Method for over-temperature control in network system

Abstract

The invention provides a method for over-temperature control in a network system. The temperature of the network system is monitored in real time during operation of the network system in a normal mode to provide a two-stage temperature active notification mechanism, and after it is determined that an over-temperature error has occurred, the network system is instructed to switch to operate in a light load mode to reduce waste heat. When it is determined that the over-temperature problem may be overcome according to the temperature change or load shedding time of the network system, or when a restart signal issued by a remote server is received, the network system is instructed to switch back to the normal mode to operate so as to provide an active system endowment mechanism.

Description

Technical Field

[0001] The present invention relates to a method for over-temperature control in a network system, and in particular to a method for predicting over-temperature events, providing over-temperature countermeasures and providing active system assignment in a network system. Background Art

[0002] A data center is a facility that an organization uses to store important applications, data, and IT equipment. It typically contains servers, storage systems, network equipment (such as network switches, routers, and firewalls), application traffic controllers, and the racks and cabling needed to organize and connect the equipment.

[0003] The infrastructure of modern data centers has evolved from traditional on-premises physical servers to virtual networks that support applications and workloads in physical infrastructure clusters, and even to multi-cloud environments. Today's data not only exists in itself, but is also connected between multiple data centers, edges, and public / private clouds. A network switch is a network hardware device responsible for network bridging. It can establish network connections within a data center or between different data centers, use packet switching to receive data, and then transmit the data to the target device in an intelligent way.

[0004] Because data center workloads rely on high-performance, scalable, and reliable network switching operations, many network switches are typically mounted in racks or cabinets. Network switches emit considerable waste heat when in operation, especially when running at full load with high bandwidth demands, which can easily lead to overheating of components, thus shutting down the entire network, resulting in a lot of downtime and outages. Therefore, network switches use several cooling strategies: passive cooling or active cooling. Passive cooling systems achieve natural convection and heat dissipation through the use of heat sinks. Active cooling, on the other hand, refers to cooling techniques that rely on external devices to enhance heat transfer. For example, fanless network switches without any built-in fans can use passive cooling systems to ensure quiet operation and do not disrupt its working environment. Network switches with built-in fans use active cooling systems, and built-in fans bring forced air when natural convection is not enough to remove heat.

[0005] For high temperature specifications of products in the information technology (IT) market, fans are usually used to solve the problem of component overheating caused by the field. On the other hand, the operation and process control of industrial control systems (ICS) are usually referred to as operational technology (OT), which is performed in isolated and independent networks. Even if an operating mechanism for system overtemperature protection is provided, the customer's control over the environment cannot be guaranteed in the harsh environment of some OT applications. There may even be dangerous operations that exceed product specifications, causing damage to components and interruption of actual link communication, ultimately resulting in the system being unable to recover.

[0006] Regardless of the cooling strategy adopted, existing network switches for OT applications do not provide any over-temperature prediction, over-temperature countermeasures, and active system reset mechanisms to enable administrators to monitor the operating status of network switches in real time. Summary of the invention

[0007] The present invention provides a method for over-temperature control in a network system, which includes the network system operating in a normal mode and monitoring the temperature of the network system in real time during the operation of the network system; when it is determined that the temperature of the network system is higher than an alarm temperature, the network system notifies a remote server of an over-temperature error that may be about to occur; when it is determined that the temperature of the network system remains higher than a maximum operating temperature for more than a first period of time, the network system notifies the remote server of an over-temperature error that has occurred and switches from the normal mode to a light-load mode. The heat energy generated by the network system when operating in the light-load mode is lower than the heat energy generated when operating in the normal mode. When the temperature of the network system is higher than the maximum operating temperature, the network system cannot operate normally, and the alarm temperature is not higher than the maximum operating temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 The present invention is a flowchart of a method for predicting over-temperature events, providing over-temperature countermeasures and providing system recovery in a network system according to an embodiment of the present invention. Figure 2 The present invention is a flowchart of a method for predicting over-temperature events, providing over-temperature countermeasures and providing system recovery in a network system in another embodiment of the present invention. DETAILED DESCRIPTION

[0009] Figure 1 The present invention is a flowchart of a method for predicting over-temperature events, providing over-temperature countermeasures, and providing active system assignment in a network system in an embodiment of the present invention. Figure 1 The flowchart shown contains the following steps:

[0010] Step 110: The network system operates in a normal mode, and the temperature of the network system is monitored in real time during the operation.

[0011] Step 120: Determine whether the temperature of the network system is higher than an alarm temperature T ALERT ; If so, execute step 130; if not, execute step 110.

[0012] Step 130: Record information about an over-temperature error that may be about to occur on the network system, and notify a remote server of the over-temperature error that may be about to occur; execute step 140.

[0013] Step 140: Determine whether the temperature of the network system is maintained above a maximum operating temperature T MAX exceeds the length of a first time period; if so, execute step 150; if not, execute step 110.

[0014] Step 150: Record relevant information about the over-temperature error that has occurred on the network system, and notify the remote server of the over-temperature error that has occurred; execute step 160.

[0015] Step 160 : The network system operates in a light load mode; and executing step 170 .

[0016] Step 170 : Determine whether the time for which the network system operates in the light load mode exceeds the length of a second period; if so, execute step 200 ; if not, execute step 170 .

[0017] Step 200: Determine whether a restart command is received; if so, execute step 210; if not, execute step 200.

[0018] Step 210: Restart the network system; execute step 110.

[0019] In the present invention, the network system may be a network switch for OT applications, but is not limited thereto. As is well known to those with general knowledge in the relevant field, the temperature tolerance range of electronic equipment refers to the temperature range in which the electronic equipment is designed to operate stably for a long time and maintain acceptable electrical characteristics stability. Generally speaking, the temperature tolerance range of industrial-grade equipment is 2-3 times larger than the temperature tolerance range of standard commercial-grade equipment of the same type.

[0020] For the purpose of illustration, it is assumed that the temperature tolerance range of the network system of the present invention is composed of a maximum operating temperature T MAX and a minimum operating temperature T MIN To define, the alarm temperature T ALERT The value is less than the maximum operating temperature T of the network systemMAX And greater than the minimum operating temperature T of the network system MIN . Alarm temperature T ALERT The value of is determined according to the temperature margin designed by the network system of the present invention for over-temperature errors. In one embodiment, the alarm temperature T ALERT The value of (T MAX -T ALERT ) is less than (T ALERT -T MIN ). In this way, when the current temperature of the network system reaches the alarm temperature T ALERT When , it means that the network system has less temperature margin before an over-temperature error occurs, that is, any temperature rise will more easily lead to an over-temperature error.

[0021] In step 110, the network system is configured to operate in normal mode and monitor its temperature in real time during operation. In step 120, the present invention determines whether the temperature of the network system during operation is higher than the alarm temperature T ALERT When the network system temperature is determined to be higher than the alarm temperature T ALERT When, it means that the network system is prone to overheating error at this time. In this case, the present invention will perform step 130 to record the information of the relevant overheating error that may be about to occur on the network system, and notify the remote server of the overheating error that may be about to occur.

[0022] In step 140, the present invention determines whether the temperature of the network system during operation reaches the alarm temperature T ALERT After that, whether it is maintained above the maximum operating temperature T MAX When operating in normal mode, if the temperature of the network system reaches the alarm temperature T ALERT Then it continues to rise and remains higher than the maximum operating temperature T MAX If the length of the first period of time is exceeded, it means that the network system cannot run stably for a long time and cannot maintain acceptable electrical characteristic stability, and an over-temperature error is determined to have occurred. In this case, the present invention will execute step 150 to record the information related to the over-temperature error on the network system and notify the remote server of the over-temperature error.

[0023] In steps 130 and 150, the present invention can provide centralized monitoring and data logging, and the data recorded by the network system can provide important event and performance data. In steps 130 and 150, the data recorded on the network system will be transmitted to the remote server for collection, analysis and correlation, thereby providing preventive measures and contingency plans for over-temperature errors. In addition, the remote server can monitor the operating status of the network system based on the notification information without directly logging into the network system.

[0024] As is well known to those with ordinary knowledge in the relevant fields, an application, a system, or even a simple electronic device is composed of several components, each of which performs a specific function to the best of its ability, leaving other tasks to other components that are specifically responsible for handling. For example, in the network system of the present invention, hardware and software components work together to analyze and forward data flows based on administrator configuration. Three planes are defined in the field of network infrastructure: data plane, control plane, and management plane.

[0025] In a network architecture, the data plane is used to forward user data packets, provide forwarding resources required by the control plane's operating policies, and support data packet transmission between end users. The control plane is the location in the network architecture where the processes used to store routing information and execute network protocols are located. The management plane allows administrators to read devices, analyze their status and health, and reconfigure devices. When administrators change the configuration, the management plane notifies the control plane to change the corresponding control plane program. The management plane can also store programs that monitor the status of the device itself, such as fan speed, power supply failure, and the status of components in other planes.

[0026] There are two ways to manage the network infrastructure: in-band management or out-of-band management (OOBM). In one embodiment, the network system of the present invention may be an in-band management network system, in which the in-band management traffic and the data traffic use the same data plane. In this case, the network system of the present invention may use the common communication protocol (such as Telnet / SSH or SNMP) adopted by its data plane to notify the remote server of an over-temperature error that may be about to occur (step 130) or an over-temperature error has occurred (step 150). However, the network infrastructure management method adopted by the network system of the present invention does not limit the scope of the present invention.

[0027] In another embodiment, the network system of the present invention may be an OOBM network system, in which out-of-band management traffic uses a dedicated channel, while data traffic uses a data plane, and the two do not share the data channel. Generally speaking, out-of-band management is a safer and more resilient method of managing network infrastructure because it can operate normally even during data traffic congestion, equipment failure, or network attacks. In an OOBM network system, an administrator can monitor and manage each device through a dedicated management port. In this case, the network system of the present invention can notify the remote server of an over-temperature error that may be about to occur (step 130) or an over-temperature error that has occurred (step 150) through a dedicated management port. However, the network infrastructure management method adopted by the network system of the present invention does not limit the scope of the present invention.

[0028] In step 160, the network system of the present invention switches from normal mode to light load mode after determining that an over temperature error has occurred. In one embodiment, after switching to light load mode, the network system is completely shut down to avoid generating waste heat. In another embodiment, after switching to light load mode, the network system disables all in-band interfaces to shut down all data traffic to reduce load and reduce waste heat. In another embodiment, after switching to light load mode, the network system disables all in-band interfaces to shut down all data traffic to reduce load and reduce waste heat, but at the same time maintains at least one OOB port or one console port enabled to receive possible user commands.

[0029] In step 170, when it is determined that the network system has been running in the light load mode for a period of time exceeding the length of the second period, the temperature of the network system is likely to have dropped to within its temperature tolerance range. In this case, the present invention will execute step 200 to determine whether a restart command has been received from the remote server through the dedicated management port, the enabled OOB port, or the enabled console port. If the administrator issues a restart command through the dedicated management port, the enabled OOB port, or the enabled console port, or a preset restart command is activated after the preset period expires, the present invention will execute step 210 to restart the network system, and then execute step 110 to resume operation in normal mode, and monitor the temperature of the network system in real time during operation.

[0030] Figure 2 The present invention is a flowchart of a method for predicting over-temperature events, providing over-temperature countermeasures, and providing active system reassignment in a network system according to another embodiment of the present invention. Figure 2 The flowchart shown contains the following steps:

[0031] Step 110: The network system operates in a normal mode, and the temperature of the network system is monitored in real time during the operation.

[0032] Step 120: Determine whether the temperature of the network system is higher than an alarm temperature T ALERT ; If so, execute step 130; if not, execute step 110.

[0033] Step 130: Record information about an over-temperature error that may be about to occur on the network system, and notify a remote server of the over-temperature error that may be about to occur; execute step 140.

[0034] Step 140: Determine whether the temperature of the network system is maintained above a maximum operating temperature T MAX exceeds the length of a first time period; if so, execute step 150; if not, execute step 110.

[0035] Step 150: Record relevant information about the over-temperature error that has occurred on the network system, and notify the remote server of the over-temperature error that has occurred; execute step 160.

[0036] Step 160 : The network system operates in a light load mode; and executing step 180 .

[0037] Step 180: Determine whether the temperature of the network system is maintained above the maximum operating temperature T MAX exceeds the length of a third time period; if so, execute step 200; if not, execute step 190.

[0038] Step 190 : Determine whether the temperature of the network system is within its temperature tolerance range; if so, execute step 110 ; if not, execute step 180 .

[0039] Step 200: Determine whether a restart command is received; if so, execute step 210; if not, execute step 200.

[0040] Step 210: Restart the network system; execute step 110.

[0041] Figure 2 The embodiments shown and Figure 1 The differences between the embodiments shown are that Figure 2 The flowchart shown also includes steps 180 and 190, and does not include step 170. After the network system of the present invention switches to the light load mode in step 160 due to determining that an over-temperature error occurs, steps 180 and 190 are then executed to determine whether the network system running in the light load mode can overcome the over-temperature problem.

[0042] During the operation of the network system in the light load mode, if it is determined in step 180 that the temperature of the network system is no longer greater than the maximum operating temperature T MAX, then step 190 is executed to determine whether the temperature of the network system is within its temperature tolerance range. If it is determined in step 180 that the temperature of the network system is no longer greater than the maximum operating temperature T MAX , and in step 190, it is determined that the temperature of the network system has returned to its temperature tolerance range, indicating that the network system has overcome the overheating problem after running in the light load mode for a period of time. In this case, the present invention will execute step 110 to resume operation in the normal mode and monitor the temperature of the network system in real time during operation.

[0043] When it is determined in step 180 that the temperature of the network system remains higher than the maximum operating temperature T MAX When the length of the third period is exceeded, it means that the network system has not overcome the overheating problem after running in the light load mode for a period of time. In this case, the present invention will execute step 200 to determine whether a restart command has been received from the remote server through the dedicated management port, the enabled OOB port, or the enabled console port. If the administrator issues a restart command through the dedicated management port, the enabled OOB port, or the enabled console port, or a preset restart command is activated after the preset period expires, the present invention will execute step 210 to restart the network system, and then execute step 110 to resume operation in normal mode, and monitor the temperature of the network system in real time during operation.

[0044] In summary, the present invention provides a two-stage temperature active notification mechanism in the network system (determining the alarm temperature T of an over-temperature error that may occur soon) ALERT and the maximum operating temperature T at which an over-temperature error has occurred MAX ), and after determining that an over-temperature error has occurred, instructs the network system to switch from normal mode to light-load mode, and tries to overcome the over-temperature problem by completely shutting down the system or disabling the data traffic of all in-band interfaces to reduce the load. In addition, the present invention also provides an active system reassignment mechanism in the network system, which instructs the network system to switch back to normal mode when it is determined that the over-temperature problem has been overcome based on the temperature change of the network system, when it is determined that the over-temperature problem may have been overcome based on the load reduction time of the network system, or when a restart signal is received from the administrator through a remote server. Therefore, the present invention can provide a method for predicting over-temperature events, providing over-temperature countermeasures, and providing active system reassignment in a network system. The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should fall within the scope of the present invention. Description of Reference Numerals

[0045] 100-210: Steps

Claims

1. A method for overtemperature control in a network system, characterized in that: Include: The network system operates in a normal mode, and the temperature of the network system is monitored in real time during the operation of the network system; When it is determined that the temperature of the network system is higher than an alarm temperature, the network system notifies a remote server of a possible impending over-temperature error; When it is determined that the temperature of the network system remains higher than a maximum operating temperature for more than a first period of time, the network system notifies the remote server of an over-temperature error that has occurred and switches from the normal mode to a light-load mode, wherein: The heat energy generated by the network system when operating in the light-load mode is lower than the heat energy generated when operating in the normal mode; When the temperature of the network system is higher than the maximum operating temperature, the network system cannot operate normally; and The alarm temperature is not higher than the maximum operating temperature.

2. The method according to claim 1, characterized in that Also includes: When it is determined that the temperature of the network system is higher than the alarm temperature, recording information related to the over-temperature error that may be about to occur on the network system; as well as When it is determined that the temperature of the network system remains higher than the maximum operating temperature for longer than the length of the first period of time, information related to the over-temperature error that has occurred is recorded on the network system.

3. The method according to claim 1, characterized in that When operating in the light load mode, the network system is completely shut down.

4. The method according to claim 1, characterized in that When operating in the light load mode, the network system disables all in-band interfaces to shut down all data traffic.

5. The method according to claim 1, characterized in that Also includes: When it is determined that the network system has been operating in the light load mode for a period of time exceeding a second period of time due to the over-temperature error that has occurred, determining whether a restart command has been received; and After determining that the restart command is received, the network system is restarted to make the network system resume to operate in the normal mode.

6. The method according to claim 1, characterized in that Also includes: After the network system switches from the normal mode to the light-load mode due to the over-temperature error that has occurred, determining whether the temperature of the network system remains higher than the maximum operating temperature for more than a third period of time; During the operation of the network system in the light load mode, when it is determined that the temperature of the network system does not remain higher than the maximum operating temperature for more than the length of the third period, determining whether the temperature of the network system is within a temperature tolerance range; as well as When it is determined that the temperature of the network system is within the temperature tolerance range, the network system resumes operating in the normal mode, wherein: When the temperature of the network system is within the temperature tolerance range, the network system can operate normally; The upper limit of the temperature tolerance range is the maximum operating temperature; The lower limit of the temperature tolerance range is a minimum operating temperature; and And the alarm temperature is not lower than the minimum operating temperature.

7. The method according to claim 1, characterized in that Also includes: After the network system switches from the normal mode to the light-load mode due to the over-temperature error that has occurred, determining whether the temperature of the network system remains higher than the maximum operating temperature for more than a third period of time; During the operation of the network system in the light load mode, when it is determined that the temperature of the network system remains higher than the maximum operating temperature for more than the length of the third period, determining whether a restart command is received; and After determining that the restart command is received, the network system is restarted to make the network system resume to operate in the normal mode.

8. The method according to any one of claims 5 and 7, characterized in that: When operating in the light load mode, the network system disables all in-band interfaces to shut down all data traffic, but keeps at least one out-of-band port or one console port enabled to receive the restart command.

9. The method according to any one of claims 5 and 7, characterized in that Also includes: The network system determines whether the reboot command is received via a dedicated management port, an out-of-band port, or a console port.

10. The method according to claim 1, characterized in that: The network system uses a universal communication protocol or a dedicated management port adopted by a data plane to notify the remote server of the over-temperature error that may be about to occur or the over-temperature error that has occurred.

Images