Slow disk identification method and computing device

Abstract

The slow disk identification method is applied to a server, the server comprises a BMC, an RAID card and a to-be-identified hard disk, the BMC is in communication connection with the RAID card, the RAID card is in communication connection with the to-be-identified hard disk, and the method comprises the steps that the BMC sends a time delay information statistics instruction to the RAID card; based on the time delay information statistical instruction, the RAID card performs multiple read-write operations on the to-be-identified hard disk to obtain time delay information, and the time delay information is used for indicating a read-write time consumption condition of the multiple read-write operations; and the BMC obtains the time delay information from the RAID card and determines whether to identify the hard disk to be identified as the slow disk based on the time delay information, or the RAID card determines whether to identify the hard disk to be identified as the slow disk based on the time delay information. In the application, the BMC and the RAID card realize the slow disk detection and identification of one or more hard disks, and the slow disk identification work does not need to be carried out under an operating system, so that the application scene is wider.

Description

Technical Field

[0001] The present application relates to the field of storage technology, and in particular to a slow disk identification method and computing device. Background Art

[0002] In the field of storage technology, it is crucial to identify slow disks in one or more hard disks for stable business operation. For mechanical hard disks, there are hardware and environmental problems such as head aging, magnetic degradation, bad sectors, fragmentation, vibration, and energy-saving design; for solid-state hard disks, there are problems such as flash media failure, frequent erasing, hard disk overheating, and charge leakage; these problems can easily cause the hard disk's read and write performance to deteriorate and the input / output (IO) response time to become longer, forming the so-called slow disk. Therefore, a slow disk identification technology is needed to identify slow disks in one or more hard disks.

[0003] In the existing slow disk identification methods, the hard disk is identified based on the IOPS (input / output operations per second) information of the operating system; or the hard disk is identified based on the driver response log; or the operating system identifies the hard disk by reading and writing the hard disk. All of these methods require the deployment of software under the operating system to identify the slow disk, and there are certain limitations in the use scenarios. Summary of the invention

[0004] The embodiments of the present application provide a slow disk identification method and a computing device, which realizes slow disk detection and identification of one or more hard disks based on BMC (Baseboard Management Controller) without the need to perform slow disk identification work under the operating system, and has a wider range of usage scenarios.

[0005] In a first aspect, an embodiment of the present application provides a slow disk identification method, which is applied to a server, wherein the server includes a BMC, a RAID card and a hard disk to be identified, the BMC is communicatively connected to the RAID card, and the RAID card is communicatively connected to the hard disk to be identified, the method comprising: the BMC sends a delay information statistics instruction to the RAID card; based on the delay information statistics instruction, the RAID card performs multiple read and write operations on the hard disk to be identified, and obtains delay information, wherein the delay information is used to indicate the read and write time consumption of the multiple read and write operations; the BMC obtains the delay information from the RAID card, and based on the delay information, determines whether to identify the hard disk to be identified as a slow disk, or, based on the delay information, the RAID card determines whether to identify the hard disk to be identified as a slow disk.

[0006] In the technical solution of the embodiment of the present application, the out-of-band management software BMC sends a delay information statistics instruction to the RAID card, and the RAID card performs multiple read and write operations on the hard disk to be identified. The BMC or RAID card identifies whether the hard disk to be identified has a slow disk phenomenon based on the delay information fed back from the multiple read and write operations. In other words, the technical solution of the embodiment of the present application realizes the slow disk detection and identification of one or more hard disks based on the BMC and the RAID card, without the need to perform the slow disk identification work under the operating system. The application scenario is more extensive and can be applied to scenarios such as the operating system does not support the slow disk identification work and the operating system does not exist.

[0007] In a possible implementation, the above-mentioned BMC obtains latency information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: the BMC obtains latency information from the RAID card; based on the latency information, the BMC determines the proportion of timeouts in the total number of multiple read and write operations, the number of timeouts indicating the number of read and write operations corresponding to the latency threshold when the ratio is higher than a preset ratio, the BMC identifies the hard disk to be identified as a slow disk. Alternatively, the above-mentioned RAID card determines whether to identify the hard disk to be identified as a slow disk based on latency information, including: based on the latency information, the RAID card determines the proportion of timeouts in the total number of multiple read and write operations, the number of timeouts indicating the number of read and write operations corresponding to the latency threshold when the ratio is higher than a preset ratio, the RAID card identifies the hard disk to be identified as a slow disk.

[0008] In this implementation, the BMC or RAID card can determine whether a slow disk phenomenon occurs in the hard disk to be identified by the relationship between the timeout times and the preset ratio. The preset times ratio can be preset by the user, making the setting of the slow disk judgment standard more flexible and improving the user experience.

[0009] In a possible implementation, the above-mentioned slow disk includes a general slow disk and a serious slow disk. The above-mentioned BMC obtains delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: BMC obtains delay information from the RAID card; based on the delay information, BMC determines the proportion of the first number of first timeouts in the number of multiple read and write operations, the first timeout number indicates the number of read and write operations corresponding to the time consumption of reading and writing exceeding the first threshold; when the proportion of the first number is higher than the first preset number ratio, BMC identifies the hard disk to be identified as a general slow disk; based on the delay information, BMC determines the proportion of the second number of second timeouts in the number of multiple read and write operations, the second timeout number indicates the number of read and write operations corresponding to the time consumption of writing exceeding the second threshold, the second threshold is greater than the first threshold; when the proportion of the second number is higher than the second preset number ratio, BMC identifies the hard disk to be identified as a serious slow disk. Alternatively, the above-mentioned determination by the RAID card based on the delay information whether to identify the hard disk to be identified as a slow disk includes: based on the delay information, the RAID card determines the proportion of the first number of first timeouts in the number of multiple read and write operations, the first number of timeouts indicating the number of read and write operations corresponding to a first threshold value in which the read and write time exceeds the ratio; when the proportion of the first number is higher than the first preset number ratio, the RAID card identifies the hard disk to be identified as a general slow disk; based on the delay information, the RAID card determines the proportion of the second number of second timeouts in the number of multiple read and write operations, the second number of timeouts indicating the number of read and write operations corresponding to a write time exceeding the second threshold value in which the second threshold value is greater than the first threshold value; when the proportion of the second number is higher than the second preset number ratio, the RAID card identifies the hard disk to be identified as a severely slow disk.

[0010] In this implementation, by setting different second thresholds and first thresholds, the judgment criteria of two types of delay information are determined; for the first timeout number exceeding the first threshold and the second timeout number exceeding the second threshold, the first preset number ratio and the second preset number ratio are set respectively, so as to realize the determination of the general slow disk phenomenon and the serious slow disk phenomenon of the hard disk to be identified. Exemplarily, by using a sliding window method, the delay information of the read and write operations within a period (such as x seconds * m times) is counted. If the count ratio exceeding y1 milliseconds exceeds the threshold of 80%, it is considered that the slow disk phenomenon has occurred; if the count ratio exceeding y2 (y2>y1) milliseconds exceeds the threshold of 70%, it is considered that the serious slow disk phenomenon has occurred. Exemplarily, it is assumed that the statistical data is: the number of times greater than the first threshold (such as 5ms) and less than or equal to the second threshold (such as 10ms) is a; the number of times greater than the second threshold is b, so that repeated statistics can be avoided. By using the sliding window method, the first timeout number is (a+b); the second timeout number is b. Exemplarily, assume that the statistical data is: the number of times c is greater than a first threshold (such as 5ms); the number of times d is greater than a second threshold (such as 10ms), so there is repeated statistics, that is, the number of times c includes the number of times d is greater than the second threshold.

[0011] In a possible implementation, the above-mentioned read and write operations include: the RAID card writes the target data to a preset area of ​​the hard disk to be identified; and the RAID card reads the target data from the preset area.

[0012] In this implementation, the RAID card reads and writes data to the preset area of ​​the hard disk to be identified, thereby realizing the read and write operation of the RAID card on the hard disk to be identified. In other implementations, the normal read and write operation on the service side can also be used as the read and write operation of the hard disk to be identified in the embodiment of the present application.

[0013] In a possible implementation, the above-mentioned BMC obtains latency information from the RAID card, and based on the latency information, determines whether to identify the hard disk to be identified as a slow disk, including: the BMC obtains latency information from the RAID card; based on the latency information, the BMC determines whether the hard disk to be identified has a slow disk phenomenon; when the hard disk to be identified has a slow disk phenomenon, the BMC determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; when there is no communication failure in the communication link from the RAID card to the hard disk to be identified, the BMC identifies the hard disk to be identified as a slow disk. Alternatively, the above-mentioned RAID card determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: based on the latency information, the RAID card determines whether the hard disk to be identified has a slow disk phenomenon; when the hard disk to be identified has a slow disk phenomenon, the RAID card determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; when there is no communication failure in the communication link from the RAID card to the hard disk to be identified, the RAID card identifies the hard disk to be identified as a slow disk.

[0014] In this implementation, a method for troubleshooting the fault location of the slow disk phenomenon is provided. The fault location that causes the slow disk phenomenon may be any location of the link from the RAID card to the hard disk to be identified, so it is necessary to handle the fault in the link from the RAID card to the hard disk to be identified. Exemplarily, the method for troubleshooting the fault in the link can be: by checking the RAID card log and the operating system log, and then determining the location of the fault; or by replacing components, such as replacing the RAID card and replacing cables.

[0015] In a possible implementation, the BMC obtains the latency information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: the BMC obtains the latency information from the RAID card; the BMC processes the latency information through a sliding window method to obtain sliding window statistics; based on the sliding window statistics, the BMC determines whether to identify the hard disk to be identified as a slow disk. Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: the RAID card processes the latency information through a sliding window method to obtain sliding window statistics; based on the sliding window statistics, the RAID card determines whether to identify the hard disk to be identified as a slow disk.

[0016] In this implementation, the sliding window method is a common method with high computational efficiency.

[0017] In a possible implementation, the RAID card performs multiple read and write operations on the hard disk to obtain delay information, including: the RAID card periodically performs multiple read and write operations on the hard disk to obtain delay information according to a first preset period.

[0018] In this implementation, the RAID card periodically performs multiple read and write operations on the hard disk to be identified, so that the BMC or RAID card can periodically perform the slow disk identification task on the hard disk to be identified, and timely discover and process the hard disk with slow disk phenomenon, so as to reduce the business impact and cost loss caused by the slow disk phenomenon.

[0019] In a possible implementation, the above-mentioned latency information is a statistical result of the read / write time consumption of each read / write operation in multiple read / write operations, and is a distribution of the read / write time consumption corresponding to the multiple read / write operations.

[0020] In this implementation, the read and write time information of multiple read and write operations may itself be complex data with a large amount of data, and the delay information is the statistical result of these read and write time-consuming data. The delay information has the characteristics of small data volume, complete information, and easy analysis.

[0021] In a possible implementation, the BMC obtains the delay information from the RAID card, including: the BMC periodically obtains the delay information from the RAID card according to a second preset period.

[0022] In this implementation, the BMC periodically obtains latency information from the RAID card, so that the BMC can periodically perform slow disk identification tasks on the hard disks to be identified, thereby timely discovering and processing hard disks with slow disk phenomena to reduce the business impact and cost losses caused by the slow disk phenomenon.

[0023] In the second aspect, the embodiment of the present application provides a slow disk identification method, which is applied to a server, the server includes a BMC and a hard disk to be identified, the method mainly includes: the BMC performs multiple read and write operations on the hard disk to be identified, and obtains delay information, the delay information is used to indicate the read and write time consumption of multiple read and write operations; based on the delay information, the BMC determines whether to identify the hard disk to be identified as a slow disk. In a possible implementation, the method also includes: when the hard disk to be identified is identified as a slow disk, the BMC determines whether a communication failure occurs in the communication link from the BMC to the hard disk to be identified, so as to handle the communication failure.

[0024] In a third aspect, an embodiment of the present application provides a slow disk identification method, which is applied to a server, the server comprising a BMC, a CPU, a volume management device (VMD) and a hard disk to be identified, the BMC being communicatively connected to the CPU; the CPU being communicatively connected to the volume management device (VMD), and the volume management device (VMD) being communicatively connected to the hard disk to be identified, the method comprising: the BMC sending a delay information statistics instruction to the volume management device (VMD) via the CPU; based on the delay information statistics instruction, the volume management device (VMD) performing multiple read and write operations on the hard disk to be identified, and obtaining delay information, the delay information being used to indicate the read and write time consumption of the multiple read and write operations; the BMC obtaining the delay information from the volume management device (VMD) via the CPU, and based on the delay information, determining whether to identify the hard disk to be identified as a slow disk.

[0025] In a possible implementation, the method further includes: when the hard disk to be identified is identified as a slow disk, the BMC determines whether a communication failure occurs in a communication link from the volume management device (VMD) to the hard disk to be identified, so as to process the communication failure.

[0026] In a fourth aspect, an embodiment of the present application provides a slow disk identification method, which is applied to a server, wherein the server includes a BMC, a CPU, a volume management device (VMD) and a hard disk to be identified, wherein the BMC is communicatively connected to the CPU; the CPU is communicatively connected to the volume management device (VMD), and the volume management device (VMD) is communicatively connected to the hard disk to be identified, and the method includes: the BMC sends a delay information statistics instruction to the volume management device (VMD) via the CPU; based on the delay information statistics instruction, the volume management device (VMD) performs multiple read and write operations on the hard disk to be identified to obtain delay information, and the delay information is used to indicate the read and write time consumption of the multiple read and write operations; based on the delay information, the volume management device (VMD) determines whether to identify the hard disk to be identified as a slow disk.

[0027] In a fifth aspect, an embodiment of the present application provides a slow disk identification method, which is applied to a server, the server comprising a BMC and a hard disk to be identified, the method comprising: the BMC sends a delay information statistics instruction to the hard disk to be identified; based on the delay information statistics instruction, the hard disk to be identified performs multiple read and write operations to obtain delay information, and the delay information is used to indicate the read and write time consumption of multiple read and write operations; the BMC obtains the delay information from the hard disk to be identified, and based on the delay information, determines whether to identify the hard disk to be identified as a slow disk.

[0028] In a sixth aspect, an embodiment of the present application provides a computing device, the computing device comprising: a processor and a memory. The memory stores instruction operations or codes. The processor is configured to communicate with the memory and execute the instruction operations or codes in the memory to perform the method of any one of the first aspect, the second aspect, the third aspect and their implementations.

[0029] In the seventh aspect, an embodiment of the present application provides a computer-readable storage medium, which includes instructions. When the instructions are executed on a computer, the computer executes a method of any one of the above-mentioned first aspect, second aspect, third aspect and their implementation methods.

[0030] In an eighth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed by a computing device, enables the computing device to execute a method of any one of the above-mentioned first aspect, second aspect, third aspect and their implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The following is a brief introduction to the drawings required for use in the embodiments or technical descriptions.

[0032] Figure 1a A schematic diagram of the composition of a computing device provided in an embodiment of the present application;

[0033] Figure 1b A schematic diagram of a communication link between a BMC and a hard disk provided in an embodiment of the present application;

[0034] Figure 2a A schematic diagram of a first communication link provided in an embodiment of the present application;

[0035] Figure 2b A schematic diagram of a second communication link provided in an embodiment of the present application;

[0036] Figure 2c A schematic diagram of a third communication link provided in an embodiment of the present application;

[0037] Figure 2d A schematic diagram of a fourth communication link provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of a flow chart of a first slow disk identification method provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of a sliding window method for processing delay information provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of a flow chart of a second slow disk identification method provided in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of a flow chart of a third slow disk identification method provided in an embodiment of the present application;

[0042] Figure 7 A schematic diagram of a fourth slow disk identification method provided in an embodiment of the present application;

[0043] Figure 8 A schematic diagram of a fifth slow disk identification method provided in an embodiment of the present application;

[0044] Fig. 9 A schematic diagram of the composition of another computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0046] The term "and / or" in this article is a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The symbol " / " in this article indicates that the associated objects are in an or relationship, for example, A / B means A or B.

[0047] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0048] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0049] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0050] In order to facilitate understanding of the solution provided by the embodiment of the present application, some terms involved in the solution are briefly introduced first.

[0051] Redundant Array of Independent Disks (RAID), also known as Redundant Array of Inexpensive Disks, refers to the use of multiple independent disks to form a large disk system, thereby achieving better storage performance and higher reliability than a single disk. RAID includes multiple types of redundant disk arrays such as RAID0, RAID1, RAID3, RAID5, RAID6, RAID10, RAID50 and RAID60. RAID group refers to the disk system composed of multiple independent disks in RAID.

[0052] VMD (Volume Management Device) function is a new technology of Purley platform. VMD is the volume management device of RAID (Redundant Arrays of Independent Disks) of NVMe (Non-Volume Memory Express) hard disk, that is, the controller of NVMe hard disk. VMD can perform RAID management and light signal control on NVMe hard disk connected to VMD. VMD is also called PCIe (Peripheral Component Interconnect Express) controller, which is an integrated controller in the root complex of CPU and PCIe. VMD is integrated into CPU and communicates with NVMe hard disk through PCIe port to realize functions such as hot-swap operation support, related storage management functions and fault tolerance.

[0053] Software RAID (Intel Virtual RAID on CPU, INTELVROC), also known as virtual disk array, software RAID group, soft RAID, software RAID controller, etc., is a simpler RAID solution implemented through Intel Volume Management Device (Intel VMD), which can implement RAID without adding an additional RAID controller card. Soft RAID aggregates several hard disks or partitions of the same size into a large virtual RAID device (or RAID0, or RAID1, or RAID5, or RAID10...) according to different usage directions. If the capacity of each hard disk or partition is inconsistent, the hard disk or partition with the smallest capacity is used as the basis. The members of soft RAID are the entire hard disk or partition. VROC is an enterprise-level RAID solution that can release the performance of NVMe solid-state drives. It is enabled by the Intel VMD function in the Intel Xeon Scalable Processor. The VMD function is an integrated controller in the CPU PCIe root complex. Intel VROC has these advantages while avoiding the complexity, high cost and high power consumption of the host bus adapter (HBA) card of the traditional hardware RAID card placed between the drive and the CPU.

[0054] The embodiment of the present application provides a slow disk identification method and related devices. The BMC performs multiple read and write operations on a hard disk to be identified through certain communication links, and identifies whether the hard disk has a slow disk phenomenon based on the IO delay fed back by the multiple read and write operations.

[0055] That is to say, the embodiment of the present application solves the problem of BMC detecting hard disks. The main working steps of BMC are: realizing communication connection with the hard disk to be identified, reading and writing the space data of the hard disk to be identified; based on the delay information when reading and writing data, BMC realizes the slow disk detection function of the hard disk to be identified.

[0056] See also Figure 1a , Figure 1a A schematic diagram of the composition of a computing device provided in an embodiment of the present application. Figure 1a As shown, the embodiment of the present application provides a computing device 100, which mainly includes: BMC 110, processing module 120 and storage module 130. Storage module 130 may include multiple hard disks. For example, Figure 1a As shown, the storage module 130 includes hard disk 1, hard disk 2, ..., hard disk n, where n is an integer greater than or equal to 1.

[0057] The computing device 100 may be a server or other types of computing devices, which is not limited in the embodiments of the present application. Among them, the server, also known as a server, is a device that provides computing services. Since the server needs to respond to service requests and process them to provide reliable services, generally speaking, the server should have the ability to undertake services and guarantee services. The server needs to have strong processing power, high stability, high reliability, high security, scalability and manageability. In the embodiments of the present application, the server may be an x86 server, which is also called a complex instruction set computer (CISC) architecture server, that is, a personal computer (PC) server commonly referred to as a PC architecture, which is a server using an Intel (Intel) or other x86 instruction set compatible processor chip and a Windows operating system.

[0058] The server may have relatively large differences due to different configurations or performances, and may include at least one central processing unit (CPU) (e.g., at least one processor) and memory, and at least one storage medium (e.g., at least one mass storage device) for storing applications or data. Among them, the memory and the storage medium may be short-term storage or persistent storage. The program stored in the storage medium may include at least one module, and each module may include a series of instruction operations on the server. Furthermore, the central processing unit may be configured to communicate with the storage medium and execute a series of instruction operations in the storage medium on the server. The server may also include at least one power supply, at least one wired or wireless network interface, at least one input and output interface, and / or at least one operating system, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, NetWare, etc. In some feasible implementations, the server may also be a cloud server, which is not limited here.

[0059] BMC (Baseboard Management Controller) 110 is a dedicated controller for monitoring and managing computing devices 100 (such as servers), which can be regarded as a small operating system. For example, it can monitor the temperature, voltage and other information of the computing device 100, and make corresponding adjustments to ensure that the computing device 100 is in a normal operating state. When the computing device 100 (such as a server) is in an abnormal state, the computing device 100 can also be restarted by resetting. BMC110 can also record various hardware information and logs to prompt users who use the computing device 100 (such as a server) and locate faults. It should be noted that BMC110 is an independent device. It does not rely on other hardware (such as a processor or memory, etc.) in the computing device 100, nor does it rely on the OS of the server, but BMC110 can interact with the OS. Generally, BMC110 includes a BMC storage medium and a BMC processor, and interacts with the OS through the BMC processor, which can play a better management role. BMC storage medium is used to store instructions required by BMC110 when running and data generated when running. The BMC storage medium may be a flash memory (flash), a random access memory (RAM), or a read-only memory (ROM), etc., which is not limited here.

[0060] It should be noted that, in other embodiments, BMC110 has different names in different computing devices. For example, the BMC of Huawei servers and Super Fusion servers is called i BMC, the BMC of HPE servers is called i LO, and the BMC of DELL servers is called iDRAC.

[0061] The processing module 120 is the control center of the computing device 100. It uses various interfaces and lines to connect various parts of the computing device 100, and executes various functions and processes data of the computing device 100 by running or executing software programs and / or modules stored in the storage module 130, and calling data stored in the storage module 130, so as to monitor the computing device 100 as a whole. Optionally, the processing module 120 may include one or more processing units; preferably, the processing module 120 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processing module 120. In some feasible implementations, the processing module 120 may be a central processing unit (CPU) or other processors, which are not limited here.

[0062] Hard disk is the actual physical carrier for storing data. The more common storage devices on the market include mechanical hard disk (HDD) and solid state disk (SSD). Among them, all the disks in HDD are mounted on a rotating shaft, and each disk is parallel to each other. There is a magnetic head on the storage surface of each disk. The distance between the magnetic head and the disk is 0.1 micrometer (μm) ~ 0.5μm, and the higher level has reached 0.005μm ~ 0.01μm. All magnetic heads are connected to a magnetic head controller, which is responsible for the movement of each magnetic head. The read and write speed of mechanical hard disk depends on the speed of the motor, because it needs to rely on the motor to drive the disk to rotate at high speed so that the magnetic head can find the specified position for reading and writing. SSD, also known as solid-state drive, is a hard disk made of solid-state electronic storage chip array.

[0063] See also Figure 1b , Figure 1b The schematic diagram of the communication link between the BMC and the hard disk provided in the embodiment of the present application. Figure 1b As shown, in different computing devices 100 , the BMC 110 may have different communication links.

[0064] Alternatively, if Figure 1bAs shown, in some computing devices 100, BMC 110 is connected to RAID card 121 for communication, and RAID card 121 is connected to one or more hard disks for communication. This communication link is recorded as a first communication link. RAID card 121 is used to manage multiple hard disks.

[0065] Alternatively, if Figure 1b As shown, in some computing devices 100, BMC 110 is connected to CPU 122 for communication, CPU 122 is connected to VMD 123 for communication, and VMD 123 is connected to one or more hard disks for communication. This communication link is recorded as a second communication link. CPU 122 is generally deployed with a soft RAID, which is used to manage multiple hard disks.

[0066] Alternatively, if Figure 1b As shown, in some computing devices 100, BMC 110 is connected to CPU 122 for communication, and CPU 122 is connected to one or more hard disks for communication. This communication link is recorded as a third communication link. Among them, CPU 122 is directly connected to one or more hard disks for communication and manages the one or more hard disks.

[0067] Alternatively, if Figure 1b As shown, in some computing devices 100, BMC 110 is connected to CPU 122 for communication, CPU 122 is connected to RAID card 121 for communication, and RAID card 121 is connected to one or more hard disks for communication. This communication link is recorded as a fourth communication link. Compared with the first communication link, the RAID card 121 communicates with BMC 110 through CPU 122, for example, the RAID card 121 can send relevant messages or instructions to CPU 122.

[0068] See also Figure 2a , Figure 2a Schematic diagram of a first communication link provided in an embodiment of the present application. Figure 2a As shown, in the computing device 100, an example of the processing module 120 is a RAID card 121. Optionally, the processing module 120 also includes a hard disk backplane. An example of the storage module 130 is a RAID group. Figure 2a As shown, the RAID group includes hard disk 1, hard disk 2, ..., hard disk n, where n is an integer greater than or equal to 1.

[0069] See also Figure 2b , Figure 2b Schematic diagram of a second communication link provided in an embodiment of the present application. Figure 2bAs shown, in the computing device 100, an example of the processing module 120 is CPU 122 and VMD 123. Optionally, through the communication connection between VMD 123 and the RAID group, CPU 122 is generally deployed with software RAID to implement RAID technology. An example of the storage module 130 is a RAID group. For example, Figure 2b As shown, the RAID group may include hard disk 1, hard disk 2, ..., hard disk n, where n is an integer greater than or equal to 1. The BMC 110 may communicate with the VMD 123 through the CPU 122, and the VMD 123 is an integrated controller that can read and write data to the hard disks in the RAID group.

[0070] See also Figure 2c , Figure 2c Schematic diagram of the third communication link provided in the embodiment of the present application. Figure 2c As shown, in the computing device 100, an example of the processing module 120 is a CPU 122; an example of the storage module 130 is one or more hard disks. Figure 2b As shown, the storage module 130 can be hard disk 1, hard disk 2, ..., hard disk n, where n is an integer greater than or equal to 1.

[0071] See also Figure 2d , Figure 2d Schematic diagram of a fourth communication link provided in an embodiment of the present application. Figure 2d As shown, in the computing device 100, an example of the processing module 120 is a CPU 122 and a RAID card 121. Optionally, the processing module 120 also includes a hard disk backplane. An example of the storage module 130 is a RAID group. Figure 2a As shown, the RAID group includes hard disk 1, hard disk 2, ..., hard disk n, where n is an integer greater than or equal to 1. Compared with the first communication link, the RAID card 121 implements communication with the BMC 110 through the CPU 122, for example, the RAID card 121 can send relevant messages or instructions to the CPU 122.

[0072] Further, Figure 2a and Figure 2d It can be understood as a hardware architecture, that is, the RAID card can communicate directly with the BMC; the RAID card can also communicate indirectly with the BMC. For example, the RAID card can first send a message to the CPU, and the CPU then sends the message to the BMC, thereby realizing indirect communication between the RAID card and the BMC.

[0073] Furthermore, combined with Figure 2a to Figure 2dThe four communication links from the BMC to the hard disk to be identified can be understood as three different hardware architectures or three different application scenarios.

[0074] Generally, the hard disk backplane is made of a printed circuit board (PCB), which is used to connect the processing module 120, the RAID card 121 and the storage module 130, and then realize data transmission among the processing module 120, the RAID card 121 and the storage module 130 by wiring. The processing module 120 and the RAID card 121 can be connected and communicated through a high-speed serial computer expansion bus standard (peripheral component interconnect express, PCI E).

[0075] Alternatively, the hard disk can be a non-volatile random access memory (NVRAM). The hard disk stores a processor and an operation instruction, an executable module or a data structure, or a subset thereof, or an extended set thereof, wherein the operation instruction may include various operation instructions for implementing various operations. It should be noted that the computing device 100 can have multiple hard disks built-in, such as 8 hard disks, which are not limited here. In some feasible implementations, each hard disk can be a hard disk of different types. Exemplarily, the hard disk can be a serial attached SCSI (SAS) hard disk, a serial port (SATA) hard disk or a non-volatile memory host controller interface specification (non-volatile memory express) hard disk.

[0076] Generally, BMC110 communicates with RAID card 121 through I2C interface. RAID card 121 manages multiple hard disks through SAS cables, obtains working parameters of hard disks and logical disks, and feeds back the working parameters of hard disks and logical disks to BMC110 in real time, so that BMC110 can monitor the working parameters of each hard disk and logical disk, and generate alarms or log records when necessary.

[0077] Furthermore, the RAID card 121 includes a logic controller and a RAID chip. The logic controller is an electronic component or an electronic assembly, and is a digital circuit chip. Common logic controllers are, for example, complex programmable logic devices (CPLD) or field programmable gate arrays (FPGA). The BMC 110 communicates with the logic controller and the RAID chip via an I2C interface, and the logic controller communicates with the RAID chip via a GPIO interface.

[0078] Generally, the computing device 100 is usually equipped with a RAID card 121, which can manage the hard disks in the computing device 100. In some possible implementations, the communication between the RAID card 121 and the storage module 130 can be RAID pass-through, just a bunch of disks (JBOD) or group RAID, which is not limited here.

[0079] Generally, BMC110 can generate an operation control signal according to the received instruction operation code and timing signal to complete the operation of obtaining and executing instructions. BMC110 may include one or more interfaces. The interface may include an inter-integrated circuit (I2C) interface, an improved integrated circuit (Improved Inter Integrated Circuit, I3C), a serial general purpose input / output (Serial Genera l Purpose I nput / output, SGPIO) interface, etc.

[0080] The BMC 110 is independent of the central processing unit CPU 122 of the computing device 100 and can communicate with the CPU 122 through the out-of-band management interface of the computing device 100, and can remotely maintain and manage the computing device 100 through a dedicated data channel. The BMC 110 controls the CPU 122 to identify slow disks of one or more hard disks connected to the CPU 122.

[0081] In some possible implementations, the computing device 100 may also include a display module and an input module (eg, a keyboard, a mouse, a touch screen, etc., which are not limited here).

[0082] The display module may be used to display information input by a user or information provided to a user and various interfaces of the computing device 100. The display module may include a display panel, and optionally, the display panel may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. Further, a touch panel may cover the display panel, and when the touch panel detects a touch operation on or near it, it is transmitted to the processing module 120 to determine the type of the touch event, and then the processing module 120 provides a corresponding visual output on the display panel according to the type of the touch event.

[0083] The input module can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the computing device 100. Specifically, the input module may include a touch panel and other input devices. The touch panel, also known as a touch screen, can collect the user's touch operations on or near it (such as the user's operation on or near the touch panel using any suitable object or accessory such as a finger, stylus, etc.), and drive the corresponding connection device according to a pre-set program. Optionally, the touch panel may include a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processing module 120, and can receive and execute commands sent by the processing module 120. In addition, the touch panel can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel, the input module may also include other input devices. Specifically, other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0084] See also Figure 3 , Figure 3 The flowchart of the first slow disk identification method provided in the embodiment of the present application is as follows. Figure 3 As shown, a slow disk identification method provided in an embodiment of the present application is applied to a server, the server including a BMC and a hard disk to be identified, and the method mainly includes the following steps:

[0085] Step S310: The BMC performs multiple read and write operations on the hard disk to be identified to obtain delay information. The delay information is used to indicate the time consumption of the multiple read and write operations.

[0086] Step S320: Based on the delay information, the BMC determines whether the hard disk to be identified is a slow disk.

[0087] In the technical solution of the embodiment of the present application, the out-of-band management software BMC is used to perform multiple read and write operations on the hard disk to be identified, and the delay information fed back from the multiple read and write operations is used to identify whether the hard disk to be identified has a slow disk phenomenon. In other words, the technical solution of the embodiment of the present application realizes the slow disk detection and identification of one or more hard disks based on the BMC, without the need to perform slow disk identification work under the operating system, and has a wider range of application scenarios.

[0088] It should be noted that if Figure 3 The slow disk identification method shown can be applied to any communication link that can realize communication between the BMC and the hard disk to be identified, such as the first communication link, the second communication link, the third communication link, and the fourth communication link mentioned above. This embodiment of the application does not limit this.

[0089] In a possible implementation, based on latency information, the BMC determines the proportion of timeouts in the total number of read and write operations, where the timeouts represent the number of read and write operations corresponding to a latency threshold whose read and write time exceeds the latency threshold. When the proportion is higher than a preset ratio, the BMC identifies the hard disk to be identified as a slow disk.

[0090] For example, by means of a sliding window, the latency information of read and write operations within a period (such as x seconds * m times) is counted, and if the count ratio exceeding y1 milliseconds exceeds a threshold of 80%, it is considered that a slow disk phenomenon has occurred. The preset number ratio can be preset by the user, making the setting of the slow disk judgment standard more flexible.

[0091] In one possible implementation, the above-mentioned slow disk includes a general slow disk and a severe slow disk. Therefore, based on the latency information, the BMC can determine the proportion of the first number of first timeouts in the number of multiple read and write operations, and the first number of timeouts indicates the number of read and write operations corresponding to the first threshold value when the first number accounts for a higher proportion than the first preset number, the BMC identifies the hard disk to be identified as a general slow disk. Based on the latency information, the BMC determines the proportion of the second number of second timeouts in the number of multiple read and write operations, and the second number of timeouts indicates the number of read and write operations corresponding to the write time exceeding the second threshold value, and the second threshold value is greater than the first threshold value; when the second number accounts for a higher proportion than the second preset number, the BMC identifies the hard disk to be identified as a severe slow disk.

[0092] In this implementation, the general slow disk phenomenon and the severe slow disk phenomenon of the hard disk to be identified are determined. Exemplarily, the delay information of the read and write operations within a period (such as x seconds * m times) is counted in a sliding window manner. If the count ratio exceeding y1 milliseconds exceeds a threshold of 80%, it is considered that the slow disk phenomenon has occurred; if the count ratio exceeding y2 (y2>y1) milliseconds exceeds a threshold of 70%, it is considered that the severe slow disk phenomenon has occurred.

[0093] In a possible implementation, the above-mentioned read and write operations include: writing target data to a preset area of ​​the hard disk to be identified; and reading target data from the preset area.

[0094] In this implementation, the read and write operations on the hard disk to be identified are performed by reading and writing data to the preset area of ​​the hard disk to be identified. In other implementations, normal read and write operations on the service side can also be used as the read and write operations on the hard disk to be identified in the embodiment of the present application.

[0095] In a possible implementation, the above-mentioned BMC determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: based on the latency information, the BMC determines whether the hard disk to be identified has a slow disk phenomenon; when the hard disk to be identified has a slow disk phenomenon, the BMC determines whether a communication failure occurs in the communication link between the BMC and the hard disk to be identified; when no communication failure occurs in the communication link between the BMC and the hard disk to be identified, the BMC identifies the hard disk to be identified as a slow disk.

[0096] Furthermore, based on the delay information, it is determined whether the hard disk to be identified is a slow disk, and the basis for the determination is whether the delay in reading and writing data is large. There are two reasons for the large delay: one is that there is a fault in the communication link, which causes communication obstruction, and then causes a large delay in reading and writing data; the other is that there is a fault in the hard disk itself (that is, the hard disk to be identified is a slow disk), resulting in a large delay in reading and writing data. Therefore, the BMC needs to first determine whether there is a fault in the communication link from the BMC to the hard disk to be identified, and then determine whether the hard disk to be identified is a slow disk. Exemplarily, the method of troubleshooting the fault in the communication link can be: by checking the BMC log, RAID card log and operating system log, and then determining the location of the fault; it can also be checked by replacing components, such as replacing the RAID card and replacing cables.

[0097] That is to say, the fault location that causes the slow disk phenomenon may be any location of the link from the BMC to the hard disk to be identified, so it is necessary to further troubleshoot the fault in the link from the BMC to the hard disk to be identified to ultimately determine whether the hard disk to be identified has a hard disk fault.

[0098] In other implementations, the BMC also directly reads the hard disk log to determine whether the hard disk to be identified is a slow disk, and then performs hard disk marking, hard disk replacement, etc. Among them, since the hard disk log is the delay information of the read and write operations recorded by the hard disk to be identified, and there is no communication link, there is no need to further check whether the communication link has a fault, and it can directly determine whether the hard disk to be identified is a slow disk based on the hard disk log.

[0099] In the embodiment of the present application, the BMC identifies the slow disk problem of the BMC to hard disk link, and the detection can be achieved without deploying software through the operating system. The usage scenarios are wider and can be applicable to scenarios such as the operating system does not support slow disk identification work and the operating system does not exist.

[0100] In the embodiment of the present application, the slow disk phenomenon refers to the phenomenon that the reading and writing time between the execution subject of the read and write operation and the hard disk to be identified is long. It is necessary to further eliminate the fault in the link from the execution subject to the hard disk to be identified. Exemplarily, when the execution subject of the read and write operation is the hard disk to be identified itself, there is no problem of link failure. Therefore, when it is determined that the hard disk to be identified is a slow disk, it can be directly determined that the hard disk to be identified is a faulty disk, that is, a slow disk failure has occurred.

[0101] In a possible implementation, the above-mentioned latency information is a statistical result of the read / write time consumption of each read / write operation in multiple read / write operations, and is a distribution of the read / write time consumption corresponding to the multiple read / write operations.

[0102] Exemplarily, an example of delay information is: the number of times less than or equal to 5ms is: 10000; the number of times greater than 5ms and less than or equal to 100ms is: 5000; the number of times greater than 100ms and less than or equal to 200ms is: 500; the number of times greater than 200ms or equal to 100ms is: 50; the number of times greater than 240ms is: 0.

[0103] Furthermore, the read and write time information of multiple read and write operations may itself be complex data with a large amount of data, and the delay information is the statistical result of these read and write time-consuming data. Therefore, the delay information has the characteristics of small data volume, complete information, and easy analysis.

[0104] In a possible implementation, the time delay information is processed by a sliding window method to obtain sliding window statistical results; and according to the sliding window statistical results, it is determined whether to identify the hard disk to be identified as a slow disk. The sliding window method is a common method with high computational efficiency.

[0105] See also Figure 4 , Figure 4 A schematic diagram of a sliding window method for processing delay information provided in an embodiment of the present application. Figure 4As shown, illustratively, the purpose of the sliding window method is to find the number of timeouts when the read and write time exceeds 100ms. The sliding window method generally includes the following steps: First, in the initial state, the left side (left) of the window and the right side (right) of the window are both 0; the index closed interval [left, right] is called a "window". Secondly, the right pointer is continuously increased to expand the range of the window [left, right] until the maximum value of the read and write time (for example, 240ms) appears in the window. Thirdly, stop increasing the right value, and continuously increase the value of the left pointer to reduce the window [left, right] until the minimum value (for example, 100ms) below the read and write time no longer appears in the window; finally, the number of times in the window [left, right] is 250, that is, the number of timeouts when the read and write time exceeds 100ms is 250. Exemplarily, the number of times in the window [left, right] can be calculated by a counter.

[0106] Generally speaking, the sliding window method, also called the ruler method, can be used to solve some problems of finding the properties (length, etc.) of continuous intervals that meet certain conditions. Since the intervals are continuous, when the intervals change, the search space can be pruned using the old calculation results, thus reducing repeated calculations and time complexity.

[0107] In a possible implementation, the BMC periodically performs multiple read and write operations on the hard disk to be identified according to a first preset period to obtain delay information.

[0108] In this implementation, the BMC periodically performs multiple read and write operations on the hard disk to be identified, thereby periodically performing the slow disk identification task on the hard disk to be identified, thereby timely discovering and processing the hard disk with slow disk phenomenon, so as to reduce the business impact and cost loss caused by the slow disk phenomenon.

[0109] Exemplarily, the BMC periodically traverses one or more hard disks to perform read and write tasks, thereby realizing that the BMC periodically performs slow disk identification tasks on one or more hard disks. Specifically, when the BMC performs an identification task on a hard disk to be identified, the BMC can write a piece of random data to the reserved area of ​​the hard disk to be identified every x seconds through a communication link such as a CPU or a RAID card; then read the data to count the read and write time (i.e., delay information); in a sliding window manner, count the read and write time within a period (such as x seconds * m times), if the count ratio of the read and write time exceeding y1 milliseconds exceeds a threshold of 80%, it is considered that a slow disk phenomenon has occurred, and if the count ratio of the read and write time exceeding y2 (y2>y1) milliseconds exceeds a threshold of 70%, it is considered that a serious slow disk phenomenon has occurred.

[0110] See also Figure 5 , Figure 5 The flowchart of the second slow disk identification method provided in the embodiment of the present application is as follows. Figure 5 As shown, a slow disk identification method provided in an embodiment of the present application is applied to a server, the server includes a BMC, a RAID card and a hard disk to be identified, the BMC is connected to the RAID card for communication, the RAID card is connected to the hard disk to be identified for communication, and the communication link refers to the first communication link mentioned above. Figure 5 As shown, the method mainly includes the following steps:

[0111] Step S510: The BMC sends a delay information statistics instruction to the RAID card.

[0112] Step S520: In response to the delay information statistics instruction, the RAID card performs multiple read and write operations on the hard disk to be identified to obtain delay information, wherein the delay information is used to indicate the read and write time consumption of the multiple read and write operations.

[0113] Step S530: The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, or the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information.

[0114] In the technical solution of the embodiment of the present application, the out-of-band management software BMC sends a delay information statistics instruction to the RAID card, and the RAID card performs multiple read and write operations on the hard disk to be identified. The BMC or RAID card identifies whether the hard disk to be identified has a slow disk phenomenon based on the delay information fed back from the multiple read and write operations. In other words, the technical solution of the embodiment of the present application realizes the slow disk detection and identification of one or more hard disks based on the BMC and the RAID card, without the need to perform the slow disk identification work under the operating system. The application scenario is more extensive and can be applied to scenarios such as the operating system does not support the slow disk identification work and the operating system does not exist.

[0115] In a possible implementation, the BMC obtains latency information from the RAID card; based on the latency information, the BMC determines the ratio of the number of timeouts in the total number of multiple read and write operations, where the number of timeouts indicates the number of read and write operations corresponding to the latency threshold when the read and write time exceeds the latency threshold; when the ratio is higher than a preset ratio, the BMC identifies the hard disk to be identified as a slow disk. Alternatively, based on the latency information, the RAID card determines the ratio of the number of timeouts in the total number of multiple read and write operations, where the number of timeouts indicates the number of read and write operations corresponding to the latency threshold when the read and write time exceeds the latency threshold; when the ratio is higher than a preset ratio, the RAID card identifies the hard disk to be identified as a slow disk.

[0116] In this implementation, the BMC or RAID card can determine whether a slow disk phenomenon occurs in the hard disk to be identified by the relationship between the timeout times and the preset ratio. The preset times ratio can be preset by the user, making the setting of the slow disk judgment standard more flexible and improving the user experience.

[0117] In a possible implementation, the above-mentioned slow disk includes a general slow disk and a serious slow disk. The above-mentioned BMC obtains delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: BMC obtains delay information from the RAID card; based on the delay information, BMC determines the proportion of the first number of first timeouts in the number of multiple read and write operations, the first timeout number indicates the number of read and write operations corresponding to the time consumption of reading and writing exceeding the first threshold; when the proportion of the first number is higher than the first preset number ratio, BMC identifies the hard disk to be identified as a general slow disk; based on the delay information, BMC determines the proportion of the second number of second timeouts in the number of multiple read and write operations, the second timeout number indicates the number of read and write operations corresponding to the time consumption of writing exceeding the second threshold, the second threshold is greater than the first threshold; when the proportion of the second number is higher than the second preset number ratio, BMC identifies the hard disk to be identified as a serious slow disk. Alternatively, the above-mentioned determination by the RAID card based on the delay information whether to identify the hard disk to be identified as a slow disk includes: based on the delay information, the RAID card determines the proportion of the first number of first timeouts in the number of multiple read and write operations, the first number of timeouts indicating the number of read and write operations corresponding to a first threshold value in which the read and write time exceeds the ratio; when the proportion of the first number is higher than the first preset number ratio, the RAID card identifies the hard disk to be identified as a general slow disk; based on the delay information, the RAID card determines the proportion of the second number of second timeouts in the number of multiple read and write operations, the second number of timeouts indicating the number of read and write operations corresponding to a write time exceeding the second threshold value in which the second threshold value is greater than the first threshold value; when the proportion of the second number is higher than the second preset number ratio, the RAID card identifies the hard disk to be identified as a severely slow disk.

[0118] In this implementation, by setting different second thresholds and first thresholds, the judgment criteria for the two types of delay information are determined; for the first timeout number exceeding the first threshold and the second timeout number exceeding the second threshold, the first preset number ratio and the second preset number ratio are set respectively, thereby realizing the determination of the general slow disk phenomenon and the severe slow disk phenomenon of the hard disk to be identified.

[0119] Exemplarily, through a sliding window method, the delay information of read and write operations within a period (such as x seconds * m times) is counted. If the count ratio exceeding y1 milliseconds exceeds the threshold of 80%, it is considered that a slow disk phenomenon has occurred; if the count ratio exceeding y2 (y2>y1) milliseconds exceeds the threshold of 70%, it is considered that a serious slow disk phenomenon has occurred. Exemplarily, it is assumed that the statistical data is: the number of times greater than the first threshold (such as 5ms) and less than or equal to the second threshold (such as 10ms) is a; the number of times greater than the second threshold is b, so that repeated statistics can be avoided. Through the sliding window method, the first timeout number is (a+b); the second timeout number is b. Exemplarily, it is assumed that the statistical data is: the number of times greater than the first threshold (such as 5ms) is c; the number of times greater than the second threshold (such as 10ms) is d, so there is repeated statistics, that is, the number of times c includes the number of times greater than the second threshold d.

[0120] In a possible implementation, the above-mentioned read and write operations include: the RAID card writes the target data to a preset area of ​​the hard disk to be identified; and the RAID card reads the target data from the preset area.

[0121] In this implementation, the RAID card reads and writes data to the preset area of ​​the hard disk to be identified, thereby realizing the read and write operation of the RAID card on the hard disk to be identified. In other implementations, the normal read and write operation on the service side can also be used as the read and write operation of the hard disk to be identified in the embodiment of the present application.

[0122] In a possible implementation, the above-mentioned BMC obtains latency information from the RAID card, and based on the latency information, determines whether to identify the hard disk to be identified as a slow disk, including: the BMC obtains latency information from the RAID card; based on the latency information, the BMC determines whether the hard disk to be identified has a slow disk phenomenon; when the hard disk to be identified has a slow disk phenomenon, the BMC determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; when there is no communication failure in the communication link from the RAID card to the hard disk to be identified, the BMC identifies the hard disk to be identified as a slow disk. Alternatively, the above-mentioned RAID card determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: based on the latency information, the RAID card determines whether the hard disk to be identified has a slow disk phenomenon; when the hard disk to be identified has a slow disk phenomenon, the RAID card determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; when there is no communication failure in the communication link from the RAID card to the hard disk to be identified, the RAID card identifies the hard disk to be identified as a slow disk.

[0123] In this implementation, a method for troubleshooting the fault location of the slow disk phenomenon is provided. The fault location that causes the slow disk phenomenon may be any location of the link from the RAID card to the hard disk to be identified, so it is necessary to handle the fault in the link from the RAID card to the hard disk to be identified. Exemplarily, the method for troubleshooting the fault in the link can be: by checking the RAID card log and the operating system log, and then determining the location of the fault; or by replacing components, such as replacing the RAID card and replacing cables.

[0124] In a possible implementation, the BMC obtains the latency information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: the BMC obtains the latency information from the RAID card; the BMC processes the latency information through a sliding window method to obtain sliding window statistics; based on the sliding window statistics, the BMC determines whether to identify the hard disk to be identified as a slow disk. Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the latency information, including: the RAID card processes the latency information through a sliding window method to obtain sliding window statistics; based on the sliding window statistics, the RAID card determines whether to identify the hard disk to be identified as a slow disk.

[0125] In this implementation, the sliding window method is a common method with high computational efficiency.

[0126] In a possible implementation, the RAID card performs multiple read and write operations on the hard disk to obtain delay information, including: the RAID card periodically performs multiple read and write operations on the hard disk to obtain delay information according to a first preset period.

[0127] In this implementation, the RAID card periodically performs multiple read and write operations on the hard disk to be identified, so that the BMC or RAID card can periodically perform the slow disk identification task on the hard disk to be identified, and timely discover and process the hard disk with slow disk phenomenon, so as to reduce the business impact and cost loss caused by the slow disk phenomenon.

[0128] In a possible implementation, the above-mentioned latency information is a statistical result of the read / write time consumption of each read / write operation in multiple read / write operations, and is a distribution of the read / write time consumption corresponding to the multiple read / write operations.

[0129] In this implementation, the read and write time information of multiple read and write operations may itself be complex data with a large amount of data, and the delay information is the statistical result of these read and write time-consuming data. The delay information has the characteristics of small data volume, complete information, and easy analysis.

[0130] In a possible implementation, the BMC obtains the delay information from the RAID card, including: the BMC periodically obtains the delay information from the RAID card according to a second preset period.

[0131] In this implementation, the BMC periodically obtains latency information from the RAID card, so that the BMC can periodically perform slow disk identification tasks on the hard disks to be identified, thereby timely discovering and processing hard disks with slow disk phenomena to reduce the business impact and cost losses caused by the slow disk phenomenon.

[0132] It should be noted that if Figure 5 For some possible implementations of the slow disk identification method shown and their technical effects, see Figure 3 Some possible implementation methods and technical effects of the slow disk identification method described in the description will not be repeated here. Figure 3 Some possible implementations of the slow disk identification method and their related descriptions are provided for example Figure 5 The slow disk identification method shown is implemented by the corresponding technical solution.

[0133] Alternatively, if Figure 5 The slow disk identification method shown in FIG. 1 can also be used for the fourth communication link mentioned above. The technical implementation of this technical solution can be referred to in Figure 5 The relevant description will not be repeated here.

[0134] It should be noted that if Figure 5 In the slow disk identification method shown in the figure, the BMC needs to send a delay information statistics instruction and receive delay information, and obtain the result of slow disk identification. The RAID card can automatically complete multiple read and write operations on the hard disk to be identified to obtain delay information. This method can reduce the workload of the BMC and improve the work efficiency of the BMC.

[0135] Exemplarily, the BMC sends a latency information statistics instruction of the hard disk to be identified to the RAID card, and the RAID card performs the statistical task of reading and writing data time (i.e. latency information). For example, the RAID card writes a piece of random data to the reserved area of ​​the hard disk through the CPU or the RAID card itself every x seconds; then reads the data and counts the reading and writing time; the BMC periodically obtains latency information from the RAID card, and the BMC uses a sliding window method. If the proportion of counts exceeding 100ms exceeds a threshold of 80%, it is considered that a slow disk phenomenon has occurred, and if the proportion of counts exceeding 200 milliseconds exceeds a threshold of 70%, it is considered that a serious slow disk phenomenon has occurred. Alternatively, the RAID card uses a sliding window method. If the proportion of counts exceeding 100ms exceeds a threshold of 80%, it is considered that a slow disk phenomenon has occurred, and if the proportion of counts exceeding 200 milliseconds exceeds a threshold of 70%, it is considered that a serious slow disk phenomenon has occurred.

[0136] It should be noted that if Figure 6 In the slow disk identification method shown, the BMC sends a delay information statistics instruction, and the RAID card can complete multiple read and write operations on the hard disk to be identified, obtain delay information, and determine whether to identify the hard disk to be identified as a slow disk based on the delay information. This method can reduce the workload of the BMC and improve the work efficiency of the BMC.

[0137] Furthermore, through this example, the BMC can identify the slow disk problem from the RAID card to the hard disk link, and the specific fault location needs to be checked. In other words, through this example, the BMC can identify the slow disk problem from the RAID card to the hard disk link, and can implement slow disk detection without deploying software through the operating system.

[0138] See also Figure 6 , Figure 6 The flowchart of the third slow disk identification method provided in the embodiment of the present application is as follows. Figure 6 As shown, an embodiment of the present application provides a slow disk identification method, which is applied to a server, the server includes a BMC, a CPU, a volume management device (VMD) and a hard disk to be identified, the BMC is communicatively connected to the CPU; the CPU is communicatively connected to the volume management device (VMD), and the volume management device (VMD) is communicatively connected to the hard disk to be identified, and the method includes:

[0139] Step S610: The BMC sends a latency information statistics instruction to the volume management device (VMD) via the CPU.

[0140] Step S620: In response to the latency information statistics instruction, the volume management device (VMD) performs multiple read and write operations on the hard disk to be identified to obtain latency information. The latency information is used to indicate the time consumption of multiple read and write operations.

[0141] Step S630: The BMC obtains latency information from the volume management device (VMD) via the CPU, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information.

[0142] In a possible implementation, the method further includes: when the hard disk to be identified is identified as a slow disk, the BMC determines whether a communication failure occurs in a communication link from the volume management device (VMD) to the hard disk to be identified, so as to process the communication failure.

[0143] It should be noted that if Figure 6The slow disk identification method shown is applicable to computing devices that use a volume management device (VMD). In a computing device that uses a volume management device (VMD), a software RAID is set on the CPU. The software RAID can implement the CPU's management of the RAID group disks in software; the volume management device (VMD) is an integrated controller that can implement the communication connection from the CPU to the RAID group disks in hardware. Further, this application scenario can be understood as a management method for RAID disk groups that combines software and hardware. The volume management device (VMD) of the computing device can provide hardware RAID configuration capabilities and hardware acceleration capabilities; in the operating system, the management capabilities for the RAID disk group are presented in the form of software RAID.

[0144] Exemplarily, the BMC sends a latency information statistics instruction of the hard disk to be identified to the volume management device (VMD), and the volume management device (VMD) performs the statistical task of reading and writing data time (i.e. latency information). For example, the volume management device (VMD) writes a piece of random data to the reserved area of ​​the hard disk through the CPU or the volume management device (VMD) itself every x seconds; then reads the data and counts the reading and writing time; the BMC periodically obtains latency information from the volume management device (VMD), and the BMC uses a sliding window method. If the count ratio of more than 100ms exceeds the threshold of 80%, it is considered that a slow disk phenomenon has occurred, and if the count ratio of more than 200 milliseconds exceeds the threshold of 70%, it is considered that a serious slow disk phenomenon has occurred.

[0145] It should be noted that if Figure 6 In the slow disk identification method shown, the BMC needs to send a delay information statistics instruction and receive delay information, and obtain the result of slow disk identification. The volume management device (VMD) can automatically complete multiple read and write operations on the hard disk to be identified to obtain delay information. This method can reduce the workload of the BMC and improve the work efficiency of the BMC.

[0146] See also Figure 7 , Figure 7 The fourth slow disk identification method provided in the present application is a flowchart. Figure 7 As shown, an embodiment of the present application provides a slow disk identification method, which is applied to a server, the server includes a BMC, a CPU, a volume management device (VMD) and a hard disk to be identified, the BMC is communicatively connected to the CPU; the CPU is communicatively connected to the volume management device (VMD), and the volume management device (VMD) is communicatively connected to the hard disk to be identified, and the method includes:

[0147] Step S710: The BMC sends a latency information statistics instruction to the volume management device (VMD) via the CPU.

[0148] Step S720: In response to the latency information statistics instruction, the volume management device (VMD) performs multiple read and write operations on the hard disk to be identified to obtain latency information. The latency information is used to indicate the time consumption of multiple read and write operations.

[0149] Step S730: The volume management device (VMD) determines whether to identify the hard disk to be identified as a slow disk based on the latency information.

[0150] It should be noted that if Figure 7 In the slow disk identification method shown, the BMC sends a latency information statistics instruction, and the volume management device (VMD) can complete multiple read and write operations on the hard disk to be identified, obtain latency information, and determine whether to identify the hard disk to be identified as a slow disk based on the latency information. This method can greatly reduce the workload of the BMC and improve the work efficiency of the BMC.

[0151] Exemplarily, the BMC sends a latency information statistics instruction of the hard disk to be identified to the volume management device (VMD), and the volume management device (VMD) performs the statistical task of reading and writing data time (i.e. latency information). For example, the volume management device (VMD) writes a piece of random data to the reserved area of ​​the hard disk through the CPU or the volume management device (VMD) itself every x seconds; then reads the data and counts the reading and writing time; the volume management device (VMD) uses a sliding window method, if the count ratio of more than 100ms exceeds the threshold of 80%, it is considered that a slow disk phenomenon has occurred, and if the count ratio of more than 200 milliseconds exceeds the threshold of 70%, it is considered that a serious slow disk phenomenon has occurred.

[0152] See also Figure 8 , Figure 8 The fifth slow disk identification method provided in the present application is a flowchart. Figure 8 As shown, an embodiment of the present application provides a slow disk identification method, which is applied to a server, the server including a BMC and a hard disk to be identified, and the method includes:

[0153] Step S810: The BMC sends a latency information statistics instruction to the hard disk to be identified.

[0154] Step S820: In response to the delay information statistics instruction, the hard disk to be identified performs multiple read and write operations to obtain delay information. The delay information is used to indicate the read and write time consumption of the multiple read and write operations.

[0155] Step S830: The BMC obtains latency information from the hard disk to be identified, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information.

[0156] It should be noted that if Figure 8The slow disk identification method shown does not limit the communication link. The BMC sends a delay information statistics instruction; the hard disk to be identified performs multiple read and write operations according to the delay information statistics instruction to obtain delay information; the BMC determines whether to identify the hard disk to be identified as a slow disk based on the delay information. Exemplarily, this method can greatly reduce the workload of the BMC and improve the work efficiency of the BMC.

[0157] Exemplarily, the BMC sends a latency information statistics instruction to the hard disk to be identified, and the hard disk to be identified performs a statistical task of the time consumed to read and write data (i.e., latency information): the BMC periodically obtains latency information from the hard disk, and uses a sliding window method. If the proportion of counts exceeding 100ms exceeds a threshold of 80%, it is considered that a slow disk phenomenon has occurred; if the proportion of counts exceeding 200 milliseconds exceeds a threshold of 70%, it is considered that a serious slow disk phenomenon has occurred.

[0158] In the present application embodiment, by Figure 8 The slow disk identification method shown in FIG. 1 can identify the slow disk problem of the hard disk. Figure 8 In the slow disk identification method shown, the BMC can identify the slow disk problem of the hard disk, and the slow disk detection can be implemented without deploying software through the operating system.

[0159] Optionally, the hard disk to be identified can count the read and write time consumption of the read and write operations of the entire hard disk. The read and write operations can be read and write operations issued by the business side, or read and write operations performed on a reserved area based on a slow disk identification task.

[0160] Optionally, hard disk log information is obtained, performance data on read and write time consumption is extracted, and latency information is obtained, thereby enabling identification and detection of slow disks.

[0161] The embodiments of the present application do not limit the channel and communication protocol between the BMC and the hard disk to be identified. The embodiments of the present application do not limit the hard disk specifications, types and other information, as long as the BMC out-of-band management is supported. The embodiments of the present application do not limit the hard disk controller type (such as software RAID, RAID card, etc.), as long as it supports out-of-band management.

[0162] In the present application embodiment, by Figure 8 In the slow disk identification method shown, the BMC directly obtains the latency information from the hard disk to be identified, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information. Among them, since the latency information is the latency information of the read and write operations recorded by the hard disk to be identified, there is no communication link. Therefore, there is no need to further check whether the communication link has a fault, and the BMC can directly determine whether the hard disk to be identified is a slow disk based on the latency information.

[0163] Based on the same concept as the aforementioned slow disk identification method, an embodiment of the present application further provides a computing device.

[0164] See also Fig. 9 , Fig. 9 A schematic diagram of another computing device provided in an embodiment of the present application. Fig. 9 As shown, the computing device 900 includes at least one processor 910, a memory 920 and a communication interface 930. The processor 910, the memory 920 and the communication interface 930 are connected in communication, and the communication connection can be achieved by wired means (such as a bus) or by wireless means. The communication interface 930 is used to receive data sent by other devices; the memory 920 stores computer instructions, and the processor 910 executes the computer instructions to execute the method in the aforementioned method embodiment.

[0165] It should be understood that in the embodiment of the present application, the processor 910 may be a central processing unit CPU, and the processor 910 may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0166] The memory 920 may include a read-only memory and a random access memory, and provides instructions and data to the processor 910. The memory 920 may also include a nonvolatile random access memory.

[0167] The memory 920 may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memories. Among them, the nonvolatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct ram bus RAM (DR RAM).

[0168] It should be understood that the computing device 900 according to the embodiment of the present application can execute the method mentioned in the embodiment of the present application. The detailed description of the implementation of the method is shown above, and for the sake of brevity, it will not be repeated here.

[0169] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer instructions are executed by a processor, the above-mentioned method is implemented.

[0170] An embodiment of the present application provides a computer program or a computer program product, wherein the computer program or the computer program product comprises instructions, and when the instructions are executed, the computer is caused to execute the above-mentioned method.

[0171] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those of ordinary skill in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0172] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented by hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0173] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A slow disk identification method, characterized in that: Applied to a server, the server includes a BMC, a RAID card and a hard disk to be identified, the BMC is communicatively connected with the RAID card, the RAID card is communicatively connected with the hard disk to be identified, and the method includes: The BMC sends a delay information statistics instruction to the RAID card; Based on the delay information statistics instruction, the RAID card performs multiple read and write operations on the hard disk to be identified to obtain delay information, where the delay information is used to indicate the read and write time consumption of the multiple read and write operations; The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, or the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information.

2. The method according to claim 1, characterized in that The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: The BMC obtains the delay information from the RAID card; Based on the latency information, the BMC determines a proportion of the number of timeouts in the total number of the multiple read and write operations, where the number of timeouts represents the number of read and write operations corresponding to the time taken to read and write exceeding a latency threshold; When the proportion is higher than a preset proportion, the BMC identifies the hard disk to be identified as a slow disk; Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: Based on the delay information, the RAID card determines the proportion of the number of timeouts in the total number of the multiple read and write operations, where the number of timeouts represents the number of read and write operations corresponding to the time taken to read and write exceeding the delay threshold; When the proportion is higher than a preset proportion, the RAID card identifies the hard disk to be identified as a slow disk.

3. The method according to claim 1, characterized in that The slow disk includes general slow disk and serious slow disk; The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: The BMC obtains the delay information from the RAID card; Based on the latency information, the BMC determines a proportion of the first number of first timeouts in the number of the multiple read and write operations, wherein the first timeout number indicates the number of read and write operations corresponding to which the read and write time exceeds a first threshold; When the first number of times accounts for a higher proportion than a first preset number of times, the BMC identifies the hard disk to be identified as a general slow disk; Based on the latency information, the BMC determines a proportion of a second number of timeouts in the number of the multiple read and write operations, the second number of timeouts indicating the number of read and write operations corresponding to a write time exceeding a second threshold, the second threshold being greater than the first threshold; When the proportion of the second number of times is higher than the second preset number of times, the BMC identifies the hard disk to be identified as a serious slow disk; Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: Based on the delay information, the RAID card determines the proportion of the first number of first timeouts in the number of times of the multiple read and write operations, wherein the first timeout number indicates the number of read and write operations corresponding to the time taken to read and write exceeding the first threshold; When the first number of times accounts for a higher proportion than a first preset number of times, the RAID card identifies the hard disk to be identified as a general slow disk; Based on the delay information, the RAID card determines a proportion of a second number of timeouts in the number of times of the multiple read and write operations, the second number of timeouts indicating the number of read and write operations corresponding to a write time exceeding a second threshold, the second threshold being greater than the first threshold; When the proportion of the second number of times is higher than the second preset number of times, the RAID card identifies the hard disk to be identified as a serious slow disk.

4. The method according to any one of claims 1 to 3, characterized in that: The read and write operations include: The RAID card writes target data to a preset area of ​​the hard disk to be identified; The RAID card reads the target data from the preset area.

5. The method according to any one of claims 1 to 4, characterized in that: The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: The BMC obtains the delay information from the RAID card; Based on the delay information, the BMC determines whether the hard disk to be identified has a slow disk phenomenon; When the hard disk to be identified is slow, the BMC determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; When there is no communication failure in the communication link between the RAID card and the hard disk to be identified, the BMC identifies the hard disk to be identified as a slow disk; Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: Based on the delay information, the RAID card determines whether the hard disk to be identified has a slow disk phenomenon; When the hard disk to be identified is slow, the RAID card determines whether a communication failure occurs in the communication link from the RAID card to the hard disk to be identified; When there is no communication failure in the communication link between the RAID card and the hard disk to be identified, the RAID card identifies the hard disk to be identified as a slow disk.

6. The method according to any one of claims 1 to 5, characterized in that: The BMC obtains the delay information from the RAID card, and determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: The BMC obtains the delay information from the RAID card; The BMC processes the delay information through a sliding window method to obtain a sliding window statistical result; According to the sliding window statistical result, the BMC determines whether to identify the hard disk to be identified as a slow disk; Alternatively, the RAID card determines whether to identify the hard disk to be identified as a slow disk based on the delay information, including: The RAID card processes the delay information through a sliding window method to obtain a sliding window statistical result; According to the sliding window statistical result, the RAID card determines whether to identify the hard disk to be identified as a slow disk.

7. A slow disk identification method, characterized in that: Applied to a server, the server includes a BMC and a hard disk to be identified, and the method includes: The BMC performs multiple read and write operations on the hard disk to be identified to obtain delay information, where the delay information is used to indicate the reading and writing time consumption of the multiple read and write operations; Based on the delay information, the BMC determines whether to identify the hard disk to be identified as a slow disk.

8. A method for identifying a slow disk, characterized in that: Applied to a server, the server includes a BMC, a CPU, a volume management device and a hard disk to be identified, the BMC is communicatively connected to the CPU; the CPU is communicatively connected to the volume management device, and the volume management device is communicatively connected to the hard disk to be identified, the method includes: The BMC sends a latency information statistics instruction to the volume management device via the CPU; Based on the delay information statistics instruction, the volume management device performs multiple read and write operations on the hard disk to be identified to obtain delay information, where the delay information is used to indicate the reading and writing time consumption of the multiple read and write operations; The BMC obtains the latency information from the volume management device via the CPU, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information, or, based on the latency information, the volume management device determines whether to identify the hard disk to be identified as a slow disk.

9. A method for identifying a slow disk, characterized in that: Applied to a server, the server includes a BMC and a hard disk to be identified, and the method includes: The BMC sends a latency information statistics instruction to the hard disk to be identified; Based on the delay information statistics instruction, the hard disk to be identified performs multiple read and write operations to obtain delay information, where the delay information is used to indicate the read and write time consumption of the multiple read and write operations; The BMC obtains the latency information from the hard disk to be identified, and determines whether to identify the hard disk to be identified as a slow disk based on the latency information.

10. A computing device, characterized in that: The computing device comprises: a processor and a memory; the memory stores instruction operations or codes; the processor is configured to communicate with the memory and execute the instruction operations or codes in the memory to perform any method in claims 1-9.

Images