacity of the data and the compression / decompression processing speed.
In the high-speed transmission control method according to the first aspect, the file format of the data is a data extension.
In the high-speed transmission control method according to the first aspect, the first transfer processing time when the data is compressed and transmitted based on the data capacity and the compression / decompression processing speed and the data is transmitted without being compressed. The second transfer processing time is calculated, and the first transfer processing time is compressed and transmitted when the first transfer processing time is shorter than the second transfer processing time.

The high-speed transmission control system according to the second embodiment of the present invention that achieves the above object is a high-speed transmission control system that transmits data via a network, and measures the state of a line that transmits data to be transmitted in advance. The size of the frame of the transmission unit is determined according to the state of the connected line.
In the high-speed transmission control method according to the second aspect, the state of the line is indicated by a line error rate, execution speed, and bandwidth.

  According to the present invention, data is transmitted in consideration of the characteristics of data to be transmitted and the characteristics of the network through which the data is transmitted. Therefore, it is possible to provide a high-speed transmission control method for optimally transmitting data.

According to the high-speed transmission control method according to the first embodiment, since it is determined whether or not the transmission data to be transmitted is compressed and transmitted by the data file format, for example, the extension of the data, it is not compressed. If the transmission time is shorter, it is possible to avoid unnecessary compression and transmission.
In addition, based on the data capacity and compression / decompression processing speed, the transfer processing time between the method of compressing and transmitting the data and the method of transmitting the data without compression is calculated, and the transfer processing time is short. Since transmission is performed, it is possible to avoid unnecessary compression and transmission.

  According to the high-speed transmission control method according to the second aspect, the frame size of the transmission unit is determined according to the measured line state such as the line error rate, execution speed and bandwidth, so that the transfer efficiency is optimal. It becomes possible to do.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 1 is a block diagram of a high-speed transmission control apparatus according to the present invention. The high-speed transmission control device 1 includes, for example, a workstation, a CPU 10, an input device 11 such as a keyboard and a mouse, an output device 12 such as a display unit and a printer, and a communication line such as the Internet between the device 1 and an external computer. Programs and data are read from a recording medium such as a communication device 13 that transmits and receives programs and data, a voice input / output device 14, a flexible disk (FD), and an optical disk CD (CD-ROM, CD-R and CD-RW). A recording medium R / W device 15 for writing on the recording medium, a main storage device 20 and an auxiliary storage device 30 are provided. The CPU 10, input device 11, output device 12, communication device 13, audio input / output device 14, recording medium R / W device 15, main memory 20, and hard disk HD 30 are connected to each other via a bus line 40.

  A main storage device (hereinafter referred to as a main memory) 20 comprises a battery-backed RAM, and is used as a temporary storage area for programs and data executed by the CPU 10, a work area for the CPU 10, and a storage area for fixed programs and data. . In the memory 20, the areas of the transfer data compression availability determination control unit 21 and the transfer frame division control unit 22 are used when transfer data compression availability determination control and transfer frame division control described later are executed. The auxiliary storage device (hereinafter referred to as a hard disk HD) 30 always stores programs and data written to the main memory as necessary, and is made of, for example, a magnetic disk. The hard disk HD 30 stores a first compression availability determination table 31, a second compression availability determination table 32, a first network characteristic table 33, and a second network characteristic table 34, which will be described later.

  A plurality of clients 50-1 to 50-n are connected to a high-speed transmission control device (hereinafter referred to as a server) 1 via the Internet 50, and the server 1 and any of the clients 50-1 to 50-n Data transmission is performed between the two. For optimum control of high-speed transmission, transfer data compression determination and transfer frame division control are performed in consideration of transfer data characteristics and network characteristics.

FIG. 2 is a flowchart of control for determining whether transfer data can be compressed according to the present invention. When data is transferred from the client 50-i to the client 50-j via the server 1, the server 1 executes the processing shown in the flowchart shown in FIG. The server 1 stores the tables shown in FIGS. 3 to 6 in the auxiliary storage device 30. FIG. 3 is a diagram showing a first compression capability determination table according to the present invention, FIG. 4 is a diagram showing a second compression capability determination table according to the present invention, and FIG. 5 is a diagram showing a first network characteristic table according to the present invention. FIG. 6 is a diagram showing a second network characteristic table of the present invention. The server 1 prepares in advance the table shown in FIG. 3 for determining whether or not compression is possible according to the file format (extension) of the transfer data and the compression rate.
The compression rate can be calculated by substituting the compressed data and the original data into a mathematical formula (compressed data / original data) × 100. This compression rate depends on the data content. For example, 1-byte alphanumeric characters such as kanji characters have a better compression rate and a smaller compression rate value as the number of repetitive data and more blanks.

  First, in step 201, transfer data preprocessing is executed. That is, the extension of the data to be transferred is read and the characteristics of the network are measured. The network characteristics include RTT measurement (execution speed measurement), bandwidth measurement, and line error rate measurement. Here, RTT (Round Trip Time) means response reception time, and bandwidth means the percentage of time that transmission target data can occupy a transmission line during transmission. The server 1 measures the bandwidth and the line error rate for each predetermined period in advance by an existing method based on the state of packets transmitted and received in the communication path used for transmission. In this step, the latest measured value is used. To do.

  In step 202, the transfer protocol corresponding to the execution speed (RTT) and the line error rate measured in step 201 is read with reference to the characteristic table of the first network shown in FIG.

In step 203, a compressibility determination process A is executed. That is, the extension table shown in FIG. 3 is referred to and compression is determined according to the extension. For example, in a data file format with an extension of ppt, compression processing has already been performed when the application generates data, and when such data is further compressed, it is compressed from the capacity of the original data by encoding a compression algorithm. The latter may have a larger capacity. In the first compressibility determination table shown in FIG. 3, data with a compression rate of 100% or more indicates such data. In this case, the original data is not compressed (the compression determination is “N ]). In FIG. 3 and subsequent figures, Y means that compression is possible, and N means that compression is not necessary.
In step 204, a compression possibility determination process B is executed. That is, when it is determined in step 203 that the data to be transmitted is compressible (the compression availability determination field is “Y”), the RTT, bandwidth, and line error rate measured in step 201 are shown in FIG. Based on the capacity of the original data to be transmitted, the data capacity after compression, the compression speed, and the decompression speed read with reference to the second compressibility determination table shown in FIG. Determine whether to compress data.

Transfer processing time (without compression processing) is Tt (sec)
The transfer processing time (compression processing is performed) is Ttcd (sec).
here,
Original data capacity A (Byte)
Capacity after compression Ac (Byte)
Compression processing speed Ct ((Byte / s)
Decompression processing speed Dt (Byte / s)
Compression processing time Tc (sec)
Decompression processing time Td (sec)
Execution line speed Te (bps / sec)
Bandwidth Bw (%)
Number of packets Pn (packet)
Line error rate Et (%)
Number of transfer packets N (times)
Packet capacity Vp (Byte / time)
Error retransmission capacity Ve (Byte)
Error retransmission capacity Ve = (N-N x Et) x Vp
Transfer processing time Tt = F (A, Bw, Te, Pn, Et, N, Vp, Ve)
Transfer processing time Ttcd is obtained as the following function.
Ttcd = F (A, Ac, Ct, Dt, Tc, Td, Te, Bw, Pn, Et, N, Vp, Ve)
Also, transfer processing time (compression processing is not performed) Tt is calculated from equation (1):
Tt = (A + Ve) x 8 / Te x Bw
The transfer processing time (compression processing) Ttcd can be calculated from equation (2).
Ttcd = A / Ct + Ac × 8 / Te × Bw + Ac / Dt + Ve × 8 / Te × Bw
When the condition of Ttcd ≦ Tt is satisfied, compression processing is performed.
When the condition of Ttcd ≧ Tt is satisfied, the compression process is not performed.

In step 205, when the condition of Ttcd ≦ Tt is satisfied, the compression processing of the compression target determination target file is executed.
In step 206, the transfer target file is transmitted by the transfer protocol referred to in the characteristic table of the first network in step 202.
The process shown in the flowchart of FIG. 2 allows the server 1 to determine whether the data to be transferred should be compressed or transmitted without being compressed, and transmit the data according to the determination. become.

FIG. 7 is a flowchart of transfer frame division control according to the present invention. When data is transferred from the client 50-i to the client 50-j via the server 1, the server 1 executes processing according to the flowchart shown in FIG.
First, in step 701, preprocessing of transfer data is executed. That is, the network characteristics are measured as in step 201 of FIG. The network characteristics include line error rate measurement, RTT measurement (execution speed measurement), and bandwidth measurement.

  In step 702, frame division capacity determination processing A is executed with reference to the characteristic table of the first network shown in FIG. That is, referring to the characteristic table of the first network, a transmission protocol corresponding to the RTT (Transmission Control Protocol: reception response time) measured in Step 701 and the line error rate is selected. In this embodiment, either TCP / IP (FTP) or TCP / (UDP + RPS) is selected as the transfer protocol. When the RTT value or the line error rate is large, UDP (: User Datagram Protocol) + RPS (: Random Packet Stream, high-speed error correction method) is selected.

In step 703, frame division capacity determination processing B is executed with reference to the characteristic table of the second network shown in FIG. That is, data corresponding to the RTT value, the line error rate, and the bandwidth measured in step 701 is searched from the characteristic table of the second network.
In step 704, frame division capacity determination processing C is executed. That is, the transfer frame size (frame division unit) corresponding to the RTT, bandwidth, and line error rate measured in step 701 searched from the characteristic table of the second network in step 703 is determined.

  FIG. 8 is a diagram showing a specific example of the frame division size, in which (a) shows an example in which the frame division size is 1 Mbyte, (b) shows an example in which 500 Kbyte is used, and (c). Is a diagram showing an example of 100 Kbytes, and (d) is a diagram showing an example of 10 Kbytes. In FIG. 8, N indicates the number of frame divisions (transmission file size units), and N, N-1, N-2, N-3, N-5, and Nm are the order in which the divided frames are transferred. And a transfer sequence counter number provided in the server 1. For example, if it is determined in step 704 that the transfer frame size is 1 Mbyte, as shown in FIG. 8A, framing is performed so that Fsaz is 1 Mbyte.

In step 705, frame transfer processing is executed. That is, compressed transfer or uncompressed transfer is performed for each transfer frame size. Here, the result of determining whether or not compression is possible by the processing shown in FIG.
In step 706, retransmission is performed when an error occurs on the receiving side. That is, when a notification of a frame in which an error has occurred is received from the receiving side, the transfer sequence counter transfers a frame N-1 that is one before the frame N in which a transfer error has occurred.
With the processing shown in the flowchart of FIG. 7, the server 1 can determine in what unit the data to be transferred should be divided into frames and transmit the data by a method according to the determination. become. Further, by combining the processing of FIG. 2 and the processing of FIG. 7, the compression availability and the frame size are determined according to the actual transmission efficiency such as the RTT, the line error rate, and the bandwidth, and the determination is made. It becomes possible to transmit the data by the method.

  FIG. 9 is a diagram showing a comparison of the transfer rate with respect to the reception response time in the TCP / IP communication system and the TCP / (UDP + RPS) communication system. FIG. 9A shows a line speed of 1 Mbps and a line error rate of 0%. (B) is a diagram showing an example in which the line speed is 1 Mbps and the line error rate is 1%, and (c) is a figure in which the line speed is 5 Mbps and the line error rate is 0.5%. (D) is an enlarged view of the relationship of the transfer rate to the reception response time with respect to the reception response time of 0 to 100 msec in (c). In transfer (FTP) using TCP / IP communication, there is no loss of transfer data and high reliability, but the longer the distance (the larger the RTT), the more useless for data transmission (to return the ACK received signal) Time) and the execution communication speed decreases. The longer the transfer distance, the longer it takes to transfer the data, making it impossible to use the data in a timely manner. On the other hand, in the transfer using TCP / (UDP + RPS) communication, the longer the distance (the larger the RTT), the unnecessary communication time (required for ACK) is unnecessary and the execution communication speed is improved.

  10 is a diagram showing verification results of effective transfer rates in the TCP / IP (FTP) communication method and the TCP / (UDP + RPS) communication method. 5-No. It is a figure which shows the value traced in the time slot | zone of 9. FIG. As can be seen from FIG. 10, the TCP / (UDP + RPS) communication method has a higher effective transfer rate than the TCP / IP (FTP) communication method.

  As described above, according to the present invention, it is possible to speed up the total transfer processing time by determining whether transfer data can be compressed and determining the division size of the transfer frame.

(Supplementary note 1) In a high-speed transmission control method for transmitting data via a network,
A method for causing a computer to execute a step of determining whether data to be transmitted is transmitted after being compressed or not based on a file format of the data and / or a capacity of the data and a compression / decompression processing speed. Program.
(Additional remark 2) The program of Additional remark 1 whose file format of the said data is an extension of data.
(Supplementary Note 3) Based on the capacity of the data and the compression / decompression processing speed, the first transfer processing time when the data is compressed and transmitted, and the second transfer processing time when the data is transmitted without compression, The program according to appendix 1 or 2, wherein the program is compressed and transmitted when the first transfer processing time is shorter than the second transfer processing time.

(Appendix 4) In a high-speed transmission control method for transmitting data via a network,
Measuring in advance the state of the line carrying the data to be transmitted;
Determining the frame size of the transmission unit according to the measured line condition;
A program that causes a computer to execute.
(Supplementary Note 5) The program according to Supplementary Note 4, wherein the state of the line is indicated by a line error rate, an execution speed, and a bandwidth.

It is a block block diagram of the high-speed transmission control apparatus by this invention. It is a flowchart of determination control of transfer data compression availability of the present invention. It is a figure which shows the 1st compression decision table of this invention. It is a figure which shows the 2nd compression decision table of this invention. It is a figure which shows the 1st network characteristic table of this invention. It is a figure which shows the 2nd network characteristic table of this invention. It is a flowchart of the division | segmentation control of the transfer frame of this invention. It is a figure which shows the example of a frame division | segmentation size, (a) is a figure which shows the example whose frame division size is 1 Mbyte, (b) is a figure which shows the example of 500 Kbyte, (c) is a figure which shows 100 Kbyte (D) is a figure which shows the example of 10 Kbytes. It is a figure which compares and shows the relationship of the transfer rate with respect to the reception response time in a TCP / IP communication system and a TCP / (UDP + RPS) communication system, (a) is an example with a line speed of 1 Mbps and a line error rate of 0%. (B) is a diagram showing an example in which the line speed is 1 Mbps and the line error rate is 1%, and (c) is an example in which the line speed is 5 Mbps and the line error rate is 0.5%. (D) is an enlarged view of the relationship of the transfer rate to the reception response time with respect to the reception response time of 0 to 100 msec in (c). It is a figure which shows the verification result of the effective transfer rate in a TCP / IP (FTP) communication system and a TCP / (UDP + RPS) communication system. No. in FIG. 5-No. It is a figure which shows the value traced in the time slot | zone of 9. FIG.

Explanation of symbols

21 Transfer data compression availability determination control unit 22 Transfer frame division control unit 31 First compression availability determination table 32 Second compression availability determination table 33 First network characteristic table 34 Second network characteristic table

Claims (5)

In a high-speed transmission control method for transmitting data over a network,
Determining whether data to be transmitted is transmitted after being compressed or not based on the file format of the data and / or the capacity of the data and the compression / decompression processing speed;
A high-speed transmission control system characterized by this.   The high-speed transmission control method according to claim 1, wherein a file format of the data is a data extension.   Based on the capacity of the data and the compression / decompression processing speed, a first transfer processing time when the data is compressed and transmitted and a second transfer processing time when the data is transmitted without compression are calculated, The high-speed transmission control method according to claim 1 or 2, wherein the first transfer processing time is compressed and transmitted when the second transfer processing time is shorter than the second transfer processing time. In a high-speed transmission control method for transmitting data over a network,
Measure the state of the line that transmits the transmission target data in advance,
Determine the frame size of the transmission unit according to the measured line status,
A high-speed transmission control method over a network characterized by this.   5. The high-speed transmission control system according to claim 4, wherein the state of the line is indicated by a line error rate, execution speed, and bandwidth.

Images