JP2001168871A - Data transfer system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To transfer a packet by utilizing transmission enable rate information from a network by a higher-order layer protocol, when a lower-order layer protocol has the information. SOLUTION: An ATM communication network is provided with an explicit congestion reporting function or an explicit transmission enable data amount reporting function. An ATM layer protocol function part 9 of the lower-order layer protocol function part 9 mounted on a transmission terminal holds transmission enable rate information or transmission enable data amount information. Then the transmission enable rate information is directly obtained by the higher- order protocol function part 5 of the higher-order layer protocol 4 mounted on the transmission terminal and the higher-order layer protocol 4 executes the shaping of a data transfer rate based on obtained transmission possible rate information. Thus, the data is transferred at high speed, while avoiding congestion in the ATM communication network and the ATM layer protocol function part 9 of a lower-order layer mounted on the transmission terminal.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to an explicit congestion notification function and an explicit transmission available rate notification function for a transmitting terminal,
Alternatively, the present invention relates to a data transfer method for transmitting data to be transferred at high speed and without loss on a network having a function of notifying a transmittable data amount.

[0002]

2. Description of the Related Art At present, a communication protocol stack generally used is based on a communication layer reference model OSI (Open Standard Edition).
It is based on a 7-layer model of system interconnection, and each layer can be designed independently. Therefore, there is an advantage that the protocol developer can develop without knowing the operation of the layer other than the layer to be developed. However, in this case, the protocol of the layer to be developed itself, for example,
It is necessary to determine the state of the lower layer indirectly by measuring the packet discard rate and the fluctuation of the feedback time, and to judge how to perform the transfer control and the like.

As a specific example, for example, TCP (Tran
A description will be given of a case in which a smission control protocol is used as the fourth layer of the OSI protocol. TCP
Then, when a packet sent by the transmitting terminal is received by the receiving terminal, a packet indicating reception confirmation (Ack = Acknow
ledge packet) to the sending terminal. The Ack packet displays the packet number that the receiving terminal expects to receive next (that is, the number of bytes of data correctly received by the receiving terminal up to the present time + 1). When the transmitting terminal confirms that the already transmitted packet has correctly arrived at the receiving terminal by receiving the Ack packet, the transmitting terminal can transmit the next data.

At this time, in order to increase the transfer efficiency, the transmitting side holds a data amount (window size) that can be transmitted at one time without waiting for reception confirmation, and by adjusting this window size, Control is performed to avoid congestion in the communication network. In the congestion control using this window size, while continuously receiving Ack packets, the number of packets flowing into the communication network is increased by gradually increasing the window size to the maximum value. To go. However, when congestion occurs in the communication network, packets are discarded in the communication network.

In this case, the following two situations can be considered. The first situation is a case where, after a transmitting terminal has transmitted a packet, an Ack packet corresponding to the packet cannot be obtained for a certain period of time, that is, a time-out. In the case of timeout, the transmitting terminal determines that the data has been lost and retransmits the data transmitted after the data whose reception was previously confirmed. Also,
This loss is determined to be due to network congestion, and the window size is reduced to reduce the amount of data that can be transmitted into the communication network, thereby avoiding congestion that is presumed to have occurred in the communication network.

The second situation is that the transmitting terminal receives Ack packets one after another, and the packet number that the receiving terminal expects to receive next is the same as the packet number received last time. This is the case when the packet numbers are the same. In this case, the packet indicated in the Ack was lost somewhere, but it is presumed that subsequent packets are being successfully transferred. Therefore, only the packet that is presumed to be lost is retransmitted. Also, a control method is performed in which the degree of congestion is determined to be small and the window size is not greatly reduced. In any of the above cases, the transmission side performs transfer control by indirectly estimating the state of the communication network from the reception state of the Ack packet.

[0007]

However, in the above prior art, since the transmitting terminal indirectly estimates the congestion state of the network, the amount of data that can be actually transferred by the communication network and the transmitting terminal are indirectly estimated. Is different from the estimated transmittable data amount. In particular, in the aforementioned prior art, the TCP
When detecting congestion by the first timeout,
As the communication network becomes a high-speed and large-capacity network, the tendency becomes remarkable.

For example, in the case of TCP, when a round-trip transfer delay time (RTT) and a band to be used (B) are given, in order to transfer data efficiently, the window size is generally set to “2”. X RTT x B "or more. However, when a round-trip transfer delay time is large and a wideband network is used, if the window size is set to this value, the amount of data that can be transferred by the transmitting terminal without waiting for reception of Ack is enormous. Amount. Therefore, once congestion occurs and data is retransmitted after waiting for timeout, T
The CP has to retransmit a huge amount of data. In addition, during the period of waiting for the timeout, the TCP transmitting terminal loses an opportunity to transfer data. As described above, with only the control method of indirectly estimating the state of the network based on packet loss, in a high-speed and wide-band network or a network with a fluctuating transmittable rate, the data transfer throughput performance due to packet discarding in the network is reduced. Due to significant deterioration,
This is a data transmission problem.

That is, ATM (Asyncronous Transfer)
Mode) layers and other protocols located in lower layers in communication have recently become more sophisticated, and are now equipped with congestion control functions and the like. However, the information obtained in the lower layer is terminated in the lower layer,
Cannot be used in upper layers. Thus, in the conventional data transfer technology, even if the network explicitly provides congestion information, the information is terminated in the lower layer protocol, and is not notified to the upper layer protocol. Even if the upper layer protocol has a rate shaping function, it does not dynamically and explicitly use information obtained from the lower layer protocol.

The present invention has been made in view of such circumstances, and a purpose of the present invention is to obtain information that is terminated in a lower layer and use it in an upper layer, so that it is more efficient than an existing method. An object of the present invention is to provide a data transfer method capable of transferring a packet.

[0011]

According to the data transfer method of the present invention, a network mainly has an explicit congestion notification function, an explicit transmittable data amount notification function or a transmittable data amount notification function, The upper-layer protocol directly obtains the transmittable rate under the condition that the lower-layer protocol implemented in the LTE holds transmittable rate information or transmittable data amount information, and the upper-layer protocol obtains the transmittable rate. By shaping the data transfer rate based on the obtained rate information, high-speed data transfer is possible while avoiding congestion in lower layer protocols implemented in the network and the transmitting terminal.

In other words, according to the data transfer method of the present invention, the upper layer protocol on the transmitting side refers to the transmittable rate information held by the lower layer protocol every time a certain condition is satisfied, and determines whether its own protocol is used. Know the rate of data that can be sent. Alternatively, the lower layer protocol interrupts the upper layer protocol each time a certain condition is satisfied, notifies the upper layer protocol of the transmittable rate information held in the lower layer protocol, and receives the notification of the upper layer protocol. Captures the transmittable rate information. If the upper layer protocol holds the data to be transmitted, the packetized data is shaped according to the rate information in order to adapt the output rate to the permissible transmission rate. And send it to the lower layer protocol.

That is, a specific means of the data transfer method according to the present invention is a data transfer method in a communication protocol constructed based on a seven-layer model of the OSI layer, and is used only between a transmitting terminal and a receiving terminal. Protocol processing is performed and an upper layer protocol having a function of performing transfer control, and protocol processing is performed between a transmitting terminal and a receiving terminal and in a network element in a communication network,
And a lower layer protocol having a function of notifying the network element or the receiving terminal of the transmittable rate information. The lower layer protocol implemented in the transmitting terminal transmits the transmittable rate information notified from the network element or the receiving terminal. The upper layer protocol implemented in the transmitting terminal holds the transmittable rate information retained in the lower layer protocol of the transmitting terminal, and the upper layer protocol transfers based on the acquired transmittable rate information. It is characterized in that data shaping is performed and a communication packet is transferred to a lower layer protocol.

That is, a communication protocol divided into layers like the OSI 7-layer model is used, and an upper layer protocol part, which is a protocol of a relatively higher layer, is basically provided only between a transmitting terminal and a receiving terminal. It has a function of performing protocol processing and performing transfer control. In addition, a lower layer protocol part, which is a protocol of a layer located at a lower level, can perform not only the protocol processing between the transmitting terminal and the receiving terminal but also a network element in a communication network in the protocol processing. The function realized by the lower layer protocol in the network element of the network or the receiving terminal has a capability of notifying the corresponding lower layer protocol function in the transmitting terminal of band information necessary for communication, that is, transmittable rate information. I have. The lower layer protocol function of the transmitting terminal holds the transmittable rate information notified from the network element or the receiving terminal, and has a capability of shaping and transmitting a communication packet according to the transmittable rate information. are doing.
Further, the upper layer protocol implemented in the transmitting terminal acquires the transmittable rate information held by the lower layer protocol function of the transmitting terminal, and uses the transmittable rate information to convert the communication packet into the lower layer. This is a data transfer method characterized by transferring data in accordance with a protocol.

According to the data transfer method of the present invention, in addition to the above-described means, the lower layer protocol implemented in the transmitting terminal includes a buffer memory having a desired capacity, and the upper layer protocol is adapted to the lower layer protocol. When the rate value of the data to be transmitted is different from the rate value of the data transmitted by the lower layer protocol into the network, the buffer temporarily stores the data to be transferred, so that the buffer between the upper layer protocol and the lower layer protocol can be used. , All data is allowed to be transferred.

A threshold value corresponding to the data amount is set in the buffer memory. When the data amount stored in the buffer memory passes the threshold value, the lower layer protocol is transmitted to the upper layer protocol. It is characterized in that a possible rate value is notified. Further, the threshold value is set as two different threshold values, and when the data amount in the buffer memory exceeds the high threshold value and when the data amount falls below the low threshold value, the upper layer protocol is set. May be notified of the transmittable rate value of the lower layer protocol. That is, by providing such a buffer, for example, even when the ACR value of the ATM layer is smaller than the ACR value acquired by the upper layer protocol, there is no possibility that data overflow occurs.

Further, in addition to the above-mentioned invention, the data transfer method of the present invention uses the transmittable rate information obtained by the upper layer protocol from the lower layer protocol.
The upper layer protocol dynamically performs shaping on the rate of communication packets transferred by the upper layer protocol to the lower layer protocol.

Further, the data transfer method of the present invention is characterized in that, in addition to the above-mentioned invention, the upper protocol stores the transmittable rate information obtained from the lower layer protocol. The transmittable rate information stored in the upper layer protocol is updated by the transmittable rate newly acquired by the upper layer protocol from the lower layer protocol. Further, the upper layer protocol may update the stored transmittable rate information for each communication packet, or may update the information for each of a plurality of packets. In this way, by updating all packets at once, the overhead of protocol processing can be further reduced.

Further, in the data transfer method of the present invention, in addition to the above-mentioned inventions, the upper layer protocol may be configured such that the stored transmittable rate information is interrupted by the lower layer protocol to the upper layer protocol. It is characterized by updating. By performing such interrupt processing, data transfer can be performed at higher speed.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of a data transfer system according to the present invention will be described in detail with reference to the drawings. That is, the communication network of the data transfer system according to the embodiment of the present invention provides communication by ATM between communication terminals, and here, as the lower layer protocol, one of the service categories of the ATM service A
The following describes a case where a BR (Available Bit Rate) service category is used as an example of a lower layer protocol having an explicit congestion notification function. The communication terminal according to the embodiment of the present invention means, for example, a terminal device such as a computer that can use the ATM protocol.

FIG. 1 is a configuration diagram showing an example of a basic configuration of an ATM communication network to which the present invention is applied. As shown in the figure, at least two opposing ATM terminals 1 and 2 which are communication terminals capable of operating the ATM protocol are connected to the ATM communication network 3. In this embodiment, A
Among ATM switches for providing BR, communication using an ABR network using an explicit rate display, that is, a switch using an ER marking which is a control signal for controlling a terminal device or the like is considered. In the following description, an embodiment of the present invention will be described with reference to FIG.
Of the communication terminals 1 and 2 using the TM communication network 3, especially ATM
The method will be described as an application method to a function of transmitting a cell.

Next, communication in the ABR service category will be described. ATM using ABR service category
In addition to the setting of various attributes at the time of connection setting which is normally performed to determine the characteristics of traffic, the RM (Re-
By using a special control cell called a “source management” cell, the communication rate that can be accepted by a network element such as a switch in an ATM network or a receiving terminal can be reduced by a transmission rate (ER: Explicit R).
ate) Notified to the transmitting terminal as a value. Note that a special control cell such as the RM cell is distinguished from other types of cells by the value of the PTI field in the header of the cell.

The transmitting terminal transmits an RM cell every time it transmits a fixed number of data cells. When the RM cell transmitted into the network arrives at the receiving terminal, it is looped back to the reverse connection and returned to the transmitting terminal. In the process of circulating the RM cell in the network, the network element sets control information such as an ER value in the RM cell. In an ATM terminal equipped with the ABR protocol, a dynamic rate control function relating to a transmission rate value of the ATM layer is realized. In this transmitting terminal, RM
ATM cells can be transmitted into the network at a rate equal to or lower than a rate according to a rate value (ACR: Allowed Cell Rate) calculated based on the ER value notified by the cell. For a detailed calculation method of the specified rate value, see “The ATM Forum, Traffic Management Specif.
ication 4.0, April 1996 ”.
In BR, as long as the transmitting terminal is transmitting a data cell according to the transmittable rate written in this RM cell,
There is assurance from the network that there is no congestion in the network and that losses during transmission in the network are very rare.

Next, a specific embodiment of the data transfer system of the present invention will be described. First, a first embodiment will be described. In this embodiment, a description will be given of realizing a method for controlling packet transmission from an upper layer protocol for ABR. for that purpose,
First, it is necessary to realize a function for the upper layer protocol to acquire the rate information (ACR value) held in the memory by the transmitting terminal that implements the above-described ABR protocol, if necessary. The ACR value is stored in a network interface card (NIC: Network Interface Card) having a function of controlling transmission of the ATM layer in the ATM terminal.
e Card), each VC (virtual Channel), or if the transmitting terminal is a VP (Virtual Path) terminal, V
It is held for each P. In the following description, VC is represented.

FIG. 2 is an explanatory diagram showing a functional model in a transmitting terminal to which the first embodiment of the present invention can be applied. That is, this figure is a functional block diagram for the upper layer to obtain the ACR value in the NIC. The upper layer protocol function unit 5 of the upper layer part obtains an ACR requesting to obtain and update the current ACR value of the existing ATM layer VC in response to a communication request from the upper layer protocol 4. A request function unit 6 and an ACR value storage function unit 7 for storing the return value are provided. That is, the ACR acquisition request function unit 6 has a function of requesting an update by acquiring the transmission available rate information held in the upper layer protocol 4 by the upper layer protocol 4 newly from the lower layer protocol.
Further, the ACR value storage function unit 7 has a function of retaining the transmittable rate information obtained from the lower layer protocol, that is, the ATM layer protocol function unit 9, in the upper layer protocol 4.

On the other hand, the ATM layer protocol function unit 9 in the ABR control unit 8 of the lower layer, together with the ACR holding function unit 10 for holding the ACR value,
An ACR acquisition request receiving function unit 11 that receives an acquisition request for a value and an ACR notification function unit 12 that notifies the upper layer protocol function unit 5 of the held ACR value.

The upper layer protocol 4 acquires the rate information of the ATM layer protocol function unit 9 by using the ACR value acquisition request function unit 6 when passing the packet to the lower layer, that is, the ATM layer protocol function unit 9. By adjusting the interval from the packet transmitted last time at the packet interval calculated based on the value, a shaping function in a packet unit is realized. At this time, in order to pass the packet to the ATM layer protocol function unit 9 at a value equal to or less than the value of the acquired rate information, the adjustment of the packet interval must be based on the following equation (1). Packet interval ≧ Max (packet size / (48 × ACR) −δ, 0) (1) Here, δ is a time required for the upper layer protocol 4 to process the packet. This processing time δ
Is different depending on the packet size, but this embodiment shows a case where a constant value is used as the value of δ for simplicity.

FIG. 3 shows the functional model shown in FIG.
FIG. 4 is an explanatory diagram showing a relationship between numerical values of packet shaping in an upper layer protocol in the transmitting terminal according to the first embodiment. That is, the value of the time δ required for the upper layer protocol 4 to process the packet is
If the value is changed in accordance with the processing time of the packet, the method of changing the value is to make the packet size correspond to a plurality of constant values according to the order of the packet size.

In the example of FIG. 3, the units of the transmittable rates A1 and A2 for which the ACR values are notified are [cell / sec], the unit of the packet interval is [time], and the unit of the packet size M is [byte]. ACR values A1, A2
Is multiplied by 48. When another unit system is used, a different multiplier is used. The control based on the above equation (1) is realized by a packet interval adjustment timer function that operates based on the system clock. In FIG. 3, the packet interval of the transmittable rate A1 [cell / sec] is M / (A1 × 48) [sec] because the packet size is M [byte].
It has become. In addition, transmittable rate A2 [cell / sec]
Since the packet size is M [bytes], the packet interval is M / (A2 × 48) [seconds].

FIG. 4 shows an example of the function model shown in FIG.
FIG. 6 is a state transition diagram of packet transmission interval control according to the first embodiment. In FIG. 4, n−1 in upper layer protocol 4
ATM (where n is a natural number of 2 or more)
It is assumed that, following the event passed to the layer protocol function unit 9, an n-th packet transfer request is sent to the upper layer protocol 4 (step S1). Then, the upper layer protocol 4 uses the NI for ABR holding the ACR information.
An ACR acquisition request relating to the VC to which the packet is to be transferred is made to the ACR acquisition request receiving function unit 11 of C (hereinafter, ABR NIC) (step S2). Then, the ABR NIC notifies the ACR value relating to the VC to the ACR acquisition request function unit 6 of the upper layer protocol 4 (Step S3).

Then, the upper layer protocol 4 calculates the packet transfer interval from the previous (n−1) th packet transmission to the current nth packet transmission by the interval determining function unit according to the above equation (1). After waiting for the time obtained by the calculation by the timer function in the system (step S4), the n-th packet is written into the NIC buffer of the ACR holding function unit 10 of the ATM layer protocol function unit 9 (step S5). Further, the ACR value written to the buffer is notified to the upper layer protocol function unit 5 (step S6). Also, when the (n + 1) th packet transfer request is issued to the upper layer protocol 4 (step S7), the same steps as described above are repeated.

FIG. 5 is a flowchart showing the control contents relating to the packet transmission method using the upper layer protocol according to the first embodiment. That is, when the upper layer protocol 4 acquires the transmittable rate (Ai) (step S11), a timer is started to count the processing time of the transmittable rate (Ai) acquired by the upper layer protocol 4 (step S12). . Then, the packet of the transmittable rate (Ai) is transmitted to the ATM layer (step S13), and the count time of the timer is calculated by dividing the packet size (M) by the transmittable rate ACR value (Ai), that is, When M / Ai [seconds] have elapsed (step S14), it is determined whether there is a next packet to be transmitted to the upper layer protocol 4 (step S1).
5) If there is, return to step S11 and repeat the same processing as described above. On the other hand, if there is no next packet to be transmitted in step S15, the apparatus enters a transmission packet waiting state (step S16).

That is, in the packet transmission method of the first embodiment, the upper layer protocol updates the transmittable rate information every time there is a packet transmission request. Therefore, once the ACR value is obtained, the ACR value in the ATM layer is changed to the AC value obtained by the upper layer protocol during the waiting time during which the packet interval is adjusted.
The value may be different from the R value. In particular, AT
If the ACR value of the M layer is smaller than the ACR value obtained by the upper layer protocol 4, data overflow may occur.

To avoid such a situation, the ATM
A buffer memory is used for the layer protocol function unit 9.
That is, an appropriate amount of buffer memory is provided in a device that realizes a lower layer function on the transmitting terminal side, and a value of a rate at which the upper layer protocol 4 sends data to the lower layer and a value of the lower layer protocol are transmitted to the network. When the transmission rate values are different, the rate values are stored in the buffer memory. That is, in order to realize this, a capacity for storing data of the maximum packet size transmitted by the upper layer protocol is prepared in the memory device so that the ATM layer can hold the capacity.

Next, a description will be given of a second embodiment of the data transfer system according to the present invention. In the first embodiment described above, each time the upper layer protocol receives a packet transmission request, it issues an ACR value acquisition request. But,
When the upper layer protocol is transferring data at high speed, if an ACR value acquisition request is made every time a packet transmission request is made to the ATM layer, the overhead in protocol processing cannot be ignored. Therefore, in the second embodiment, the upper layer protocol updates the transmittable rate information for each of a plurality of packets. As described above, by making an ACR acquisition request for each of a plurality of packet transmission requests, it is possible to reduce protocol processing overhead.

FIG. 6 is an example of a state transition diagram in the case where packet transmission interval control is performed for each of a plurality of packets in the second embodiment. In FIG. 6, the upper layer protocol issues an ACR acquisition request once (step S2).
1) After acquiring the ACR value (step S22), A
Without acquiring the CR value, each packet transfer interval based on the above-described equation (1) (step S
23) The packet is transmitted up to m times. Then, in the m-th packet, again, the upper layer protocol is A
A CR acquisition request is made (step S24), and an ACR value is acquired (step S25). Further, the ACR value information stored in the upper layer protocol is updated, and the AC
Packet transmission is performed at a new packet transfer interval calculated based on the R value. Here, m is a number determined by the upper layer protocol.

The above-described second embodiment can be modified as shown in FIG. FIG. 7 is another example of a state transition diagram in the case where packet transmission interval control is performed for each of a plurality of packets in the second embodiment. That is, this case shows a state transition in a case where packets to be transmitted in the upper layer protocol have already been accumulated. Therefore, the upper layer protocol issues m ACR acquisition requests (step S31), acquires the ACR value (step S32), passes the m packet transfer intervals (step S33), and returns to the ACR acquisition request again. Is performed (step S34), and an ACR value is obtained (step S34).
35).

FIG. 8 is a state transition diagram shown in FIG.
9 is a flowchart illustrating control contents relating to a packet transmission method in an upper layer protocol. That is, upon acquiring the transmittable rate (Ai) (step S41), the upper layer protocol calculates the number m of packets that can be transferred (step S42). Here, m is a value obtained by dividing the lower layer buffer amount by the packet size. Furthermore, the transmittable rate (Ai) obtained by the upper layer protocol
A timer is started to count the processing time (step S43). Then, the m packets are transmitted to the ATM layer (step S44), and the count time of the timer is set so that the packet size (M) is equal to the m transmittable rates A
When the value divided by the CR value (A), that is, M / A [seconds] has elapsed (step S45), it is determined whether there is a next packet to be transmitted to the upper layer protocol (step S46). In this case, the process returns to step S41, and the same processing as described above is repeated. On the other hand, if there is no next packet to be transmitted in step S46, the apparatus enters a transmission packet waiting state (step S47).

That is, in the case of FIG. 8, once the upper layer protocol has obtained the ACR value, it immediately transmits the packet m times without obtaining the ACR value, and after transmitting the m packets, Wait for the interval based on equation (2),
A request to acquire the next ACR value and subsequent packet transmission are performed. Packet interval = Max (m × packet size / (48 × ACR) −γ, 0) (2) Here, γ is a value determined by a processing delay of an upper layer protocol required to transmit m packets. . M is a value obtained by dividing the buffer amount obtained in the ATM layer by the packet size.

FIG. 9 is a flowchart showing the second operation shown in the flowchart of FIG.
FIG. 10 is an explanatory diagram showing a relationship between numerical values of packet shaping in an upper layer protocol in a transmitting terminal in the embodiment of FIG. In the figure, the transmittable rate A1 [cell
/ S] is m for packets of packet size M [bytes].
Are sent out. The packet interval at this time is m × M
/ (A1 × 48) [seconds]. The same applies to the transmittable rate A2 [cell / sec]. That is, in the transfer method shown in FIG. 7, compared to the transfer method shown in FIG.
Since many packets are transferred from the upper layer protocol at a time, the NIC of the ATM layer requires a larger buffer, while the accuracy of the timer to be maintained in the upper layer protocol is smaller than that of the transfer method of FIG. Can be set loosely.

Next, a third embodiment of the present invention will be described. That is, in the first and second embodiments described above, the method is to acquire the ACR value held by the NIC from the upper layer protocol. However, when performing higher-speed control, A
Instead of obtaining the CR value, the A
It is more advantageous to update the CR value. That is, this is a method in which the lower layer protocol interrupts the upper layer protocol to update the transmittable rate information held in the upper layer protocol.

In this case, an appropriate amount of buffer memory is provided in a device for realizing the lower layer function on the transmitting terminal side, and the value of the rate at which the upper layer protocol sends data to the lower layer is determined. Even if the protocol sends different rate values into the network,
Enables storing rate values. One threshold value is set in the buffer memory, and when the amount of data accumulated in the buffer memory exceeds the threshold value, the upper layer protocol is notified of the transmittable rate value of the lower layer protocol. Generates an interrupt signal, and sets the value held in the upper layer protocol to AC.
Update is performed by the R notification interrupt function. In addition, two thresholds having different values are provided in place of one threshold, and when the amount of data in the buffer memory exceeds the high threshold and falls below the low threshold, respectively, the upper layer protocol is used. Alternatively, a transmittable rate value in a lower layer protocol may be notified.

FIG. 10 is an explanatory diagram showing a functional model in a transmitting terminal to which the third embodiment of the present invention can be applied. In the figure, an upper layer protocol function unit 22 of an upper layer protocol 21 includes an ACR value storage function unit 2.
The lower layer ATM layer protocol function unit 24 includes an interrupt generation function unit 25 and an ACR holding function unit 2.
6 and an ACR notification interrupt function unit 27. In the state transition diagram in the case of this figure, the ACR acquisition request and the ACR value notification are read as the ACR value read and the ACR value read in FIGS. 4 to 6 and 7 in the first embodiment and the second embodiment. In other words, the other configuration is the same, and a duplicate description will be omitted.

The above-described embodiment is an example for describing the present invention, and the present invention is not limited to the above-described embodiment, and the technical scope of the present invention is limited to the scope of the invention. The range can be variously modified and implemented.

[0045]

As described above, according to the data transfer method of the present invention, it is possible to perform data transfer control in which the upper layer protocol directly reflects the congestion status of the network, and during the transfer after transmission from the upper layer protocol. Packet loss can be minimized. Also, since the overhead in the control of the present invention is only the acquisition of information from the network interface card and packet shaping, the processing load on the terminal due to the protocol processing for congestion control performed in the upper layer is also reduced. Therefore, application to higher-speed communication can be realized as compared with the conventional data transfer method.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing an example of a basic configuration of an ATM communication network to which the present invention is applied.

FIG. 2 is an explanatory diagram showing a functional model in a transmitting terminal to which the first embodiment of the present invention can be applied.

FIG. 3 is an explanatory diagram showing a relationship between numerical values of packet shaping in an upper layer protocol in the transmitting terminal according to the first embodiment in the functional model of FIG. 2;

FIG. 4 is a state transition diagram of packet transmission interval control according to the first embodiment in the functional model shown in FIG. 2;

FIG. 5 is a flowchart showing control contents relating to a packet transmission method in an upper layer protocol according to the first embodiment.

FIG. 6 is an example of a state transition diagram in the case where packet transmission interval control is performed for each of a plurality of packets in the second embodiment.

FIG. 7 is another example of a state transition diagram in the case where packet transmission interval control is performed for each of a plurality of packets in the second embodiment.

FIG. 8 is a flowchart showing control contents relating to a packet transmission method in an upper layer protocol in the state transition diagram shown in FIG. 7;

FIG. 9 is an explanatory diagram showing a relationship between numerical values of packet shaping in an upper layer protocol in a transmitting terminal in the second embodiment shown in the flowchart of FIG.

FIG. 10 is a view to which a third embodiment of the present invention is applicable;
FIG. 3 is an explanatory diagram illustrating a functional model in a transmission terminal.

[Explanation of symbols]

 1, 2 ATM terminal 3 ATM communication network 4, 21 Upper layer protocol 5, 22 Upper layer protocol function unit 6 ACR acquisition request function unit 7, 23 ACR value storage function unit 8, 28 ABR control unit 9, 24 ATM layer protocol function Unit 10, 26 ACR holding function unit 11 ACR acquisition request receiving function unit 12 ACR notification function unit 25 Interrupt generation function unit 27 ACR notification interrupt function unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5K030 GA01 HB13 KA03 KX11 LC02 LC11 LD11 MA13 MB15 5K034 AA01 HH50 KK28 MM08 MM11 9A001 BB04 CC06 DD10 JJ12 KK56 LL09

Claims (10)

[Claims] 1. A data transfer method in a communication protocol constructed based on a seven-layer model of an OSI layer, wherein a protocol process is performed only between a transmitting terminal and a receiving terminal, and a function of performing transfer control is provided. An upper layer protocol, a lower layer having a function of performing protocol processing between the transmitting terminal and the receiving terminal and in a network element in a communication network, and notifying the network element or the receiving terminal of transmittable rate information; And a lower layer protocol implemented in the transmitting terminal,
Holding the transmittable rate information notified from the network element or the receiving terminal, an upper layer protocol implemented in the transmitting terminal,
Acquiring transmittable rate information held in a lower layer protocol of the transmitting terminal, the upper layer protocol shaping transfer data based on the acquired transmittable rate information, and transmitting a communication packet to the lower layer protocol. A data transfer method characterized in that data is transferred to a data transfer device. 2. The lower-layer protocol implemented in the transmitting terminal includes a buffer memory having a desired capacity, a rate value of data transmitted from the upper-layer protocol to the lower-layer protocol, and a lower-layer protocol. When the rate value of the data to be transmitted into the network is different, the buffer temporarily stores the data to be transferred, thereby transferring all data between the upper layer protocol and the lower layer protocol. The data transfer method according to claim 1, wherein the data transfer method is allowed. 3. A threshold value corresponding to a data amount is set in a memory of the buffer, and when the data amount accumulated in the buffer memory passes the threshold value, a threshold value is set for the upper layer protocol. 3. The data transfer method according to claim 2, wherein a transmittable rate value of the lower layer protocol is notified. 4. The threshold value comprises two threshold values having different values. When the amount of data in the buffer memory exceeds a high threshold value and when a data amount falls below a low threshold value,
4. The data transfer method according to claim 3, wherein the upper layer protocol is notified of a transmittable rate value of the lower layer protocol. 5. The method according to claim 5, wherein the upper layer protocol uses the transmittable rate information acquired from the lower layer protocol, and the upper layer protocol relates to a rate of a communication packet transferred to the lower layer protocol. 5. The data transfer method according to claim 1, wherein the protocol performs shaping dynamically. 6. The data transfer method according to claim 1, wherein the upper protocol stores transmittable rate information acquired from the lower layer protocol. 7. The transmittable rate information stored in the upper layer protocol is updated by a transmittable rate newly acquired by the upper layer protocol from the lower layer protocol. Data transfer method. 8. The data transfer method according to claim 7, wherein the upper layer protocol updates stored transmittable rate information for each communication packet. 9. The data transfer method according to claim 8, wherein the upper layer protocol updates stored transmittable rate information for each of a plurality of packets. 10. The upper layer protocol updates the stored transmittable rate information by causing the lower layer protocol to interrupt the upper layer protocol. Item 10. The data transfer method according to any one of Items 9.