Monitoring index generation method and device for disk array card, equipment and medium

Abstract

The invention discloses a method, a device and equipment for generating a monitoring index of a disk array card and a medium, and relates to the technical field of monitoring, and the method comprises the following steps: obtaining each business constraint rule, and obtaining disk array card configuration information corresponding to each business constraint rule; wherein the service constraint rule comprises each configuration attribute and a constraint relationship between the configuration attributes; determining the satisfaction condition of each service constraint rule according to the configuration information of the disk array card; according to a performance index subset corresponding to the target rule in the performance index set, generating a monitoring index set so as to perform performance monitoring on the disk array card chip by using the monitoring index set; according to the method and the device, the performance index subset required by the service scene can be automatically screened from the performance index set by utilizing the service constraint rule corresponding to the performance index subset of each performance monitoring engine, so that the automatic configuration generation of the monitoring index set for performance monitoring is realized, the workload of testers is reduced, and the test stability is improved.

Description

Technical Field

[0001] The present invention relates to the field of monitoring technology, and in particular to a method, device, equipment and medium for generating monitoring indicators of a disk array card. Background Art

[0002] With the advancement of integrated circuit technology, RAID (Redundant Array of Independent Disks) card chips have become the core components of data centers. RAID card chips contain multiple functional units such as processors, memory, peripheral storage devices, and internal submodules, so they face complex performance management challenges in design and application. In order to ensure the efficient operation of RAID card chips, monitoring their performance is an important means to ensure the smooth operation of data centers.

[0003] In the related art, the process of monitoring the performance of the RAID card chip is as follows: Figure 1 As shown in the figure, there are two channels in the RAID card chip (RAID chip): the business channel and the performance monitoring channel. The business channel is mainly used to configure business scenarios and issue business IO (Input / Output); the performance monitoring channel is used to issue performance monitoring instructions and report performance data. The conventional performance monitoring process is as follows: 1) The tester configures the business scenario in the business channel, and the business sends IO data normally; 2) The tester selects a subset of performance indicators that need to be monitored from the full performance set (such as Figure 1 The host sends the feature map to the performance monitoring framework, which then forwards it to the RAID card chip. The RAID card chip receives the feature map from each sub-module of the RAID card chip (such as Figure 1 The performance monitoring data is obtained from Modules 1, 2, and 3) in the host computer, and stored in the SOC (System on a Chip) data cache area; 4) when the host computer queries the data, the performance monitoring data is returned to the host computer, and the performance analysis and data display are performed on the host computer.

[0004] However, the process of manually selecting a subset of performance indicators from the full set of performance indicators based on business scenarios makes the workload of testers too heavy and cumbersome due to the large number of performance indicators in the full set of indicators. In addition, human behavior is an incompletely controllable element in the workflow, and manually constructing a subset of performance indicators will bring many unstable factors to the test process. How to automatically configure and generate a monitoring indicator set for performance monitoring, reduce the workload of testers, and improve test stability is an urgent problem to be solved today. Summary of the invention

[0005] The object of the present invention is to provide a method, device, equipment and computer-readable storage medium for generating monitoring metrics of a disk array card, which automatically configures and generates a set of monitoring metrics for performance monitoring, reduces the workload of testers, and improves test stability.

[0006] To solve the above technical problems, the present invention provides a method for generating monitoring metrics of a disk array card, including:

[0007] Obtain each service constraint rule and obtain the disk array card configuration information corresponding to each of the service constraint rules; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between the configuration attributes;

[0008] Determine the satisfaction status of each service constraint rule according to the disk array card configuration information;

[0009] Generate a set of monitoring metrics according to the subset of performance metrics corresponding to the target rule in the performance metric set, so as to use the set of monitoring metrics to perform performance monitoring on the disk array card chip; wherein, the performance metric set includes subsets of performance metrics of each performance monitoring engine, and each subset of performance metrics corresponds to a corresponding service constraint rule, and the target rule is a service constraint rule whose satisfaction status is a satisfied state.

[0010] On the other hand, the service constraint rule includes a configuration attribute AND set, a configuration attribute OR set, and an AND / OR flag bit; between two adjacent configuration attributes in the configuration attribute AND set, an AND logical operation is performed, between two adjacent configuration attributes in the configuration attribute OR set, an OR logical operation is performed, and the AND / OR flag bit is a flag bit for OR logical operation or a flag bit for AND logical operation.

[0011] On the other hand, the number of the configuration attribute AND set, the configuration attribute OR set, and the AND / OR flag bit in each service constraint rule is 1; the determining the satisfaction status of each service constraint rule according to the disk array card configuration information includes:

[0012] If the AND / OR flag bit in the current service constraint rule is a flag bit for AND logical operation, then through flag and &flag or , determine the satisfaction status of the current service constraint rule;

[0013] If the AND / OR flag bit in the current service constraint rule is a flag bit for OR logical operation, then through flag and |flag or , determine the satisfaction status of the current service constraint rule; wherein, flag and &flag or =1 or flag and | flag orWhen it is equal to 1, the satisfaction status of the current service constraint rule is the satisfied status; when the configuration attribute and set are empty, flag and is 1, and when the configuration attribute and set are non-empty, flag and is the output of the configuration attribute and set; when the configuration attribute or set is empty, flag or is 1, and when the configuration attribute or set is non-empty, flag or is the output of the configuration attribute and set.

[0014] On the other hand, the method further includes:

[0015] The host device distributes the monitoring metric set to the disk array card chip through the performance monitoring framework, and records the distribution order of the performance metrics in the monitoring metric set, so as to use the disk array card chip to obtain the performance metric data of each performance metric according to the monitoring metric set, and fill it into the on-chip system data buffer according to the distribution order;

[0016] Obtain the performance metric data from the on-chip system data buffer through the performance monitoring framework;

[0017] According to the distribution order, determine the correspondence between each of the performance metrics and the corresponding performance metric data.

[0018] On the other hand, the disk array card configuration information includes at least one of the disk presence status, the disk array level, and the distributed input / output type; the performance monitoring engine includes at least one of a downlink path engine, an uplink path engine, a basic engine, a verification engine, a disk management engine, and a resource management engine.

[0019] On the other hand, generating the monitoring metric set according to the subset of performance metrics corresponding to the target rule in the performance metric set includes:

[0020] Take the union of the subsets of performance metrics corresponding to all the target rules to obtain an initial metric set;

[0021] Judge whether the number of metrics in the initial metric set is greater than the number threshold;

[0022] If not, use the initial metric set as the monitoring metric set;

[0023] If so, use the initial metric set to perform pre-performance monitoring on the disk array card chip to obtain the pre-monitoring data corresponding to each performance metric in the initial metric set;

[0024] According to the pre-monitoring data, delete the performance metrics in the initial metric set to obtain the monitoring metric set; wherein, the number of metrics in the monitoring metric set is less than or equal to the number threshold.

[0025] On the other hand, using the initial index set to perform performance pre - monitoring on the disk array card chip to obtain pre - monitoring data corresponding to each performance index in the initial index set, including:

[0026] Determine the current performance index subset in the initial index set; wherein, the current performance index subset is any performance index subset in the initial index set;

[0027] Use the current performance index subset to perform performance pre - monitoring on the disk array card chip to obtain pre - monitoring data corresponding to each performance index in the current performance index subset;

[0028] Correspondingly, deleting the performance indexes in the initial index set according to the pre - monitoring data to obtain the monitoring index set, including:

[0029] According to the pre - monitoring data, calculate the data variance corresponding to each performance index in the current performance index subset;

[0030] Remove the performance index with the smallest data variance in the current performance index subset, and execute the step of determining whether the number of indexes in the initial index set is greater than the number threshold.

[0031] The present invention also provides a monitoring index generation device for a disk array card, including:

[0032] An acquisition module, configured to acquire each service constraint rule and acquire the disk array card configuration information corresponding to each service constraint rule; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between the configuration attributes;

[0033] A determination module, configured to determine the satisfaction situation of each service constraint rule according to the disk array card configuration information;

[0034] A generation module, configured to generate a monitoring index set according to the performance index subset corresponding to the target rule in the performance index collection, so as to perform performance monitoring on the disk array card chip by using the monitoring index set; wherein, the performance index collection includes the performance index subsets of each performance monitoring engine, each performance index subset corresponds to a corresponding service constraint rule respectively, and the target rule is the service constraint rule whose satisfaction situation is the satisfied state.

[0035] The present invention also provides a monitoring index generation device for a disk array card, including:

[0036] A memory, configured to store a computer program;

[0037] A processor, which is configured to implement the steps of the method for generating monitoring metrics of a disk array card as described in the above claims when executing the computer program.

[0038] In addition, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the method for generating monitoring metrics of a disk array card as described above.

[0039] A method for generating monitoring metrics of a disk array card provided by the present invention includes: obtaining each service constraint rule and obtaining the disk array card configuration information corresponding to each service constraint rule; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between configuration attributes; determining the satisfaction status of each service constraint rule according to the disk array card configuration information; generating a monitoring metric set according to the subset of performance metrics corresponding to the target rule in the performance metric set, so as to use the monitoring metric set to perform performance monitoring on the disk array card chip; wherein, the performance metric set includes the subset of performance metrics of each performance monitoring engine, each subset of performance metrics corresponds to the corresponding service constraint rule respectively, and the target rule is the service constraint rule whose satisfaction status is the satisfied state.

[0040] It can be seen that by generating a monitoring metric set according to the subset of performance metrics corresponding to the target rule in the performance metric set, the present invention can automatically screen the subset of performance metrics required by the service scenario from the performance metric set by using the service constraint rules corresponding to the subset of performance metrics of each performance monitoring engine, realize the automatic configuration generation of the monitoring metric set for performance monitoring, not only simplifies the operation process of performance monitoring, but also avoids the instability brought by manually constructing the metric set, reduces the workload of testers, and improves the test stability; in addition, the present invention further provides a device, equipment and computer-readable storage medium for generating monitoring metrics of a disk array card, which also have the above beneficial effects. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0042] Figure 1 It is a schematic flowchart of the performance monitoring of the RAID card chip in the related art;

[0043] Figure 2 It is a flowchart of a method for generating monitoring metrics of a disk array card provided by an embodiment of the present invention;

[0044] Figure 3 A display diagram for dividing a subset of performance indicators provided by an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of an index screening process for RAID resource constraints provided by an embodiment of the present invention;

[0046] Figure 5 A schematic diagram of a process for database construction and performance monitoring provided by an embodiment of the present invention;

[0047] Figure 6 A schematic diagram of a process for performance monitoring of a RAID card chip provided by an embodiment of the present invention;

[0048] Figure 7 A schematic diagram of an improvement in the storage method of performance indicator data provided by an embodiment of the present invention;

[0049] Figure 8 A structural block diagram of a monitoring index generation device for a disk array card provided by an embodiment of the present invention;

[0050] Figure 9 A structural schematic diagram of a monitoring index generation device for a disk array card provided by an embodiment of the present invention. Detailed implementation manners

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] Please refer to Figure 2 , Figure 2 A flowchart of a method for generating monitoring indicators for a disk array card provided by an embodiment of the present invention. The method may include:

[0053] Step 101: Obtain each service constraint rule and obtain the disk array card configuration information corresponding to each service constraint rule; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between configuration attributes.

[0054] It can be understood that the set of all performance indicators in this embodiment (i.e., the performance indicator collection) is divided into several performance indicator subsets according to the module engines required for the RAID card service (i.e., the performance monitoring engine), that is, all the performance indicators required for the performance test of the RAID card chip are divided into several performance indicator subsets according to functions; that is to say, the performance indicator collection can include the performance indicator subsets of each performance monitoring engine respectively.

[0055] Correspondingly, for the specific division method of the performance indicator subsets, that is, the specific configuration of the performance monitoring engine, it can be set by the designer according to the practical scenario and user requirements. For example, the performance monitoring engine can include at least one of a downlink path engine, an uplink path engine, a basic engine, a parity engine, a disk management engine, and a resource management engine; as Figure 3 shown, the performance monitoring engine can include a downlink path engine, an uplink path engine, a basic engine, a parity engine, a disk management engine, and a resource management engine; the performance indicator collection can include the performance indicator subsets corresponding to these performance monitoring engines respectively, and each performance indicator subset can include several performance indicators (such as Performance 1, Performance 2, and Performance 3, etc.).

[0056] Correspondingly, the service constraint rules in this embodiment can be the constraint rules related to the preset service scenarios, such as the constraint rules for RAID card configuration. In this embodiment, by setting the service constraint rules, the module engines to be monitored can be determined according to the software and hardware configuration of the RAID card, and then the corresponding performance indicator subsets can be determined to obtain the set of performance indicators required for the final performance monitoring (i.e., the monitoring indicator set). That is to say, each service constraint rule in this embodiment can be respectively corresponding to the corresponding performance indicator subset, and the performance indicator collection is screened according to the satisfaction situation of the service constraint rules.

[0057] It should be noted that for the specific setting of each service constraint rule in this embodiment, it can be set by the designer according to the practical scenario and user requirements. For example, each service constraint rule includes each configuration attribute and the constraint relationship between the configuration attributes, so as to determine the configuration attributes in the service constraint rule according to the RAID card configuration information, and detect whether it meets the required constraint relationship, so as to determine the satisfaction situation of the service constraint rule, and to determine whether the performance indicators in the corresponding performance indicator subset have monitoring value.

[0058] For example, the service constraint rules include configuration attribute sets, configuration attribute sets, and AND / OR flags; for adjacent configuration attributes in a configuration attribute set, an AND logical operation is performed between them, and for adjacent configuration attributes in a configuration attribute set, an OR logical operation is performed between them. The AND / OR flag is a flag for an OR logical operation or an AND logical operation flag. Correspondingly, this embodiment does not limit the specific number of configuration attribute sets, configuration attribute sets, and AND / OR flags in each service constraint rule. For example, the number of configuration attribute sets, configuration attribute sets, and AND / OR flags in each service constraint rule is 1.

[0059] Among them, the disk array card configuration information in this step can be information related to the software and hardware configuration of the disk array card. This embodiment does not limit the specific number and type of the disk array card configuration information. For example, the disk array card configuration information can include at least one of the disk presence status, disk array level, and issued input type.

[0060] Step 102: Determine the satisfaction status of each service constraint rule according to the disk array card configuration information.

[0061] Among them, in this step, the disk array card configuration information can be used to determine each configuration attribute in the service constraint rule, so as to determine the satisfaction status of each service constraint rule, so as to determine whether the performance indicators in the corresponding performance indicator subset have monitoring value, so as to screen out the performance indicator subsets that do not have monitoring value, and realize the RAID card configuration constraint.

[0062] Correspondingly, for the specific method of determining the satisfaction status of each service constraint rule according to the disk array card configuration information in this step, it can be set by the designer himself. For example, when the number of configuration attribute sets, configuration attribute sets, and AND / OR flags in each service constraint rule is 1, if the AND / OR flag in the current service constraint rule is an AND logical operation flag, then through flag and &flag or , determine the satisfaction status of the current service constraint rule; if the AND / OR flag in the current service constraint rule is an OR logical operation flag, then through flag and |flag or , determine the satisfaction status of the current service constraint rule; among them, when flag and &flag or =1 or flag and |flag or =1, the satisfaction status of the current service constraint rule is the satisfied state; when the configuration attribute set is empty, flag and is 1, and when the configuration attribute set is not empty, flag and is the output of the configuration attribute set; when the configuration attribute set is empty, flag oris 1 when the configuration attribute or set is not empty, flag or is the output of the configuration attribute and set.

[0063] That is to say, when any configuration attribute of the configuration attribute and set does not meet the requirements (i.e., the output is 0), the output (flag and ) is 0. When all configuration attributes meet the requirements (i.e., the output is 1) or the set is empty, the output (flag and ) is 1, that is . When all configuration attributes of the configuration attribute or set do not meet the requirements, the output (flag or ) is 0. When any configuration attribute meets the requirements or the set is empty, the output (flag or ) is 1. . The AND / OR flag is used to determine the relationship between the configuration attribute and set and the configuration attribute or set. For example, when the configuration attribute and set play a decisive role, this flag can be the flag for the AND logical operation (&); if the configuration attribute or set plays a decisive role, this flag can be the flag for the OR set logical operation (|); the satisfaction status of the service constraint rule (flag valid ) can be the logical operation result of flag and &flag or or flag and |flag or , that is . For example, if flag valid = 1, it can indicate that the satisfaction status can be the satisfied state; if flag valid = 0, it can indicate that the satisfaction status can be the unsatisfied state.

[0064] For example, in the case where the service constraint rule is the disk constraint rule, only when the corresponding disk is in place, the performance metrics in the performance metric subset of the disk management engine have monitoring value; the presence of the corresponding disk can be added to the configuration attribute and set of the disk constraint rule, with the presence being 1 and the absence being 0; the configuration attribute or set is empty; the AND / OR flag is AND (&). According to the above content, it can be calculated whether the performance metric subset of the disk management engine should be added to the monitoring metric set.

[0065] In the case where the service constraint rule is the constraint rule of the impact of the RAID card on the parity engine, since for the parity engine, only when the parity value needs to be calculated is the parity engine required, and it is valuable to monitor the performance metrics of the parity engine; if RAID0 does not need to calculate the parity value, and RAID 5 / 6 / TP all need the parity value; RAID 0 can be added to the configuration attribute and set, with RAID 0 configured as 0 and non-RAID 0 configured as 1; RAID 5 / 6 / TP are added to the configuration attribute or set, with the corresponding RAID configured as 1 and the non-corresponding RAID configured as 0; the AND / OR flag is OR (|).

[0066] When the service constraint rule is the IO constraint rule corresponding to the upstream path engine, the upstream path engine has monitoring value only when the read IO exists; when the service constraint rule is the IO constraint rule corresponding to the downstream path engine, the downstream path engine has monitoring value only when the write IO exists. Taking the IO constraint rule corresponding to the upstream path engine as an example, the IO read / write attribute can be added to the configuration attribute set. The read IO is 1 and the non-read IO is 0; if the configuration attribute set is empty, the AND / OR flag bit is AND (&).

[0067] Correspondingly, the specific number of service constraint rules in this embodiment is not limited and can be set by testers according to needs. For example, testers can add new service constraint rules according to newly discovered service constraints.

[0068] Step 103: Generate a monitoring index set according to the performance index subset corresponding to the target rule in the performance index set, so as to use the monitoring index set to monitor the performance of the disk array card chip; wherein, the performance index set includes the performance index subsets of each performance monitoring engine, and each performance index subset corresponds to the corresponding service constraint rule respectively, and the target rule is the service constraint rule whose satisfaction status is the satisfied state.

[0069] It can be understood that the target rule in this step can be the service constraint rule whose satisfaction status is the satisfied state, such as the service constraint rule with flag valid = 1. In this step, the performance index subset corresponding to the target rule can be used to generate a monitoring index set, so as to realize the automatic screening of the performance indexes in the performance index set and generate the monitoring index set required for the performance monitoring of the RAID card chip.

[0070] Correspondingly, for the specific method of generating the monitoring index set according to the performance index subset corresponding to the target rule in the performance index set in this embodiment, it can be set by the designer himself. For example, the union of the performance index subsets corresponding to all target rules can be directly taken to obtain the monitoring index set. Due to the limitations of the internal resources of the RAID card chip, such as the buffer size and CPU (Central Processing Unit) resources, the total amount of data queries allowed is limited. Therefore, the amount of data monitored by the monitoring index set cannot exceed the amount of monitoring data actually supported by the system, that is, the monitoring index set needs to meet the RAID resource constraints.

[0071] For example Figure 4As shown, in this step, the union of the subsets of performance indicators corresponding to all target rules can be taken to obtain the initial indicator set (subset of performance indicators 1); check whether the initial indicator set meets the RAID resource constraints; if it meets, use the initial indicator set as the monitoring indicator set (subset of target performance indicators); if it does not meet, the initial indicator set can be used to perform pre-monitoring of the performance of the disk array card chip to obtain the pre-monitoring data corresponding to each performance indicator in the initial indicator set; according to the pre-monitoring data, delete the performance indicators in the initial indicator set to obtain the monitoring indicator set; where the number of indicators in the monitoring indicator set is less than or equal to the quantity threshold.

[0072] Correspondingly, this embodiment does not limit the specific content of the above RAID resource constraints. For example, the RAID resource constraints can limit that the number of performance indicators cannot exceed the quantity threshold; correspondingly, the above check of whether the initial indicator set meets the RAID resource constraints can include determining whether the number of indicators in the initial indicator set is greater than the quantity threshold; if so, it is determined that the RAID resource constraints are met, and the initial indicator set can be used as the monitoring indicator set; if not, it can be determined that the RAID resource constraints are not met, and the initial indicator set can be used to perform pre-monitoring of the performance of the disk array card chip to obtain the pre-monitoring data corresponding to each performance indicator in the initial indicator set; according to the pre-monitoring data, delete the performance indicators in the initial indicator set to obtain the monitoring indicator set.

[0073] Correspondingly, for the specific method of deleting the performance indicators in the initial indicator set using the pre-monitoring data, it can be set by the designer according to the practical scenario and user requirements. For example, in this embodiment, the principle of deleting performance indicators can be to retain the performance indicators with relatively large data changes and reduce the performance indicators with relatively gentle data changes; that is, the data fluctuation of the pre-monitoring data corresponding to the performance indicators deleted from the initial indicator set can be greater than the data fluctuation of the pre-monitoring data corresponding to the performance indicators retained in the initial indicator set (or the same subset of performance indicators). For example, in this embodiment, the pre-monitoring data corresponding to all performance indicators in the initial indicator set can be directly used to calculate the data variance corresponding to each performance indicator in the current subset of performance indicators; delete the corresponding performance indicators in the initial indicator set in ascending order of the corresponding data variance until the number of indicators in the initial indicator set is equal to the quantity threshold or other preset values less than or equal to the quantity threshold.

[0074] Correspondingly, in order to evenly distribute the performance indicators to be deleted among each subset of performance indicators, the process of pre-monitoring the performance of the disk array card chip using the initial indicator set to obtain the pre-monitoring data corresponding to each performance indicator in the initial indicator set may include: determining the current subset of performance indicators in the initial indicator set; where the current subset of performance indicators is any subset of performance indicators in the initial indicator set; using the current subset of performance indicators to pre-monitor the performance of the disk array card chip to obtain the pre-monitoring data corresponding to each performance indicator in the current subset of performance indicators. Correspondingly, the process of deleting the performance indicators in the initial indicator set according to the pre-monitoring data to obtain the monitoring indicator set may include: calculating the data variance corresponding to each performance indicator in the current subset of performance indicators according to the pre-monitoring data; removing the performance indicator with the smallest corresponding data variance in the current subset of performance indicators, and performing the step of judging whether the number of indicators in the initial indicator set is greater than the quantity threshold.

[0075] That is to say, in this step, the union of the subsets of performance indicators corresponding to all target rules is taken to obtain the initial indicator set; it is judged whether the number of indicators in the initial indicator set is greater than the quantity threshold; if it is not greater than the quantity threshold, the current initial indicator set is used as the monitoring indicator set; if it is greater than the quantity threshold, the current subset of performance indicators in the initial indicator set is determined, for example, according to the preset reduction order (such as the order from largest to smallest in terms of the number of indicators), the current subset of performance indicators is determined from all subsets of performance indicators in the initial indicator set; using the current subset of performance indicators to pre-monitor the performance of the disk array card chip to obtain the pre-monitoring data corresponding to each performance indicator in the current subset of performance indicators; calculating the data variance corresponding to each performance indicator in the current subset of performance indicators according to the pre-monitoring data; removing the performance indicator with the smallest corresponding data variance in the current subset of performance indicators, and returning to the step of judging whether the number of indicators in the initial indicator set is greater than the quantity threshold, so as to continue to reduce the next subset of performance indicators when it is greater than the quantity threshold until the number of indicators in the initial indicator set is not greater than the quantity threshold.

[0076] Among them, the above process measures the degree of data fluctuation by data variance, and other methods can also be used to measure the degree of data fluctuation, and this embodiment does not make any restrictions on this. For the specific calculation method of data variance, it can be set by the designer himself. For example, after pre-monitoring the performance for a period of time, for each performance indicator in the current subset of performance indicators, a corresponding set of discrete performance data (i.e., pre-monitoring data) can be obtained. Using these data, the data variance corresponding to each performance indicator can be calculated through , where δ 2 can represent the data variance corresponding to the current performance indicator, x can represent the x-th pre-monitoring data corresponding to the current performance indicator, It can represent the average value of N pre-monitoring data corresponding to the current performance indicator, and N can represent the total amount of pre-monitoring data corresponding to the current performance indicator.

[0077] It should be noted that the monitoring index generation method of the disk array card provided in this embodiment can be applied to a database, such as Figure 5 As shown in the figure, during the development period, developers can build and maintain a simplified database for testing, a subset of performance indicators and business constraints in the database. This stage mainly verifies the development process. In order to build test scenarios, virtual data can be added, which does not necessarily conform to the real scenario. Testers maintain the database and add real business constraints and performance indicator subsets that meet business requirements according to the test plan. When testers monitor the performance of RAID card chips, they can issue test commands to drive feature generation tools (such as Figure 6 The feature generation tool reads the business constraint rules from the test library and reads the raid card configuration information from the SoC chip (such as the RAID card chip); the indicator generation tool filters out the final monitoring indicator set from the indicator collection in the database based on the read business constraint rules and RAID card configuration information; the monitoring indicator set is input into the performance monitoring framework for performance monitoring.

[0078] Correspondingly, the monitoring indicator generation method for the disk array card provided in this embodiment can also be applied to a host device (Host) or other devices. For example, the above-mentioned feature generation tool can be run in the host device to use the obtained business constraint rules and RAID card configuration information to filter out the final monitoring indicator set from the indicator collection in the database, so as to input the monitoring indicator set into the performance monitoring framework for performance monitoring.

[0079] It should be noted that after the monitoring indicator set is generated, the method provided in this embodiment can also use the monitoring indicator set to monitor the performance of the disk array card chip; for example, the monitoring indicator set can be sent to the performance monitoring framework, and the monitoring indicator set can be sent to the disk array card chip through the performance monitoring framework, so that the RAID card chip can obtain the performance data of each performance monitoring engine (i.e., performance indicator data) according to the monitoring indicator set, and fill the data into the cache area (e.g., Figure 6 Afterwards, the host device can obtain the data in the cache area through the performance monitoring framework to implement performance monitoring of the RAID chip for subsequent performance analysis.

[0080] Furthermore, in order to increase the cache area in the RAID card chip (such as Figure 6In the case of the SOC data buffer), the present embodiment can also provide a mechanism that allows data to be closely arranged to maximize the utilization of the buffer and increase the number of performance metrics that can be monitored simultaneously. For example, before or during the process of sending the monitoring metric set to the performance monitoring framework, the host device can record the sending order of the performance metrics in the monitoring metric set. For example, the host device can directly back up a copy of the monitoring metric set to ensure consistency with the sending order. Correspondingly, the RAID card chip can store the performance metric data corresponding to each performance metric in the buffer according to this sending order, so that the performance metric data corresponding to each performance metric in the buffer can be closely arranged according to this sending order. As Figure 7 shown, compared with the storage method in the related art where the buffer stores performance metric data according to the performance metric id (encoding), performance metric length, and performance metric data, the ability to store performance metric data in the buffer can be tripled. Correspondingly, the host device can obtain the performance metric data in the buffer through the performance monitoring framework according to the sending order and parse it to determine the correspondence between the performance metric data and the corresponding performance metric (such as the performance metric id).

[0081] That is to say, the performance metric data in the buffer of the RAID card chip can be directly stored according to the sending order of the performance metrics in the monitoring metric set, and the buffer does not need to store the performance metric id and the performance metric length.

[0082] For example, when the method provided by the present embodiment is applied to a host device, after generating the monitoring metric set, the host device can also send the monitoring metric set to the disk array card chip through the performance monitoring framework and record the sending order of the performance metrics in the monitoring metric set, so as to use the disk array card chip to obtain the performance metric data of each performance metric according to the monitoring metric set and fill it into the on-chip system data buffer according to the sending order; obtain the performance metric data from the on-chip system data buffer through the performance monitoring framework; determine the correspondence between each performance metric and the corresponding performance metric data according to the sending order.

[0083] As Figure 6As shown in the figure, the process of performance monitoring of the RAID card chip may include: ① After the test developer configures the RAID scenario and stably runs the service, the IO service can be issued; ② The test developer triggers the feature generator to start performance monitoring; the feature generator obtains the prior information required to generate the monitoring metric set from the outside world, including reading the performance metric collection and business constraint rules from the database, and obtaining the real RAID card configuration information from the RAID card chip; ③ Generate the monitoring metric set according to the prior information; ④ The host device backs up the monitoring metric subset and records the issuing order of the monitoring metric subset. Since the memory of the host device is relatively sufficient, the memory occupation of this part is an acceptable cost for the host device; ⑤ According to the issuing order, the monitoring metric subset is sent to the performance monitoring framework, and the performance monitoring framework can perform preprocessing such as verification on the monitoring metric subset; ⑥ The performance monitoring framework sends the monitoring metric subset to the RAID card chip; ⑦ The RAID card chip obtains the performance data points (i.e., performance metric data) of each module (i.e., the performance monitoring engine) according to the monitoring metric subset, and fills the data into the SOC data buffer according to the issuing order; ⑧ The performance monitoring framework obtains the performance metric data from the SOC data buffer; ⑨ The host device obtains the performance metric data from the performance framework, and determines the correspondence between the performance metric id and the performance metric data according to the backed-up monitoring metric subset, and supplies it to the domain expert for performance analysis later.

[0084] In this embodiment, the embodiment of the present invention can generate a monitoring metric set according to the performance metric subset corresponding to the target rule in the performance metric collection, and can automatically screen the performance metric subset required by the business scenario from the performance metric collection by using the business constraint rules corresponding to the performance metric subsets of each performance monitoring engine, realizing the automatic configuration generation of the monitoring metric set for performance monitoring, which not only simplifies the operation process of performance monitoring, but also avoids the instability brought by manually constructing the metric set, reduces the workload of testers, and improves the test stability.

[0085] Corresponding to the above method embodiment, the embodiment of the present invention also provides a monitoring metric generation device for a disk array card. A monitoring metric generation device for a disk array card described below can be correspondingly referred to the monitoring metric generation method for a disk array card described above.

[0086] Please refer to Figure 8 , Figure 8 which is a structural block diagram of a monitoring metric generation device for a disk array card provided by the embodiment of the present invention. The device may include:

[0087] An obtaining module 10, configured to obtain each business constraint rule and obtain the disk array card configuration information corresponding to each business constraint rule; wherein, the business constraint rule includes each configuration attribute and the constraint relationship between configuration attributes;

[0088] A determination module 20, configured to determine the satisfaction of each service constraint rule according to the disk array card configuration information;

[0089] A generation module 30, configured to generate a monitoring metric set according to a subset of performance metrics corresponding to a target rule in a performance metric collection, so as to use the monitoring metric set to perform performance monitoring on a disk array card chip; wherein, the performance metric collection includes subsets of performance metrics of respective performance monitoring engines, each subset of performance metrics corresponds to a corresponding service constraint rule, and the target rule is a service constraint rule whose satisfaction status is a satisfied state.

[0090] In some embodiments, the service constraint rule includes a configuration attribute AND set, a configuration attribute OR set, and an AND / OR flag bit; between two adjacent configuration attributes in the configuration attribute AND set, an AND logical operation is performed, between two adjacent configuration attributes in the configuration attribute OR set, an OR logical operation is performed, and the AND / OR flag bit is a flag bit for an OR logical operation or a flag bit for an AND logical operation.

[0091] In some embodiments, the number of the configuration attribute AND set, the configuration attribute OR set, and the AND / OR flag bit in each service constraint rule is 1;

[0092] The determination module 20 may specifically be configured to, if the AND / OR flag bit in the current service constraint rule is a flag bit for an AND logical operation, determine the satisfaction of the current service constraint rule through flag and &flag or , and if the AND / OR flag bit in the current service constraint rule is a flag bit for an OR logical operation, determine the satisfaction of the current service constraint rule through flag and |flag or ; wherein, when flag and &flag or =1 or flag and |flag or =1, the satisfaction of the current service constraint rule is a satisfied state; when the configuration attribute AND set is empty, flag and is 1, when the configuration attribute AND set is non-empty, flag and is the output of the configuration attribute AND set; when the configuration attribute OR set is empty, flag or is 1, when the configuration attribute OR set is non-empty, flag or is the output of the configuration attribute AND set.

[0093] In some embodiments, the apparatus may be applied to a host device, and further includes:

[0094] A distribution module, configured to distribute a set of monitoring metrics to a disk array card chip through a performance monitoring framework, and record the distribution order of the performance metrics in the set of monitoring metrics, so as to enable the disk array card chip to obtain performance metric data of each performance metric according to the set of monitoring metrics and fill it into the on-chip system data buffer in the distribution order;

[0095] A cache acquisition module, configured to acquire performance metric data from the on-chip system data buffer through a performance monitoring framework;

[0096] An index correspondence module, configured to determine the correspondence between each performance metric and the corresponding performance metric data according to the distribution order.

[0097] In some embodiments, the disk array card configuration information includes at least one of the disk presence status, the disk array level, and the distributed input / output type; the performance monitoring engine includes at least one of a downlink path engine, an uplink path engine, a basic engine, a verification engine, a disk management engine, and a resource management engine.

[0098] In some embodiments, the generation module 30 may include:

[0099] A union sub-module, configured to take the union of the subsets of performance metrics corresponding to all target rules to obtain an initial index set;

[0100] A judgment sub-module, configured to judge whether the number of metrics in the initial index set is greater than a quantity threshold;

[0101] A determination sub-module, configured to, if it is not greater than the quantity threshold, use the initial index set as the set of monitoring metrics;

[0102] A pre-monitoring sub-module, configured to, if it is greater than the quantity threshold, perform performance pre-monitoring on the disk array card chip using the initial index set to obtain pre-monitoring data corresponding to each performance metric in the initial index set;

[0103] A reduction sub-module, configured to delete performance metrics in the initial index set according to the pre-monitoring data to obtain a set of monitoring metrics; wherein, the number of metrics in the set of monitoring metrics is less than or equal to the quantity threshold.

[0104] In some embodiments, the pre-monitoring sub-module may include:

[0105] A selection and determination unit, configured to determine the current subset of performance metrics in the initial index set; wherein, the current subset of performance metrics is any subset of performance metrics in the initial index set;

[0106] A pre-monitoring unit, configured to perform performance pre-monitoring on the disk array card chip using the current subset of performance metrics to obtain pre-monitoring data corresponding to each performance metric in the current subset of performance metrics;

[0107] Correspondingly, the reduction sub-module may include:

[0108] A variance calculation unit, configured to calculate the data variance corresponding to each performance metric in the current subset of performance metrics according to pre-monitored data;

[0109] A reduction unit, configured to remove the performance metric with the smallest corresponding data variance in the current subset of performance metrics, and send a start signal to the judgment sub-module.

[0110] In this embodiment, the embodiment of the present invention can generate a monitoring metric set through the generation module 30 according to the subset of performance metrics corresponding to the target rule in the set of performance metrics, and can utilize the business constraint rules corresponding to the subsets of performance metrics of each performance monitoring engine to automatically screen out the subset of performance metrics required for the business scenario from the set of performance metrics, realizing the automatic configuration generation of the monitoring metric set for performance monitoring. This not only simplifies the operation process of performance monitoring, but also avoids the instability caused by manual construction of the metric set, reduces the workload of testers, and improves the test stability.

[0111] Corresponding to the above method embodiment, the embodiment of the present invention further provides a monitoring metric generation device for a disk array card. The monitoring metric generation device for a disk array card described below can be correspondingly referred to with the monitoring metric generation method for a disk array card described above.

[0112] Please refer to Figure 9 , Figure 9 , which is a schematic structural diagram of a monitoring metric generation device for a disk array card provided by the embodiment of the present invention. The device may include:

[0113] A memory D1, configured to store a computer program;

[0114] A processor D2, configured to implement the steps of the monitoring metric generation method for a disk array card provided by the above method embodiment when executing the computer program.

[0115] Among them, the monitoring metric generation device for a disk array card provided in this embodiment may specifically be a database or a host device.

[0116] Corresponding to the above method embodiment, the embodiment of the present invention further provides a computer program product. The computer program product described below can be correspondingly referred to with the monitoring metric generation method for a disk array card described above.

[0117] A computer program product includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the steps of the monitoring metric generation method for a disk array card provided by the above method embodiment are implemented.

[0118] Corresponding to the above method embodiments, an embodiment of the present invention further provides a computer-readable storage medium. A computer-readable storage medium described below corresponds and refers to a method for generating monitoring metrics of a disk array card described above.

[0119] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the method embodiment for generating monitoring metrics of the disk array card described above are implemented.

[0120] The computer-readable storage medium may specifically be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc., various readable storage media that can store program codes.

[0121] The embodiments in the specification are described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the devices, equipment, computer program products, and computer-readable storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple. For the relevant parts, reference can be made to the description in the method part.

[0122] The above has introduced in detail a method, device, equipment, and computer-readable storage medium for generating monitoring metrics of a disk array card provided by the present invention. Specific examples are used herein to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the present invention.

Claims

1. A method for generating monitoring metrics of a disk array card, characterized in that, Including: Obtain each service constraint rule, and obtain the disk array card configuration information corresponding to each of the service constraint rules; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between the configuration attributes; Determine the satisfaction status of each of the service constraint rules according to the disk array card configuration information; Generate a monitoring index set according to the performance index subset corresponding to the target rule in the performance index set, so as to use the monitoring index set to perform performance monitoring on the disk array card chip; wherein, the performance index set includes the performance index subsets of each performance monitoring engine, and each of the performance index subsets corresponds to the corresponding service constraint rule, and the target rule is the service constraint rule whose satisfaction status is the satisfied state.

2. The method for generating monitoring metrics of the disk array card according to claim 1, wherein, The service constraint rule includes a configuration attribute AND set, a configuration attribute OR set, and an AND / OR flag bit; between two adjacent configuration attributes in the configuration attribute AND set, an AND logical operation is performed, between two adjacent configuration attributes in the configuration attribute OR set, an OR logical operation is performed, and the AND / OR flag bit is the flag bit of the OR logical operation or the flag bit of the AND logical operation.

3. The method for generating monitoring metrics of the disk array card according to claim 2, wherein The number of the configuration attribute AND set, the configuration attribute OR set, and the AND / OR flag bit in each service constraint rule is 1; the determining the satisfaction status of each of the service constraint rules according to the disk array card configuration information includes: If the AND / OR flag bit in the current service constraint rule is the flag bit for logical AND operation, then determine the satisfaction status of the current service constraint rule through flag and &flag or , and determine the satisfaction status of the current service constraint rule; If the AND / OR flag bit in the current service constraint rule is the flag bit for OR logical operation, then determine the satisfaction status of the current service constraint rule through flag and |flag or , and determine the satisfaction status of the current service constraint rule; where flag and &flag or = 1 or flag and |flag or = 1, the satisfaction status of the current service constraint rule is the satisfied state; when the configuration attribute AND set is empty, flag and is 1, and when the configuration attribute AND set is non-empty, flag and is the output of the configuration attribute AND set; when the configuration attribute OR set is empty, flag or is 1, and when the configuration attribute OR set is non-empty, flag or is the output of the configuration attribute AND set.

4. The method for generating monitoring metrics of the disk array card according to claim 1, wherein Also including: The host device issues the monitoring index set to the disk array card chip through the performance monitoring framework, and records the issuing order of the performance indexes in the monitoring index set, so as to use the disk array card chip to obtain the performance index data of each performance index according to the monitoring index set, and fill it into the on-chip system data buffer according to the issuing order; Obtain the performance index data from the on-chip system data buffer through the performance monitoring framework; Determine the correspondence between each of the performance indexes and the corresponding performance index data according to the issuing order.

5. The method for generating monitoring metrics of the disk array card according to claim 1, characterized in that The disk array card configuration information includes at least one of the disk presence status, the disk array level, and the issued input / output type; the performance monitoring engine includes at least one of a downlink path engine, an uplink path engine, a basic engine, a check engine, a disk management engine, and a resource management engine.

6. The method for generating monitoring metrics of the disk array card according to any one of claims 1 to 5, characterized in that, The generating a monitoring index set according to the performance index subset corresponding to the target rule in the performance index set includes: Take the union of the performance index subsets corresponding to all the target rules to obtain an initial index set; Judge whether the number of indexes in the initial index set is greater than the number threshold; If not, use the initial index set as the monitoring index set; If so, perform performance pre-monitoring on the disk array card chip by using the initial index set to obtain the pre-monitoring data corresponding to each performance index in the initial index set; Delete the performance indexes in the initial index set according to the pre-monitoring data to obtain the monitoring index set; wherein, the number of indexes in the monitoring index set is less than or equal to the number threshold.

7. The method for generating monitoring metrics of the disk array card according to claim 6, wherein Performing performance pre - monitoring on the disk array card chip using the initial index set to obtain pre - monitoring data corresponding to each performance index in the initial index set, including: Determining the current performance index subset in the initial index set; wherein, the current performance index subset is any performance index subset in the initial index set; Performing performance pre - monitoring on the disk array card chip using the current performance index subset to obtain pre - monitoring data corresponding to each performance index in the current performance index subset; Correspondingly, the step of pruning the performance indexes in the initial index set according to the pre - monitoring data to obtain the monitoring index set includes: Calculating the data variance corresponding to each performance index in the current performance index subset according to the pre - monitoring data; Removing the performance index with the smallest data variance in the current performance index subset, and performing the step of determining whether the number of indexes in the initial index set is greater than the number threshold.

8. A monitoring metric generation device for a disk array card, characterized in that, Including: An acquisition module, configured to acquire each service constraint rule and acquire the disk array card configuration information corresponding to each service constraint rule; wherein, the service constraint rule includes each configuration attribute and the constraint relationship between the configuration attributes; A determination module, configured to determine the satisfaction status of each service constraint rule according to the disk array card configuration information; A generation module, configured to generate a monitoring index set according to the performance index subset corresponding to the target rule in the performance index collection, so as to perform performance monitoring on the disk array card chip using the monitoring index set; wherein, the performance index collection includes the performance index subsets of each performance monitoring engine, and each performance index subset corresponds to a corresponding service constraint rule, and the target rule is a service constraint rule whose satisfaction status is the satisfied state.

9. A monitoring index generation device for a disk array card, characterized in that Including: A memory, configured to store a computer program; A processor, configured to implement the steps of the method for generating monitoring indexes of the disk array card according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer - readable storage medium, and when the computer program is executed by a processor, the steps of the method for generating monitoring indexes of the disk array card according to any one of claims 1 to 7 are implemented.

Images