JP2021087170A - Arithmetic device - Google Patents

Abstract

To provide a technique that reduces variations in loads and averages the processing load in data communication.SOLUTION: An ECU 10 includes: at least two or more microcomputers (for example, a transmitting-side microcomputer 11 and a receiving-side microcomputer 12), and the microcomputers are connected to each other by two types of communication buses of a first communication bus 30 and a second communication bus 31. The first communication bus transmits and receives environment information on a load on the ECU 10, and the second communication bus transmits and receives data between the microcomputers, selects a data compression method so as to distribute the processing load according to the environment information, and transmits the compressed data via the second communication bus.SELECTED DRAWING: Figure 1

Description

The present invention relates to a data communication technique between microcomputers in an arithmetic unit.

Electronic control devices such as ECUs (Electronic Control Units), which are a type of arithmetic unit, carry out various data communications using microcomputers (hereinafter referred to as microcomputers).
Currently, various communication buses (for example, the Internet, LAN (Local Area Network), etc.) are used for data communication.

In recent years, sensing of people and objects in the vicinity has become important in realizing autonomous driving technology, and it is possible to acquire and control data from cameras, RADAR (Radio Detecting and Ranging), LiDAR (Light Detection and Ranging), etc. It is needed. The amount of these data is increasing in proportion to the increase in the level of autonomous driving and driving support. Due to this increase in the amount of data, the load on the ECU is also increasing. In order to reduce this load, in data communication, data is acquired by compressing the data on the transmitting side and decompressing the data on the receiving side. For example, in Patent Document 1, the data compression rate in data communication is determined by comparing the network load and the CPU load of a computer, which are connected by one network line.

JP-A-2002-229886

However, in data communication, depending on the CPU load status of the transmitting side microcomputer and the receiving side microcomputer, it takes time for the data compression (decompression) processing, so that the CPU load varies. Further, depending on the load of the communication bus, the amount of transmitted data, and the like, the load balance of the entire arithmetic unit may be lost depending on the data transmission control method. As a result, the processing may not be completed within the cycle time (for example, 10 milliseconds), which may affect the subsequent processing (for example, the cycle processing).

Therefore, an object of the present invention is to provide a technique for reducing variation in load balance and further leveling the processing load (for example, distributing the load).

In order to achieve the above object, in the present invention, data transmission control according to the environmental information is performed by using different communication buses, each of which transmits / receives the data to be processed and the environmental information regarding the load in the arithmetic unit. Is to do.

More specifically, in an arithmetic unit that executes a predetermined calculation, a plurality of microcomputers and the plurality of microcomputers are connected to each other and have two types of communication buses that are different from each other. A first communication bus for transmitting and receiving environmental information regarding a load in the arithmetic unit and a second communication bus for transmitting and receiving data to be calculated between the plurality of microcomputers, and environmental information transmitted by the first communication bus. Is an arithmetic unit that specifies a transmission control method when transmitting the data on the second communication bus, and executes transmission control according to the specified transmission control method.

Further, the transmission control method includes control of a compression method such as a data compression rate and transmission timing.

The present invention also includes a processing method in the arithmetic unit and a program product for executing the method.

According to the present invention, it is possible to reduce the variation in the load balance and control the transmission so that the processing load is more leveled. Issues, configurations and effects other than those described above will be clarified by the description of the following embodiments.

The figure which shows the structure of the ECU 10 in the Example of this invention. The figure which shows the structure of the microcomputer (the transmitting side microcomputer 11, the receiving side microcomputer 12) in the Example of this invention. The flowchart which shows the process of Example 1. The figure which shows the judgment table (the 1) used in Examples 1 and 3. The flowchart which shows the process of Example 2. The figure which shows the judgment table (the 2) used in Example 2. The flowchart which shows the process of Example 3. The figure which shows the correlation of the data amount, compression time, decompression time, compression rate, and CPU load factor. It is a figure which shows an example of the load situation during periodic processing in CPU 40 of each microcomputer. The flowchart which shows the process of Example 4.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing a configuration of an ECU 10 which is a kind of arithmetic unit in the first embodiment. Two microcomputers (transmitting side microcomputer 11 and receiving side microcomputer 12) are mounted on the ECU, and communication is performed via the first communication bus 30 and the second communication bus 31. The first and second communication buses include, for example, SPI communication and Ethernet® communication. The communication of the second communication bus 31 and the communication with the external network 20 are performed via the Switch 13.

The first communication bus 30 is used for transmitting and receiving environmental information, and the second communication bus 31 is used for transmitting and receiving data. In this embodiment, two different types of communication buses are used, but one of them may be used for transmitting and receiving environmental information and the other may be used for transmitting and receiving data as the same type of communication bus. Further, the type of communication bus is not limited to SPI communication and Ethernet. Further, it is more desirable that the second communication bus supports high speed or large capacity than the first communication bus, but the present invention is not limited to this. The environmental information is information related to the processing load of the ECU 10 or each microcomputer (transmitting side microcomputer 11, receiving side microcomputer) such as CPU load information.

FIG. 2 is a diagram showing a configuration of each microcomputer included in the ECU 10. In the present specification, the microcomputers are a transmitting side microcomputer 11 for transmitting data and a receiving side microcomputer 12 for receiving data. Here, since the transmitting side microcomputer 11 and the receiving side microcomputer 12 have the same configuration, the outline of their functions will be described together. Therefore, the display below the hyphen of the code assigned to each microcomputer is omitted here. The detailed functions of the transmitting side microcomputer 11 and the receiving side microcomputer 12 will be described later when the flowchart of FIG. 3 is described.

Each microcomputer is composed of a CPU 40 that performs arithmetic processing, a volatile RAM 41 (Random Access Memory), and a non-volatile ROM 42 (Read Only Memory). Further, a CPU load calculation unit 43 that calculates a CPU load indicating a CPU load, a compression / decompression control unit 44 that performs data compression / decompression processing, a communication bus load calculation unit 45 that calculates a communication bus load, and data communication are performed. It includes a communication control unit 46. In FIG. 2, the first and second communication buses are collectively represented.

Here, the CPU load calculation unit 43 calculates, for example, the CPU load from the processing time of the program per unit time. The communication bus load calculation unit 45 calculates the communication bus load from, for example, the number of packets transmitted / received per unit time in the network band and the packet size. The communication control unit 46 transmits the CPU load acquired from the CPU load calculation unit 43 to the other microcomputer in a fixed cycle or in response to a request from the other microcomputer. For example, it notifies every 10 milliseconds. Further, the communication control unit 46 receives the CPU load transmitted from another microcomputer.

In FIG. 2, the CPU load calculation unit 43, the compression / expansion control unit 44, and the communication bus load calculation unit 45 for calculating the communication bus load are represented by different configurations, but they may be realized by one processor, or the CPU 40. May have that function.

Next, a flowchart showing the processing of this embodiment is shown in FIG. In this embodiment, the data compression method (compression rate) and the transmission timing are determined using the CPU load and the communication bus load as the environmental information. Such environmental information is used because the CPU load and the communication bus load are likely to fluctuate, that is, other environmental information (for example, the amount of transmitted data) is constant. Therefore, when the amount of transmitted data is constant, this flowchart is executed.

The configurations of FIGS. 1 and 2 described above are also common to each of the examples after the second embodiment.

Hereinafter, the processing content will be described with reference to FIG. This process is executed every transmission timing or every process cycle in the ECU 10. Further, here, it is assumed that data for calculation is transmitted from the transmitting side microcomputer 11 to the receiving side microcomputer 12.

First, in step S201, the CPU 40-1 of the transmitting side microcomputer 11 acquires the transmitting side CPU load indicating the load of itself (CPU 40-1) from the CPU load calculating unit 43-1.

Next, in step S202, the CPU 40-1 acquires the receiving side CPU load indicating the load of the CPU 40-2 of the receiving side microcomputer 12 from the communication control unit 46-1.

For this purpose, the CPU 40-1 transmits a CPU load request to the receiving side microcomputer 12 via the communication control unit 46-1 via the first communication bus. In the receiving-side microcomputer 12, the communication control unit 46-2 receives the CPU load request. Then, the receiving side CPU load calculated by the CPU load calculating unit 43-2 is transmitted from the communication control unit 46-1 to the transmitting side microcomputer 11 via the first communication bus 30. In this way
The CPU load calculation unit 43-2 may calculate the CPU load on the receiving side for each processing cycle in the ECU 10, or may calculate it when a CPU load request is received.

Next, in step S203, the CPU 40-1 acquires a communication bus load indicating the load of the second communication bus from the communication bus load calculation unit 45-1.

Finally, in step S204, the CPU 40-1 uses the transmitting side CPU load, the receiving side CPU load, and the communication bus load acquired in steps S201 to S203 according to the determination table (FIG. 4) stored in the ROM 42 to perform data. Determine the compression method and transmission timing.

Hereinafter, the details of step S204 will be described with reference to the determination table shown in FIG. FIG. 4 is a determination table for determining the data compression method and the transmission timing from the CPU load and the communication bus load when the amount of transmission data is constant.

Here, the constant transmission data amount indicates a case where a predetermined condition is satisfied. Specifically, it includes that the fluctuation of the amount of transmitted data is within a certain range. This indicates the fluctuation for each cycle, and it is particularly preferable to use whether the difference from the amount of transmitted data in the previous cycle is within the range.

In this determination table, the transmission side CPU load and the reception side CPU load are on the vertical axis, and the communication bus load is on the vertical axis. That is, the environment information including the set of the transmission side CPU load and the reception side CPU load, the combination of the communication bus load, and the communication control method including the data compression method and the transmission timing are associated with each other.

Therefore, in S204, it is specified which of the determination tables corresponds to the values of the transmission side CPU load and the reception side CPU load acquired in S201 and S202, and the communication bus load acquired in S203, respectively.

Here, for example, a CPU load of 50% or more is defined as "high", and a CPU load of less than 50% is defined as "low". Further, the threshold value of the communication bus load is set to 80%.

Therefore, it is determined whether the acquired transmission side CPU load or reception side CPU load is "high" or "low". Further, it is determined whether the communication bus load acquired in S203 is "greater than or equal to the threshold value" or "less than or equal to the threshold value". The information corresponding to these judgment results, that is, the data compression method and the transmission timing are specified from the judgment table.

Here, the concept of the determination table will be described with reference to FIG. 8 showing the relationship between the environmental information and the transmission control method. Specifically, FIG. 8 summarizes the correlation between the amount of transmitted data, the compression time, the compression rate, the CPU load, and the decompression time. According to this figure, it can be understood that the compression rate, the CPU load, and the compression time are inversely proportional to the decompression time. That is, the higher the compression rate, the higher the CPU load and compression time, but the shorter the decompression time.

Therefore, the relationship between the transmission side CPU load, the reception side CPU load, the communication bus load, and the compression method can be grasped as follows.
(1) When the transmission side CPU load is high (for example, 50% or more), the compression rate is relatively low, and when the transmission side CPU load is low (for example, less than 50%), the compression rate is relatively high. Is desired.
(2) When the CPU load on the receiving side is high, the compression rate is relatively high so as to shorten the decompression time, and when the CPU load on the receiving side is low, the decompression time can be long, so the compression rate is relatively low. It is possible.

Further, as the communication bus load is larger, it takes more time and load to transmit new data. Therefore, it is desirable to control the transmission timing according to the communication bus load.

In consideration of these, the judgment table of FIG. 4 has the following characteristics.

When the communication bus load is equal to or higher than the threshold value (for example, 80% or higher) and the transmission side CPU load is high, the data compression / transmission processing may affect the subsequent periodic processing. Further, since the communication bus load is also equal to or higher than the threshold value, there is no guarantee that the data can be transmitted normally even if the data is transmitted. Therefore, the transmission at the relevant timing is postponed. Then, the data compression / transmission process is performed at the next timing.

In addition, the next timing may include the next and subsequent times, and may be the timing one after another. In addition, the judgment may be made again at the next timing.

If the communication bus load is equal to or higher than the threshold value and the CPU load on the transmitting side is low, the subsequent periodic processing is not affected. Therefore, data compression is performed and data transmission is performed at the next timing. In this case, the compression rate of data compression may be either high or low.

If the communication bus load is less than the threshold value and both the transmitting side and receiving side CPU loads are high, performing data compression may affect the subsequent periodic processing. Further, since the communication bus load is also equal to or higher than the threshold value, there is no guarantee that the data can be transmitted normally even if the data is transmitted. Therefore, the data compression / transmission process is performed at the next timing.

When the communication bus load is less than the threshold value and one of the CPU loads is high, a data compression method in which the processing load is leveled is selected. That is, when the transmission side CPU load is low, compression is performed by a method having a relatively high compression rate so that the load on the CPU 40-1 is reduced. On the contrary, when the CPU load on the receiving side is high, compression is performed by a method having a relatively high compression rate so that the load on the CPU 40-2 is reduced.

If the communication bus load is less than the threshold value and both the transmitting side and receiving side CPU loads are low, either the compression rate is increased or decreased, which does not affect the subsequent periodic processing. Therefore, which data compression method is used. May be selected.

When the communication bus load is less than the threshold value, the CPU load on the transmitting side is low, and the CPU load on the receiving side is high, data is transmitted by a method having a high compression rate. This is because the CPU load of the receiving side microcomputer 12 is high, so that the load due to the decompression processing is not applied to the CPU 40-2 of the receiving side microcomputer 12 as much as possible.

In the flowchart of FIG. 3, when both the transmitting side and the receiving side CPU load are “high” according to S201 and S202, step S203 may be skipped and the process may proceed to S204. That is, when both the transmitting side and the receiving side CPU load are "high", the transmission is performed at the next timing regardless of the communication bus load, and the transmission control method is specified.

FIG. 9 is a diagram showing an example of a load status during periodic processing in the CPU 40 of each microcomputer. This is an example in which the CPU load of the transmitting side microcomputer 11 is low and the CPU load of the receiving side microcomputer 12 is high. Since the CPU 40-1 of the transmitting side microcomputer 11 has a low load (there is a margin in free time), the compression rate is increased and data is transmitted. After receiving the data, the CPU 40-2 of the receiving-side microcomputer 12 performs decompression processing. If a method having a low compression rate is selected in this case, the data decompression time becomes long, and the decompression process may not be completed within the free time of the CPU of the receiving microcomputer.

In this way, by checking the CPU load of the transmitting side microcomputer 11, the CPU load of the receiving side microcomputer 12, and the communication bus load information, and selecting the compression method in which the processing load is leveled, the processing is not completed within the cycle time. You can get closer to completion. On the contrary, the present embodiment also includes performing transmission control so that the processing is completed within the cycle time.

In this embodiment, the amount of transmitted data is used instead of the communication bus load information in step S203 as the selection of the data compression method in the first embodiment. That is, the data compression method (compression rate) and the transmission timing are determined by using the CPU load and the amount of transmission data as the environment information. Such environmental information is used because the CPU load and the amount of transmitted data are likely to fluctuate, that is, other environmental information (for example, the communication bus load) is constant. Therefore, when the communication bus load is constant, this flowchart is executed. In addition, a certain idea in this Example is the same as Example 1.

FIG. 5 shows a flowchart of this embodiment. Since steps S201 and S202 execute the same processing as the flowchart shown in FIG. 3, the description thereof will be omitted.

In step S303, in step S203, the CPU 40-1 acquires the transmission data amount from the communication control unit 46-1. This will acquire the amount of data transmitted by the second communication bus in this cycle.

Finally, in step S304, the data compression method and the transmission timing are determined. At this time, the CPU 40-1 uses the transmission side CPU load, the reception side CPU load, and the transmission data amount acquired in steps S201, S202, and S303 according to the determination table (FIG. 6) stored in the ROM 42.

Hereinafter, the details of step S304 will be described with reference to the determination table shown in FIG. FIG. 6 is a determination table for determining the data compression method and the transmission timing from the CPU load and the amount of transmission data when the communication bus load is constant.

In this determination table, the transmission side CPU load and the reception side CPU load are on the vertical axis, and the transmission data amount is on the vertical axis. That is, the environment information including the set of the transmission side CPU load and the reception side CPU load, the combination of the transmission data amount, and the communication control method including the data compression method and the transmission timing are associated with each other.

Therefore, in S304, it is specified which of the determination table corresponds to the transmission side CPU load and the reception side CPU load acquired in S201 and S202, and the value of the communication data amount acquired in S303, respectively.

Here, for example, a CPU load of 50% or more is defined as “high”, and a CPU load of less than 50% is defined as “low”. In addition, the threshold value for the amount of transmitted data is set to 1 megabyte or more.

Therefore, it is determined whether the acquired transmission side CPU load or reception side CPU load is "high" or "low". Further, it is determined whether the amount of transmitted data acquired in S303 is "greater than or equal to the threshold value" or "less than or equal to the threshold value". The information corresponding to these judgment results, that is, the data compression method and the transmission timing are specified from the judgment table.

Here, reference is made to FIG. 8 again. FIG. 8 summarizes the correlation between the amount of transmitted data, the compression time, the compression rate, the CPU load, and the decompression time. According to FIG. 8, the amount of data is proportional to the compression time. The compression rate, CPU load, and compression time are inversely proportional to the decompression time.

Therefore, according to FIG. 8, in addition to the above-mentioned (1) and (2), the following can be grasped.
(3) The larger (larger) the amount of transmitted data, the longer it takes to transmit new data, and therefore it is desirable to control the transmission timing according to the amount of transmitted data.

In consideration of these, the judgment table of FIG. 6 has the following specifications.

Regardless of the amount of data to be transmitted, if both the CPU load on the transmitting side and the CPU load on the receiving side are high, performing data compression / transmission processing may affect the subsequent periodic processing, so transmission at that timing is postponed. Then, the data compression / transmission process is performed at the next timing. In addition, the next timing may include the next and subsequent times, and may be the timing one after another. In addition, the judgment may be made again at the next timing.

If the amount of data to be transmitted is equal to or greater than the threshold value (for example, 1 megabyte or more) and one of the CPU load on the transmitting side and the receiving side is high, a data compression method in which the processing load is leveled is selected. That is, when the transmission side CPU load is low, compression is performed by a method having a relatively high compression rate so that the load on the CPU 40-1 is reduced. On the contrary, when the CPU load on the receiving side is high, compression is performed by a method having a relatively high compression rate so that the load on the CPU 40-2 is reduced.

When the amount of transmitted data is equal to or greater than the threshold value (for example, 1 megabyte or more) and both the transmitting side and receiving side CPU loads are low, whether the compression rate is increased or decreased affects the subsequent periodic processing. Any data compression method may be selected because there is no data.

When the amount of transmitted data is less than the threshold (for example, less than 1 megabyte) and the CPU load on both the transmitting side and the receiving side is not high, the amount of transmitted data is small and the efficiency is improved even if data compression is performed. Since it is not limited, data transmission is performed without compressing the data.

Also in this embodiment, if both the transmitting side and the receiving side CPU load are "high" according to S201 and S202 in the flowchart of FIG. 5, step S203 may be skipped and the process may proceed to S204. That is, when both the transmitting side and the receiving side CPU load are "high", the transmission is performed at the next timing regardless of the amount of transmission data, and the transmission control method is specified.

In this way, also in the second embodiment, the processing is completed or completed within the cycle time by checking the transmitting side CPU load, the receiving side CPU load, and the amount of transmitted data and selecting a compression method in which the processing load is leveled. You can get closer. Further, as in the first embodiment, the present embodiment also includes performing transmission control so as to complete the process within the cycle time.

In the first embodiment, the value actually notified is used for the receiving CPU load, but there is a possibility that the receiving CPU load cannot be acquired due to a failure of the communication bus 1 or the like. Therefore, in this embodiment, the receiving side CPU load predicted by the transmitting side microcomputer 11 is used instead of the receiving side CPU load in step S202 of the first embodiment.

FIG. 7 shows a flowchart of this embodiment. Here, the description of steps S201, S202, and S204 will be omitted because they overlap with those of the first embodiment.

In step S402, the CPU 40-1 acquires the receiving side CPU load predicted by the communication control unit 46-1, or predicts the receiving side CPU load by itself. For this purpose, the CPU 40-1 determines whether or not the receiving side CPU load can be acquired from the receiving side microcomputer 12 a certain number of times (for example, three times). This determination may be executed using the number of times the CPU load request is transmitted, or may be executed using the number of times of acquisition timing. As a result, if you cannot get it
Prediction is performed using a representative value including the average value of the CPU load information on the receiving side for the last three times. This prediction may be executed by using a CPU load other than the receiving side microcomputer 12, or may use the historical data and use the receiving side CPU load of the same period in the past.

In addition, this embodiment is not limited to the case where it cannot be obtained. For example, the prediction may be made by omitting the judgment as to whether or not the receiving side CPU load can be acquired. Further, this embodiment can be applied to the case where the transmitted data amount of the second embodiment is used.

By checking the CPU load on the transmitting side and the CPU load on the communication bus, predicting the CPU load on the receiving side, and selecting a compression method in which the processing load is leveled, the processing can be completed or approached within the cycle time. On the contrary, the transmission control is also included in this embodiment so that the processing is completed within the cycle time.

In each of the above-described embodiments, processing is shown on the premise that the communication bus load or the amount of transmitted data is constant. Therefore, in this embodiment, it is determined whether the communication bus load or the transmission data amount is constant, and the transmission control method is specified according to the result. Hereinafter, the contents will be described with reference to FIG.

First, in step S501, the CPU 40-1 determines whether the transmission timing is reached. This determines whether it is the next cycle of the periodic transmission. As a result, if it is the transmission timing, the process proceeds to step S502. If it is not the transmission timing, it returns to the original. The step S501 may be omitted and the step S502 may be started. In this case, it is preferable to determine the transmission timing when starting FIGS. 3 and 5, which will be described later.

Next, in step S502, the CPU 40-1 determines whether the fluctuation of the communication bus load or the amount of transmitted data is constant. Here, the fluctuation indicates the fluctuation for each cycle as described above, and it is particularly preferable to use whether the difference from the communication bus load or the amount of transmitted data in the previous cycle is within the range. is there. Further, it may be selected that the variation is smaller in the past fixed number of transmissions. In this case, it is possible to use that the difference between the minimum value and the maximum value is small and the deviation is small as the variation.

As a result, if it is determined in step S502 that the amount of transmitted data is constant, the process proceeds to the processes of FIGS. 3 to 7. If it is determined that the communication bus load is constant, the process proceeds to FIG. Note that the process of FIG. 5 includes step S402 instead of step S202, that is, predicting the CPU load on the receiving side. Further, the process of FIG. 10 and the processes of steps S201 and S202 (or S402) in FIGS. 3, 5 and 7 may be processed in parallel. In this case, the processes after step S502 proceed to S203 to S303, respectively. This is the end of the description of Example 4.

In each of the above embodiments, the CPU load and the transmission data amount or the communication bus load are used as the environmental information, but at least one of the CPU load, the transmission data amount, and the communication bus load may be used. In particular, when any one of these is used, the information in which the environmental information and the transmission control method are associated with each other on a one-to-one basis is used instead of the determination table corresponding to FIGS.

In this case, it is preferable to execute the determination of which environmental information to use, the CPU load, the amount of transmitted data, or the communication bus load, as follows. Whether to use the transmission data amount or the communication bus load (which is constant) is also determined for the CPU load, and the determination is made according to the conditions. As described above, it is preferable to use whether the load is constant. Further, each environmental information may be used by omitting the above-mentioned determination process, or other environmental information may be used.

Further, in each embodiment, communication between two microcomputers is taken as an example, but three or more microcomputers may be used, and not all microcomputers need to be the same. Further, each embodiment can be similarly used regardless of the medium for communicating data (for example, communication between ECUs and communication between cores).

The present invention is not limited to the above-described examples, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to the one including all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration. Further, each of the above configurations, functions, processing units, processing procedures, and the like may be realized by hardware by designing a part or all of them by, for example, an integrated circuit. Information such as programs, tables, and files that realize each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an SD card.

Further, each embodiment includes an automobile control module including the ECU 10 and an automobile, and also includes a control method using these. Furthermore, when each embodiment is applied to an arithmetic unit instead of the ECU 10, an information device (computer) including the arithmetic unit and a system including the arithmetic unit are also included in the category of each embodiment. In this case, the processing method in the information device or the system is also included in the category of each embodiment.

10 ... ECU
11 ... Transmitter microcomputer 11
12 ... Receiving microcomputer 12
13 ... Switch
20 ... External network 30 ... First communication bus 31 ... Second communication bus 40 ... CPU
41 ... RAM
42 ... ROM
43 ... CPU load calculation unit 44 ... Compression / decompression control unit 45 ... Communication bus load calculation unit 46 ... Communication control unit

Claims (10)

In an arithmetic unit that executes a predetermined operation,
With multiple microcomputers
It connects between the plurality of microcomputers and has two types of communication buses that are different from each other.
The two types of communication buses are a first communication bus for transmitting and receiving environmental information regarding a load in the arithmetic unit and a second communication bus for transmitting and receiving data to be calculated between the plurality of microcomputers, respectively.
Using the environmental information transmitted by the first communication bus, the transmission control method when transmitting the data by the second communication bus is specified.
An arithmetic unit characterized in that transmission control is executed according to the specified transmission control method. In the arithmetic unit according to claim 1,
An arithmetic unit characterized in that the transmission control method is specified by using a determination table showing a correspondence relationship between the environment information and the transmission control method. In the arithmetic unit according to any one of claims 1 or 2.
As the transmission control method, an arithmetic unit characterized in that at least one of a compression rate and a transmission timing for the data is specified. In the arithmetic unit according to any one of claims 1 to 3.
An arithmetic unit characterized in that the environmental information includes any one of the CPU load of the microcomputer, the communication bus load of the communication bus, and the amount of transmitted data. In the arithmetic unit according to any one of claims 1 to 3.
An arithmetic unit characterized in that the environmental information includes either the CPU load of the microcomputer and the communication bus load of the communication bus or the amount of transmitted data. In the arithmetic unit according to claim 5.
It is determined whether the communication bus load or the transmitted data amount is constant, and the result is determined.
As a result of the determination, an arithmetic unit characterized in that non-constant information is used as the environmental information. In the arithmetic unit according to any one of claims 5 or 6.
As the transmission control method, the transmission timing for the data is specified, and the transmission timing is specified.
When the CPU load is higher than a predetermined reference in both the first microcomputer for transmitting the data and the second microcomputer for receiving the data, the data is transmitted from the predetermined transmission timing. An arithmetic unit characterized by transmitting at a later transmission timing. In the arithmetic unit according to claim 7.
As the transmission control method, the compression ratio for the data is further specified.
An arithmetic unit characterized in that at least the compression rate is specified according to the relationship between the communication bus load or the transmitted data amount and a predetermined threshold value. In the arithmetic unit according to claim 8.
When the CPU load is higher than the reference by either the first microcomputer for transmitting the data or the second microcomputer for receiving the data, the compression ratio is specified. Arithmetic logic unit. In the arithmetic unit according to any one of claims 1 to 9.
An arithmetic unit characterized in that the environmental information uses a predicted value predicted by the microcomputer.

Images