A server onboard IO link monitoring device and method

Abstract

The application relates to the technical field of servers, and discloses a server on-board IO link monitoring device and method, which comprises a signal processing module, a signal scheduling module, a BMC diagnosis module, a programmable detachable power supply module and a detachable IO controller test module. The device can monitor the link state of IO equipment in a server, so that a user can learn about abnormal links in the server, can quickly and accurately locate the IO controller that fails in the abnormal IO link, can turn off the power supply of the faulty IO controller, can avoid affecting normal business of other IO links of the server, and can shorten the fault maintenance time and reduce the maintenance labor cost.

Description

Technical Field

[0001] This invention relates to the field of server technology, and in particular to a server onboard I / O link monitoring device and method. Background Technology

[0002] A server is a device that provides computing services. Because servers need to respond to and process service requests, they possess the capability to undertake and guarantee services. The hardware components of a server are similar to those of a general-purpose computer architecture, with onboard I / O controllers such as hard drive controllers, USB controllers, and network controllers on the motherboard. However, due to the need to provide highly reliable services, servers have higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0003] Server motherboards typically have various onboard I / O controllers with corresponding link types. These controllers communicate with the processor via these links and, in response to the processor's control, provide computing services to clients. Because servers have various types of onboard I / O controllers, the malfunction of these links determines whether various I / O devices (hard drives, serial devices, USB devices, and network devices, etc.) can respond correctly to the processor's control, thus affecting the functionality of some server services. Currently, I / O link malfunctions in servers usually exist during the server startup phase, but are only discovered by users when the server provides computing services through these faulty I / O devices. Therefore, potential I / O link malfunctions often impact the normal operation of the server, and current technologies struggle to detect these malfunctions in a timely manner. Summary of the Invention

[0004] This invention provides a server-side onboard I / O link monitoring device and method to solve the problem in the prior art that it is difficult to detect link anomalies of I / O devices in servers in a timely manner.

[0005] To achieve the above objectives, the present invention employs the following technical solution:

[0006] In a first aspect, the present invention provides a server onboard I / O link monitoring device, applied to a server onboard I / O link, wherein the server onboard I / O link includes a Phytium processor, a primary I / O controller, a secondary I / O controller, and I / O devices. The monitoring device includes: a signal processing module, a signal scheduling module, a BMC diagnostic module, a programmable discrete power supply module, and a discrete I / O controller signal testing module; the Phytium processor, the primary I / O controller, the secondary I / O controller, and the I / O devices are all connected through the signal processing module.

[0007] The signal processing module is configured to collect IO link signals through an internal signal collector and feed back to a signal scheduling module in the case of power-on of the server.

[0008] The signal scheduling module is configured to monitor the IO link signals in real time to obtain monitoring information and synchronously feed back the monitoring information to a BMC diagnosis module in real time.

[0009] The BMC diagnosis module is configured to diagnose the onboard IO link according to the monitoring information, switch the IO link of the signal processing module through the signal scheduling module in the case of existence of an abnormality in the IO link, and separately test the primary IO controller and the secondary IO controller through the signal scheduling module. The separable IO controller signal test module injects test IO link data signals to the front end of each IO controller and monitors the state of the back end IO link data communication of each IO controller to lock the IO controller with no output data communication signal in the back end IO link, and closes the abnormal IO controller through the programmable separable power supply module. In the case of non-existence of an abnormality in the IO link, the BMC diagnosis module monitors the server onboard IO link in segments through the signal scheduling module and the signal processing module in real time.

[0010] Optionally, the signal processing module comprises a first signal processing module, a second signal processing module and a third signal processing module, the Feiteng processor is connected with the primary IO controller through the first signal processing module, the primary IO controller is connected with the secondary IO controller through the second signal processing module, the secondary IO controller is connected with the IO device through the third signal processing module, and the first signal processing module, the second signal processing module and the third signal processing module are connected with the signal scheduling module.

[0011] Optionally, the signal scheduling module is a PFX switch, and the PFX switch supports 100 channels, 52 ports, 26 virtual switch partitions and 48 non-transparent bridges.

[0012] Optionally, the signal processing module is an equalizer, and the equalizer comprises a signal collector and a signal switcher, the signal collector is connected with the RX signal channel and the TX signal channel of the signal switcher respectively, the signal switcher is connected with the Feiteng processor, the signal scheduling module and the primary IO controller respectively, and the signal collector is further connected with the signal scheduling module.

[0013] Optionally, the signal scheduling module comprises a first signal management module, a second signal management module, a signal monitoring module and a signal selector, the first signal management module is connected with the detachable IO controller signal test module, the first signal processing module, the signal selector, the signal monitoring module and the second signal management module, the signal monitoring module is further connected with the second signal management module and the BMC diagnosis module, the second signal management module is further connected with the signal selector, the third signal processing module and the detachable IO controller signal test module, and the signal selector is further connected with the second signal processing module.

[0014] Optionally, the monitoring information comprises IO link segment performance, IO link communication state and real-time captured eye diagram.

[0015] Optionally, the Feiteng processor, the first signal processing module and the primary IO controller constitute a first IO link segment, the primary IO controller, the second signal processing module and the secondary IO controller constitute a second IO link segment, and the secondary IO controller, the third signal processing module and the IO device constitute a third IO link segment.

[0016] In a second aspect, the application provides a server on-board IO link monitoring method, applied to the server on-board IO link monitoring device of the first aspect, the method comprising:

[0017] The signal processing module collects IO link signals through an internal signal collector and feeds back to the signal scheduling module in the case of starting on-board IO link data communication on the server;

[0018] The signal scheduling module obtains monitoring information by real-time monitoring of the IO link signals and synchronizes the monitoring information to the BMC diagnosis module in real time;

[0019] The BMC diagnosis module performs abnormal diagnosis on the on-board IO link according to the monitoring information, switches the IO link of the signal processing module through the signal scheduling module in the case of existence of IO link abnormality, and the signal scheduling module tests the primary IO controller and the secondary IO controller separately; the detachable IO controller signal test module injects test IO link data signals to the front end of each IO controller, monitors the data communication state of the rear end of each IO controller, locks the IO controller without output data communication signal at the rear end of the IO link, and closes the abnormal IO controller through the programmable detachable power supply module; in the case of non-existence of IO link abnormality, the BMC diagnosis module performs segmented monitoring on the server on-board IO link in real time through the signal scheduling module and the signal processing module.

[0020] Optionally, the BMC diagnosis module switches the IO link of the signal processing module through the signal scheduling module, comprising:

[0021] The BMC diagnosis module isolates the communication state of the upstream and downstream IO controller connection in the current IO link, and switches to the signal scheduling module for signal input.

[0022] Optionally, the signal scheduling module separately tests the primary IO controller and the secondary IO controller, comprising:

[0023] The signal scheduling module detects the in-place state of the detachable IO controller signal test module, loads the detachable IO controller signal test module driver according to the in-place state, and enables the detachable IO controller signal test module to communicate with the IO controller.

[0024] Advantages:

[0025] The server on-board IO link monitoring device provided by the application comprises a signal processing module, a signal scheduling module, a BMC diagnosis module, a programmable detachable power supply module, and a detachable IO controller signal test module, can realize monitoring of the link state of the IO device of the server, thereby ensuring that the user can learn about the abnormal link in the server, and can quickly and accurately locate the IO controller that fails in the abnormal IO link, and turn off the power supply of the faulty IO controller, so as to avoid affecting the normal business of other IO links of the server, and shorten the fault maintenance time and reduce the maintenance labor cost. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 FIG. 1 is a structural schematic diagram of a server on-board IO link monitoring device according to a preferred embodiment of the application;

[0027] Figure 2 FIG. 2 is a structural schematic diagram of a signal processing module according to a preferred embodiment of the application;

[0028] Figure 3 FIG. 3 is a structural schematic diagram of a signal scheduling module according to a preferred embodiment of the application;

[0029] Figure 4 FIG. 4 is a flowchart of a server on-board IO link monitoring method according to a preferred embodiment of the application. DETAILED DESCRIPTION

[0030] The technical solutions of the application will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0031] Unless otherwise defined, technical terms and scientific terms used in the present application shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. Unless otherwise defined, the terms "first", "second", and the like, used in the present application do not necessarily have any sequence or order, but are used to differentiate one area from another. Similarly, the terms "one", "another", and the like, do not necessarily refer to a single or individual item, but can refer to one or more items. The terms "connected", "coupled", and the like, do not necessarily mean physically or mechanically connected, but can include electrical connection, whether direct or indirect. The terms "upper", "lower", "left", "right", and the like, are used only to indicate relative positional relationships, and when the absolute positions of the described objects are changed, the relative positional relationships are also changed accordingly.

[0032] Referring to Figure 1 The server on-board IO link monitoring device provided by the present application is applied to a server on-board IO link, the server on-board IO link comprises a Feiteng processor, a first IO controller, a second IO controller, and an IO device, and the monitoring device comprises a signal processing module, a signal scheduling module, a BMC diagnosis module, a programmable detachable power supply module, and a detachable IO controller signal test module; the Feiteng processor, the first IO controller, the second IO controller, and the IO device are all connected through the signal processing module;

[0033] The signal processing module is used for collecting IO link signals through an internal signal collector and feeding back to the signal scheduling module in the case of starting the on-board IO link data communication on the server;

[0034] The signal scheduling module is used for monitoring the IO link signals in real time to obtain monitoring information and synchronizing the monitoring information to the BMC diagnosis module in real time;

[0035] The BMC diagnosis module is used for diagnosing the on-board IO link abnormity according to the monitoring information, switching the IO link of the signal processing module through the signal scheduling module in the case of IO link abnormity, separately testing the first IO controller and the second IO controller through the signal scheduling module, injecting test IO link data signals to the front end of each IO controller through the detachable IO controller signal test module, monitoring the back end IO link data communication state of each IO controller, locking the IO controller without output data communication signal of the back end IO link, and closing the abnormal IO controller through the programmable detachable power supply module; in the case of no IO link abnormity, the BMC diagnosis module is used for monitoring the server on-board IO link in sections through the signal scheduling module and the signal processing module in real time.

[0036] The programmable detachable power module is connected with the Feiteng processor, the first IO controller, the second IO controller, the IO device and the BMC diagnosis module respectively, the first signal processing module, the second signal processing module and the third signal processing module are connected with the signal scheduling module, and the signal scheduling module is further connected with the BMC diagnosis module and the detachable IO controller signal test module.

[0037] In the embodiment, the IO link refers to a bus in a physical sense in a server, and can be a common PCI-Express (peripheral component interconnect express) bus, a USB bus, a SAS (Serial Attached SCSI) bus or a RGMII (Reduced Gigabit Media Independent Interface) network bus.

[0038] The programmable detachable power module refers to a power module that can adjust the output of a plurality of power supply voltages / currents through software programming and can be replaced. The programmable detachable power module can individually turn off the power supply of a faulty IO controller in the on-board IO link of the server system during operation, so as to avoid the influence of normal business of other IO link controllers of the server.

[0039] The server on-board IO link monitoring device described above includes a signal processing module, a signal scheduling module, a BMC diagnosis module, a programmable detachable power module and a detachable IO controller signal test module, and can realize monitoring of the link state of the IO device of the server, so as to ensure that the user can learn about the abnormal link in the server and quickly and accurately locate the faulty IO controller in the abnormal IO link, turn off the power supply of the faulty IO controller, avoid the influence on the normal business of other IO links of the server, and shorten the fault repair time and reduce the maintenance labor cost.

[0040] Optionally, the signal processing module includes a first signal processing module, a second signal processing module and a third signal processing module, the Feiteng processor is connected with the first signal processing module, the first IO controller is connected with the second signal processing module, the second IO controller is connected with the third signal processing module, and the first signal processing module, the second signal processing module and the third signal processing module are connected with the signal scheduling module.

[0041] Optionally, the signal scheduling module is a PFX switch (Microchip Technology Switchtec PFX Gen 4 fanout switch), which supports 100 channels, 52 ports, 26 virtual switch partitions and 48 non-transparent bridges.

[0042] In the optional embodiment, the PFX switch provides each port with hot plug and sudden plug controllers, advanced error control, comprehensive diagnosis and debugging functions, various I / O interfaces and an integrated MIPS processor. The comprehensive diagnosis and debugging can include signal generators, synthesizers and analyzers, per-port performance and error counters, various loopback modes and real-time eye diagram capture.

[0043] Optionally, referring to Figure 2 , the signal processing module is an equalizer, which includes a signal collector and a signal switcher, the signal collector is connected with the RX signal channel and the TX signal channel of the signal switcher respectively, the signal switcher is connected with the Feiteng processor, the signal scheduling module and the primary IO controller respectively, and the signal collector is further connected with the signal scheduling module.

[0044] In the embodiment, the RX (Receive) signal refers to the received signal in the bus differential signal, and the TX (Transport) signal refers to the transmitted signal in the bus differential signal.

[0045] Optionally, referring to Figure 3 , the signal scheduling module includes a first signal management module, a second signal management module, a signal monitoring module and a signal selector, the first signal management module is connected with the detachable IO controller signal test module, the first signal processing module, the signal selector, the signal monitoring module and the second signal management module, the signal monitoring module is further connected with the second signal management module and the BMC diagnosis module, the second signal management module is further connected with the signal selector, the third signal processing module and the detachable IO controller signal test module, and the signal selector is further connected with the second signal processing module.

[0046] Optionally, the monitoring information includes IO link segment performance, IO link communication state and real-time captured eye diagram.

[0047] Optionally, the Feiteng processor, the first signal processing module and the primary IO controller form a first IO link segment, the primary IO controller, the second signal processing module and the secondary IO controller form a second IO link segment, and the secondary IO controller, the third signal processing module and the IO device form a third IO link segment.

[0048] In this embodiment, the IO link is divided into three link segments, which can facilitate segmented monitoring of different link segments and make monitoring more flexible.

[0049] As a general inventive concept, please refer to Figure 4 The application also provides a server onboard IO link monitoring method, which is applied to the server onboard IO link monitoring device and includes the following steps.

[0050] The signal processing module collects IO link signals through an internal signal collector and feeds them back to the signal scheduling module when the server is powered on and starts onboard IO link data communication.

[0051] The signal scheduling module obtains monitoring information by monitoring IO link signals in real time and synchronizes the monitoring information to the BMC diagnosis module in real time.

[0052] The BMC diagnosis module diagnoses abnormalities of the onboard IO link according to the monitoring information. When there is an abnormality in the IO link, the BMC diagnosis module switches the IO link of the signal processing module through the signal scheduling module, and the signal scheduling module tests the primary IO controller and the secondary IO controller separately. The detachable IO controller signal test module injects test IO link data signals to the front end of each IO controller and monitors the data communication state of the rear end of each IO controller at the same time, locks the IO controller with no output data communication signal at the rear end of the IO link, and turns off the abnormal IO controller through the programmable detachable power module. When there is no abnormality in the IO link, the BMC diagnosis module monitors the server onboard IO link in real time through the signal scheduling module and the signal processing module.

[0053] The BMC diagnosis module switches the IO link of the signal processing module through the signal scheduling module, including:

[0054] The BMC diagnosis module isolates the communication state of the upstream and downstream IO controllers connected in the current IO link and switches it to the signal scheduling module for signal input.

[0055] The signal scheduling module tests the primary IO controller and the secondary IO controller separately, including:

[0056] The signal scheduling module detects the in-place state of the detachable IO controller signal test module, loads the detachable IO controller signal test module driver according to the in-place state, and enables the detachable IO controller signal test module to communicate with the IO controller.

[0057] The server onboard IO link monitoring method can realize each embodiment of the server onboard IO link monitoring system and achieve the same beneficial effects, which will not be described here.

[0058] The preferred embodiments of the application have been described above in detail. It should be understood that modifications and variations to the preferred embodiments could be made by those skilled in the art in light of the teachings above. It is therefore contemplated that the application can encompass other variations and modifications that fall within the scope of the claims.

Claims

1. A server onboard I / O link monitoring device, applied to a server onboard I / O link, wherein the server onboard I / O link includes a Phytium processor, a primary I / O controller, a secondary I / O controller, and I / O devices, characterized in that, The monitoring device comprises a signal processing module, a signal scheduling module, a BMC diagnosis module, a programmable separated power supply module and a separable IO controller signal test module; the Feiteng processor, the primary IO controller, the secondary IO controller and the IO device are connected through the signal processing module; The signal processing module is configured to collect IO link signals through an internal signal collector and feed back to the signal scheduling module in the case of starting the on-board IO link data communication on the server; The signal scheduling module is configured to monitor the IO link signals in real time to obtain monitoring information and synchronously feed back the monitoring information to the BMC diagnosis module in real time; The BMC diagnosis module is configured to diagnose the on-board IO link according to the monitoring information, switch the IO link of the signal processing module through the signal scheduling module in the case of IO link abnormality, and separately test the primary IO controller and the secondary IO controller through the signal scheduling module; the separable IO controller signal test module injects test IO link data signals to the front end of each IO controller and monitors the back end IO link data communication state of each IO controller, locks the IO controller with no output data communication signal in the back end IO link, and closes the abnormal IO controller through the programmable separated power supply module; in the case of no IO link abnormality, the BMC diagnosis module monitors the server on-board IO link in real time through the signal scheduling module and the signal processing module; The signal processing module comprises a first signal processing module, a second signal processing module and a third signal processing module; the Feiteng processor is connected with the primary IO controller through the first signal processing module; the primary IO controller is connected with the secondary IO controller through the second signal processing module; the secondary IO controller is connected with the IO device through the third signal processing module; and the first signal processing module, the second signal processing module and the third signal processing module are connected with the signal scheduling module; The Feiteng processor, the first signal processing module and the primary IO controller form a first IO link segment; the primary IO controller, the second signal processing module and the secondary IO controller form a second IO link segment; and the secondary IO controller, the third signal processing module and the IO device form a third IO link segment.

2. The server onboard IO link monitoring device of claim 1, wherein, The signal scheduling module is a PFX switch which supports 100 channels, 52 ports, 26 virtual switch partitions and 48 non-transparent bridges.

3. The server onboard IO link monitoring device of claim 1, wherein, The signal processing module is an equalizer which comprises a signal collector and a signal switcher; the signal collector is connected with the RX signal channel and the TX signal channel of the signal switcher respectively; the signal switcher is connected with the Feiteng processor, the signal scheduling module and the primary IO controller respectively; and the signal collector is also connected with the signal scheduling module.

4. The server onboard IO link monitoring device of claim 1, wherein, The signal scheduling module comprises a first signal management module, a second signal management module, a signal monitoring module and a signal selector, the first signal management module is connected with the separable IO controller signal test module, the first signal processing module, the signal selector, the signal monitoring module and the second signal management module, the signal monitoring module is further connected with the second signal management module and the BMC diagnosis module, the second signal management module is further connected with the signal selector, the third signal processing module and the separable IO controller signal test module, and the signal selector is further connected with the second signal processing module.

5. The server onboard IO link monitoring device of claim 1, wherein, The monitoring information comprises IO link segment performance, IO link communication state and real-time captured eye diagram.

6. A method for monitoring a server onboard IO link, applied to the server onboard IO link monitoring device in any one of claims 1-5, characterized in that, The method comprises: The signal processing module collects IO link signals through an internal signal collector and feeds back to the signal scheduling module in the case of power-on start of the server onboard IO link data communication; The signal scheduling module obtains monitoring information by real-time monitoring of the IO link signals and synchronizes the monitoring information to the BMC diagnosis module in real time; The BMC diagnosis module performs abnormal diagnosis on the onboard IO link according to the monitoring information, switches the IO link of the signal processing module through the signal scheduling module in the case of IO link abnormality, and the signal scheduling module tests the primary IO controller and the secondary IO controller separately; the separable IO controller signal test module injects test IO link data signals to the front end of each IO controller, monitors the rear end IO link data communication state of each IO controller, locks the IO controller without rear end IO link output data communication signal, and closes the abnormal IO controller through the programmable separable power module; in the case of no IO link abnormality, the BMC diagnosis module performs segmented monitoring on the server onboard IO link in real time through the signal scheduling module and the signal processing module.

7. The server onboard IO link monitoring method of claim 6, wherein, The BMC diagnosis module switches the IO link of the signal processing module through the signal scheduling module, comprising: The BMC diagnosis module isolates the communication state of the upstream and downstream IO controllers connected in the current IO link and switches to the signal scheduling module for signal input.

8. The server onboard IO link monitoring method of claim 6, wherein, The signal scheduling module tests the primary IO controller and the secondary IO controller separately, comprising: The signal scheduling module detects the in-place state of the separable IO controller signal test module, loads the separable IO controller signal test module driver according to the in-place state, and enables the separable IO controller signal test module to communicate with the IO controller.

Images