JP2024108417A - Network interface card and transmission performance monitoring method - Google Patents

Abstract

[Problem] To provide a technology that makes it possible to monitor the performance of data transmission in a NIC with respect to the stability of an optical output signal. [Solution] A network interface card includes a controller that controls the transmission and reception of data between the server and an external server via a bus interface inside the server, an optical communication module that is connected to the controller and converts data input from the controller from an electrical signal to an optical signal and outputs the converted data to an optical branching circuit connected to a network, a monitoring optical module that converts return data branched by the optical branching circuit from an optical signal to an electrical signal, and a monitoring unit that monitors the performance of data transmission based on the normality of the return data input from the monitoring optical module, and if the performance of data transmission deteriorates, the monitoring unit notifies the operating system or application inside the server. [Selected Figure] Figure 1

Description

The present disclosure relates to a network interface card and a transmission performance monitoring method.

There are cases where a server is expected to be used without any problems. However, if the server continues to be used even though the performance of its network interface card (hereafter referred to as NIC) has deteriorated, this can cause a major problem.

Patent document 1 describes a method for detecting transmission anomalies using packets to check communication path connectivity, using a dedicated loopback circuit provided inside an interface connected to a LAN.

JP 2021-034907 A

However, while the method described in Patent Document 1 can detect abnormalities occurring within the NIC, it cannot detect abnormalities such as unstable optical output signals in the optical module actually used for communication with the outside of the server.

One example of the objective of the present disclosure is to provide a technique that enables monitoring the performance of data transmission in a NIC with respect to the stability of an optical output signal.

A network interface card according to one aspect of the present disclosure includes a controller that controls the transmission and reception of data between the server and an external server via a bus interface inside the server; a communication optical module that is connected to the controller and converts data input from the controller from an electrical signal to an optical signal and outputs the converted data to an optical branching circuit that is connected to a network; a monitoring optical module that converts return data branched by the optical branching circuit from an optical signal to an electrical signal; and a monitoring unit that monitors the performance of data transmission by monitoring the normality of the return data input from the monitoring optical module, and the monitoring unit notifies an operating system or application inside the server when the performance of data transmission deteriorates.

In one aspect of the present disclosure, a transmission performance monitoring method is provided in a server, and in a network interface card that transmits and receives data between the server and a network, a computer monitors the normality of return data that is output by a controller provided in the network interface card, converted into an optical signal by a communication optical module, and further converted into an electrical signal by a monitoring optical module and input, thereby monitoring the performance of data transmission, and if the data transmission performance is degraded as a result of the monitoring, a notification is given to an operating system or application within the server.

One example of the effect of the present disclosure is the ability to monitor the performance of data transmission in a NIC with respect to the stability of the optical output signal.

2 is a block diagram showing an example of a configuration of a network interface unit including a NIC and an optical branching circuit; 1 is a block diagram showing an example of a communication system including a server including a NIC and a connected device. FIG. 11 is a sequence diagram showing the operation of each component related to monitoring of transmission performance. 13 is a flowchart showing an operation of a monitoring unit. FIG. 2 is a block diagram showing an example of the configuration of a NIC. 2 is a block diagram showing an example of a configuration of a network interface unit including a NIC and an optical branching circuit;

Embodiments of the present disclosure will be described in detail with reference to the drawings.

[Embodiment]
Fig. 1 is a block diagram showing an example of the configuration of a network interface unit including a NIC 100 and an optical branching circuit 200 in an embodiment. Referring to Fig. 1, the NIC 100 includes a controller 101, a communication optical module 102, a monitoring optical module 103, and a monitoring unit 104. The communication optical module 102 and the monitoring optical module 103 of the NIC 100 are communicatively connected to the optical branching circuit 200. The optical branching circuit 200 is communicatively connected to a network 300.

The NIC 100 is a device for realizing communication between network-related devices and a network. In this disclosure, the NIC 100 is provided in a server, which is a network-related device.

2 is a block diagram showing an example of a communication system including a server 10 including a NIC 100 in an embodiment and a connected device. Referring to FIG. 2, the server 10 includes a memory 11, a memory control unit 12, a processor 13, an input/output control unit 14, an input/output card slot 15, a system bus 18, and a bus interface 19. An active NIC 16 and a standby NIC 17 are inserted into the input/output card slot 15. The server 10 is connected to a network switch 30 so as to be able to communicate with the network switch 30. The active NIC 16 of the server 10 is connected to the network switch 30 via an optical branching circuit 200. The standby NIC of the server 10 is connected to the network switch 30. The network switch 30 is further connected to another server 40 so as to be able to communicate with the other server 40. The network switch 30 connects multiple servers and other devices. The network 300 is connected to the network switch 30, the server 10 connected to the network switch 30, and multiple other servers 40, and transmits data between them.

In FIG. 2, the active NIC 16 is the NIC 100 in FIG. 1. The standby NIC 17 may be the NIC 100 or a known NIC. The standby NIC 17 may be connected to the network switch 30 via the optical branching circuit 200, just like the active NIC 16.

Other configurations of the server 10 in FIG. 2 will be described. In the server 10, the memory 11 stores data. The memory 11 may be a RAM (Random Access Memory) or a ROM (Read Only Memory). The memory control unit 12 manages the memory 11. For example, the memory control unit 12 allocates a portion of the memory 11 and releases the allocated portion of the memory 11 upon request from the processor 13. The processor 13 performs information processing described in a program. The program is, for example, an operating system (hereinafter also referred to as an OS (Operating System)) or an application. When the input/output control unit 14 receives an input/output command from the processor, it inputs/outputs data between the server and an input/output device that inputs/outputs data. The input/output card slot 15 has an insertion port for an expansion device for connecting to various input/output devices. As described above, the active NIC 16 and the standby NIC 17 are inserted into the I/O card slot 15 in the server 10. The system bus 18 is a data transmission path that connects the processor 13 to other components. The bus interface 19 is a data transmission path that connects the I/O control unit 14 to the active NIC 16 and the standby NIC 17, respectively.

In FIG. 2, server 10 is equipped with an active NIC 16 and a backup NIC 17 as NICs. A configuration in which multiple devices are prepared in this manner is called a redundant configuration. In the server 10 of this disclosure, by having a redundant NIC configuration, even if one NIC fails, it is possible to repair or replace the failed NIC while continuing to operate the server 10 using the other NIC. This type of configuration is used in mission-critical servers, i.e., servers that cannot be stopped or malfunctioned.

So far, we have explained the server 10 in Figure 2, but the configuration of the server equipped with the NIC 100 is not limited to this.

The general functions of the NIC 100 will be described with reference to Figures 1 and 2. In Figures 1 and 2, the NIC 100 transmits data input from the server 10 via the bus interface 19 to the network 300. For example, the NIC 100 transmits data input from the server 10 via the bus interface 19 via the optical branching circuit 200 and the network switch 30 to another server 40. The NIC 100 also receives data from the network 300. For example, the NIC 100 receives data transmitted by another server 40 via the network switch 30 and the optical branching circuit 200, and inputs the data to the server 10.

Next, each component of the NIC 100 of this embodiment will be described in detail.

The controller 101 controls the transmission and reception of data between the server 10 and the external server 40 via the bus interface 19 inside the server 10. If the NIC 100 is a LAN interface card, the controller 101 is also called a LAN controller.

In FIG. 1, the communication optical module 102 is a device that converts data from an electrical signal to an optical signal and from an optical signal to an electrical signal. The communication optical module 102 is connected to the controller 101, and converts data input from the controller 101 from an electrical signal to an optical signal. The communication optical module 102 then outputs the converted data to an optical branching circuit 200 that is connected to the network. The optical branching circuit 200 will be described later.

The optical communication module 102 is, for example, a small form-factor pluggable (SFP) module, an SFP+ module, etc., but is not limited to these examples. The optical communication module 102 may suffer from a particular failure in which the optical output signal is unstable when converting data from an electrical signal to an optical signal. An unstable optical signal is visualized by a disturbance in the eye pattern, which is a display of multiple overlapping transitions of signal waveforms in transmissions including optical communications.

Here, the optical branching circuit 200 is a circuit that physically branches an input optical signal and outputs each of the branched optical signals. The optical branching circuit 200 is connected to the NIC 100 of the server 10 and to the network 300. More specifically, the optical branching circuit 200 is connected to the communication optical module 102 and the monitoring optical module 103 of the NIC 100 by the optical interface 20. The optical branching circuit 200 is also connected to a network switch 30 that is connected to another server 40.

The optical branching circuit 200 branches the optical signal input from the communication optical module 102. The optical branching circuit 200 outputs one of the branched optical signals to a connection destination included in the network 300, for example, the network switch 30. The optical branching circuit 200 then outputs the other of the branched optical signals to the monitoring optical module 103. In the following description, data transmitted from the server 10 to the network 300 or another server 40 is called "transmitted data," and data branched by the optical branching circuit 200 and output to the monitoring optical module 103 is called "loopback data."

The monitoring optical module 103 is a device that converts data from an optical signal to an electrical signal. The monitoring optical module 103 is connected to the optical branching circuit 200, and converts the return data branched by the optical branching circuit 200 from an optical signal to an electrical signal. The monitoring optical module 103 outputs the return data converted into an electrical signal to the monitoring unit 104.

In this disclosure, Rx and Tx included in optical modules for communication, optical branching circuits, optical modules for monitoring, etc. are receiving connectors and transmitting connectors, respectively.

The monitoring unit 104 monitors the data transmission performance. For example, the monitoring unit 104 determines that the data transmission performance has deteriorated. More specifically, the monitoring unit 104 monitors the data transmission performance of the NIC 100. The monitoring unit 104 monitors the data transmission performance based on the normality of the return data input from the monitoring optical module 103. The monitoring of the normality of the return data and the monitoring of the data transmission performance by the monitoring unit 104 will be described later. Furthermore, if the data transmission performance has deteriorated, the monitoring unit 104 notifies the operating system or application within the server 10 via the controller 101 that the data transmission performance has deteriorated. This notification will be described later.

First, the monitoring of the performance of data transmission by the monitoring unit 104 will be described. The monitoring unit 104 monitors the performance of data transmission based on a value related to abnormality detection calculated by a predetermined method from the error detection code of the return data. The error detection code is configured to be able to detect the occurrence of an error in data during data communication. For example, when transmitting data, the data transmitting side adds data calculated from the data to be transmitted in a certain procedure as an error detection code to the data to be transmitted. Then, the data receiving side detects an error by the error detection code added to the received data. For example, the monitoring unit 104 may detect an error in the return data by comparing the error detection code added to the frame of the return data with a value calculated from the frame data of the return data by the same method as the error detection calculation used when transmitting data. The error detection code is, for example, a Frame Check Sequence (FCS) or a Cyclic Redundancy Code (CRC), but is not limited to these examples.

When the monitoring unit 104 detects an abnormality in the return data by the error detection code, it counts the number of times the abnormality is detected as the first error count. When the number of first errors relative to the number of transmitted data is equal to or greater than a threshold value, the monitoring unit 104 may determine that the performance of data transmission has deteriorated.

The monitoring unit 104 may also monitor the performance of data transmission by taking into account abnormalities in the transmission data input to the controller 101 in addition to abnormalities in the return data. Here, the second error number is defined as the number of abnormalities in the transmission data input to the controller 101. The second error number may be monitored by the controller 101, or may be monitored by another additional configuration. In this case, the monitoring unit 104 may calculate the total number of errors, including the first error number and the second error number, and may determine that the performance of data transmission has deteriorated if the total error number relative to the number of transmitted data is equal to or greater than a threshold value.

The threshold value for determining that the data transmission performance has deteriorated may be set in advance. The threshold value may be, for example, 0.025%. The threshold value is not limited to this example and may be set as appropriate.

Next, a notification by the monitoring unit 104 regarding the degradation of data transmission performance when the data transmission performance is degraded will be described. As described above, when the data transmission performance is degraded, the monitoring unit 104 notifies the operating system or application in the server 10 via the controller 101 regarding the degradation of data transmission performance. This allows the server 10 to detect the degradation of the data transmission performance of the NIC 100 and take necessary measures accordingly. Here, the necessary measures include switching the NIC in a redundant configuration, or notifying the administrator that repair or replacement is necessary. In addition, for example, when the data transmission performance is degraded, the monitoring unit 104 may notify the operating system or application in the server 10 via the controller 101 that the NIC needs to be switched. Switching the NIC means switching the NIC from the currently used NIC (e.g., the active NIC 16) to the spare NIC (e.g., the spare NIC 17).

In addition, when the data transmission performance is degraded, the monitoring unit 104 may stop the transmission and reception of data by the NIC 100. For example, the monitoring unit 104 stops the data transmission of the controller 101, and stops the data transmission by the network interface card by returning an error return in response to a transmission request from a network adapter driver provided inside the server 10.

The operation of data transmission performance monitoring using the NIC 100 and peripheral devices configured as described above will be explained with reference to the sequence diagram in Figure 3.

As shown in FIG. 3, first, the communication optical module 102 converts the transmission data (electrical signal) input from the controller 101 from an electrical signal to an optical signal (step S101). Then, the communication optical module 102 outputs the converted transmission data (optical signal) to the optical branching circuit 200 (step S102).

Next, the optical branching circuit 200 branches the transmission data (optical signal) input from the communication optical module 102 (step S103). The optical branching circuit 200 outputs one of the branched optical signals to the network 300 as the transmission data (optical signal) (step S104). Then, the optical branching circuit 200 outputs the other of the branched optical signals to the monitoring optical module 103 as return data (optical signal) (step S105).

Next, the monitoring optical module 103 converts the input return data (optical signal) from an optical signal to an electrical signal (step S106). Then, the monitoring optical module 103 outputs the converted return data (electrical signal) to the monitoring unit 104 (step S107).

Then, the monitoring unit 104 monitors the normality of the return data (electrical signal) input from the monitoring optical module 103 (step S108). The operation of the monitoring unit 104 after step S108 will be described later with reference to the flowchart in FIG. 4.

This completes the series of operations performed by the NIC 100 and the peripheral device. This series of operations may be performed sequentially in response to data transmission by the server 10.

Next, the operation of the monitoring unit 104 will be described with reference to the flowchart in FIG.

First, the monitoring unit 104 receives return data (electrical signal) from the monitoring optical module 103 (step S201).

The monitoring unit 104 determines whether there is an abnormality in the return data (electrical signal) (step S202). The monitoring unit 104 determines whether there is an abnormality by using, for example, an error detection code such as FCS or CRC.

If there is an abnormality in the return data (electrical signal) (step S202: Yes), the monitoring unit 104 counts the first error number by performing an error counter registration (step S203). If there is no abnormality in the return data (electrical signal) (step S202: No), the monitoring unit 104 returns to the operation of step S201.

Then, the monitoring unit 104 determines whether the rate of data abnormality occurrence exceeds a threshold value (step S204). The rate of data abnormality occurrence may be the number of first errors relative to the number of transmitted data, or the total number of first errors and the number of second errors relative to the number of transmitted data, as described above.

When the monitoring unit 104 determines that the rate of data abnormality occurrence exceeds the threshold (step S204: Yes), it determines that the data transmission performance of the NIC 100 has deteriorated, and notifies the operating system and applications of the server 10 (step S205). When the monitoring unit 104 determines that the rate of data abnormality occurrence exceeds the threshold, the monitoring unit 104 may stop data transmission and reception by the NIC 100, as described above. When the monitoring unit 104 determines that the rate of data abnormality occurrence exceeds the threshold (step S204: No), the monitoring unit 104 returns to the operation of step S201.

The monitoring unit 104 then completes the series of operations.

The NIC in the above-described embodiment includes a controller, a communication optical module, a monitoring optical module, and a monitoring unit. The communication optical module and the monitoring optical module are each connected to an optical branching circuit. The communication optical module converts the server's transmission data from an electrical signal to an optical signal, and inputs the converted optical signal to the optical branching circuit. The optical signal, which is the return data branched by the optical branching circuit, is input to the monitoring optical module. The monitoring optical module converts the return data from an optical signal to an electrical signal, and inputs it to the monitoring unit. The monitoring unit then monitors the normality of the return data, thereby monitoring the performance of data transmission.

As a result, the NIC in this embodiment can monitor the performance of data transmission in the NIC.

NICs equipped with optical modules that convert electrical signals into optical signals can experience failures specific to optical modules, such as unstable optical output signals.

If the optical output signal is unstable, the destination receiving the data sent from the NIC in the server, for example a LAN (Local Area Network) switch, will not detect a link-down due to a drop in optical reception strength. The destination receiving the data sent from the NIC in the server will detect an abnormality in the received data and discard the received data. An abnormality in the data received at the destination is, for example, a bit error such as an FCS (Frame Check Sequence) error. Then, the error count in the MIB (Management Information Base) of the corresponding receiving port of the network switch is incremented, leaving a history of abnormalities in the received frame. Furthermore, at the upper TCP (Transmission Control Protocol) layer between servers, due to the specifications of the TCP/IP (Internet Protocol) network, if there is no reception response from the other party, recovery is achieved by resending the frame.

In this case, however, the NIC installed in the server has no way of recognizing that the data it sent has been detected as abnormal and discarded by the connected party. As a result, the server's higher-level application software cannot recognize that a degradation in communication performance has occurred.

In response to this, the monitoring unit monitors the normality of the return data that is converted from an electrical signal to an optical signal by the communication optical module and branched off by the optical branching circuit. This makes it possible to detect errors that occur in the communication optical module that could not be detected by the server side, which is the source of the data. This allows the server equipped with the NIC to take measures against degradation of the NIC's performance, including errors that occur in the communication optical module.

[Modification]
Next, a modification of the embodiment will be described with reference to the drawings.

Figure 5 is a diagram showing the functional configuration of a NIC 100A including an optical branching circuit 200. As shown in Figure 5, the NIC may be configured to include an optical branching circuit within the NIC.

Also, FIG. 6 is a block diagram showing the functional configuration of the NIC 100 and the optical branching circuit 200A. As shown in FIG. 6, the optical branching circuit 200A may be configured to branch and output an input optical signal. That is, compared to the optical branching circuit 200 in the embodiment, the optical branching circuit 200A does not include a receiving connector that receives data from the network 300 and a transmitting connector that outputs the input data. In this case, the communication optical module 102 of the NIC 100 may receive data directly from the network 300.

In addition, the optical branching circuit 200A may also be provided within the NIC, similar to the optical branching circuit 200.

The present invention has been described above with reference to each embodiment, but the present invention is not limited to the above-mentioned embodiment. The configuration and details of the present invention can be modified in various ways that are understandable to those skilled in the art and are within the scope of the present invention.

Although the operations are described in sequence in the form of a flowchart, the order of description does not limit the order in which the operations are performed. Therefore, when implementing each embodiment, the order of the operations may be changed as long as it does not interfere with the content.

Some or all of the above embodiments may be described as follows, but are not limited to the following:

(Appendix 1)
a controller that controls transmission and reception of data between the server and an external server via a bus interface within the server;
an optical communication module connected to the controller, converting data input from the controller from an electrical signal to an optical signal, and outputting the converted data to an optical branching circuit connected to a network;
a monitoring optical module that converts the return data branched by the optical branch circuit from an optical signal to an electrical signal;
a monitoring unit that monitors performance of data transmission based on the normality of the return data input from the monitoring optical module;
The monitoring unit notifies an operating system or an application in the server when performance of data transmission is degraded.
Network interface card.

(Appendix 2)
The network interface card according to claim 1, wherein the monitoring unit monitors data transmission performance based on a value relating to abnormality detection calculated by a predetermined method from the error detection code of the return data.

(Appendix 3)
The monitoring unit counts an abnormality in the return data as a first error number, and determines that data transmission performance has deteriorated if the number of first errors relative to the amount of transmitted data is equal to or greater than a threshold.

(Appendix 4)
The monitoring unit calculates a total number of errors, which is a second error number, which is the number of abnormalities in the transmission data input to the controller, and the first error number, and if the total error number relative to the number of transmission data is equal to or greater than a threshold, it determines that data transmission performance has deteriorated.

(Appendix 5)
The network interface card according to any one of claims 1 to 4, wherein the monitoring unit stops data transmission and reception by the network interface card when data transmission and reception performance is degraded.

(Appendix 6)
The network interface card described in Appendix 5, wherein when data transmission performance deteriorates, the monitoring unit stops data transmission from the controller and returns an error return in response to a transmission request from a network adapter driver provided inside the server, thereby stopping data transmission and reception by the network interface card.

(Appendix 7)
When the performance of the data transmission is degraded, the monitoring unit notifies an operating system or an application in the server of the degradation of the performance of the data transmission.
7. The network interface card of claim 1.

(Appendix 8)
When the performance of data transmission is degraded, the monitoring unit notifies an operating system or an application in the server that a network interface card needs to be switched.
8. The network interface card according to any one of claims 1 to 7.

(Appendix 9)
The optical branch circuit is further provided.
9. The network interface card of claim 1.

(Appendix 10)
A network interface card provided in a server for transmitting and receiving data between the server and a network,
The computer
monitoring performance of data transmission and reception based on normality of return data outputted from a controller included in the network interface card, converted into an optical signal by a communication optical module, and further converted into an electrical signal by a monitoring optical module and inputted;
If the monitoring results in a deterioration in data transmission/reception performance, a notification is sent to the operating system or application within the server.
Monitoring methods.

REFERENCE SIGNS LIST 10 Server 11 Memory 12 Memory control unit 13 Processor 14 Input/output control unit 15 Input/output card slot 16 Active network interface card 17 Backup network interface card 18 System bus 19 Bus interface 20 Optical interface 30 Network switch 40 Server 100 Network interface card 100A Network interface card 101 Controller 102 Communication optical module 103 Monitoring optical module 104 Monitoring unit 200 Optical branching circuit 200A Optical branching circuit 300 Network

Claims (10)

a controller that controls transmission and reception of data between the server and an external server via a bus interface within the server;
an optical communication module connected to the controller, converting data input from the controller from an electrical signal to an optical signal, and outputting the converted data to an optical branching circuit connected to a network;
a monitoring optical module that converts the loopback data branched by the optical branch circuit from an optical signal to an electrical signal;
a monitoring unit that monitors performance of data transmission based on the normality of the return data input from the monitoring optical module;
The monitoring unit notifies an operating system or an application in the server when performance of data transmission is degraded.
Network interface card. The network interface card according to claim 1 , wherein the monitoring unit monitors data transmission performance based on a value relating to abnormality detection calculated by a predetermined method from the error detection code of the return data. 2. The network interface card according to claim 1, wherein the monitoring unit counts an abnormality in the return data as a first error number, and determines that data transmission performance has deteriorated if the number of first errors relative to the amount of transmitted data is equal to or greater than a threshold value. 4. The network interface card of claim 3, wherein the monitoring unit calculates a total number of errors including a second error number, which is a count of abnormalities in the transmission data input to the controller, and the first error number, and determines that data transmission performance has deteriorated if the total error number relative to the number of transmission data is equal to or greater than a threshold value. The network interface card according to claim 1 , wherein the monitoring unit stops data transmission/reception by the network interface card when data transmission/reception performance is degraded. 6. The network interface card according to claim 5, wherein the monitoring unit, when data transmission performance deteriorates, stops data transmission from the controller and returns an error return in response to a transmission request from a network adapter driver provided inside the server, thereby stopping data transmission and reception by the network interface card. When the performance of the data transmission is degraded, the monitoring unit notifies an operating system or an application in the server of the degradation of the performance of the data transmission.
2. The network interface card of claim 1. When the performance of data transmission is degraded, the monitoring unit notifies an operating system or an application in the server that a network interface card needs to be switched.
2. The network interface card of claim 1. The optical branch circuit is further provided.
9. A network interface card according to any one of claims 1 to 8. A network interface card provided in a server for transmitting and receiving data between the server and a network,
The computer
monitoring performance of data transmission and reception based on normality of return data outputted from a controller included in the network interface card, converted into an optical signal by a communication optical module, and further converted into an electrical signal by a monitoring optical module and inputted;
If the monitoring results in a deterioration in data transmission/reception performance, a notification is sent to the operating system or application within the server.
A method for monitoring performance degradation.

Images