Information recommendation method, data processing method, device, and storage medium

Abstract

Provided in the embodiments of the present disclosure are an information recommendation method, a data processing method, a device, and a storage medium. The method comprises the following steps: acquiring historical usage data of a cloud disk, wherein the cloud disk is configured with pre-configured read-write performance and burst read-write performance, the pre-configured read-write performance is billed on the basis of a pre-configured performance value and a billing duration, and the burst read-write performance is billed on the basis of the number of instances of burst reading and writing of the cloud disk; on the basis of the historical usage data, predicting a target pre-configuration performance value meeting a preset cost optimization condition; and on the basis of the target pre-configuration performance value, sending to a user recommendation information related to cost optimization. The technical solution provided in the embodiments of the present disclosure can optimize the cost of using a cloud disk by a user.

Description

[0001] Information Recommendation Method, Data Processing Method, Device, and Storage Medium Technical Field The present disclosure relates to the field of cloud computing technology, and in particular to an information recommendation method, data processing method, device, and storage medium. Background Cloud server ECS (Elastic Compute Service) is a simple, efficient, secure, reliable, and elastically scalable cloud computing service provided by cloud service providers. Block storage (e.g., cloud disk) is a block device product provided by cloud service providers for cloud server ECS. It features high performance and low latency, supports random read and write, and meets the data storage requirements of most common project scenarios. Users can use block storage by formatting and creating a file system just like using a physical hard drive. After a cloud disk is mounted to an ECS instance, the pressure borne by the cloud disk comes from the instance to which it is mounted; the performance of all applications in the instance using the cloud disk is included in performance statistics. In actual applications, the following scenarios exist:

[0002] 1. The cloud disk capacity required by the instance is fixed, but higher cloud disk performance is required to support the operation of the instance.

[0003] 2. Instance operation fluctuates significantly, with frequent peaks, requiring cloud disks to be able to cope with emergencies. To address these scenarios, some cloud service providers offer cloud disks that support customizing pre-configured and burst performance based on project requirements. This type of cloud disk decouples cloud disk capacity from performance to meet user needs. During cloud disk operation, burst performance is triggered only when the pre-configured performance of the cloud disk fails to meet the instance's operating requirements. Once pre-configured performance is activated, it is billed based on the billing duration (also understood as the activation duration), regardless of whether the user uses it. Burst performance is billed based on the number of burst reads and writes performed by the user. As can be seen in the prior art, improper configuration of pre-configured performance by users can result in high cloud disk usage costs. SUMMARY OF THE INVENTION In view of the above issues, the present disclosure provides an information recommendation method, data processing method, device, and storage medium that address, or at least partially address, these issues. A first aspect of the present disclosure provides an information recommendation method, comprising: obtaining historical usage data of a cloud disk; the cloud disk being configured with preconfigured read and write performance and burst read and write performance; the preconfigured read and write performance being billed based on a preconfigured performance value and a billing duration; the burst read and write performance being billed based on the number of burst reads and writes of the cloud disk; the number of burst reads and writes referring to the number of reads and writes involved in a portion of an actually used performance value that exceeds a preconfigured performance value; predicting, based on the historical usage data, a target preconfigured performance value that meets a preset cost optimization condition; and sending recommendation information about cost optimization to a user based on the target preconfigured performance value. Optionally, based on the historical usage data, predicting a target preconfigured performance value that meets a preset cost optimization condition includes: determining, based on the historical usage data, a performance peak value of the actual usage of the cloud disk during a historical time period; when the preconfigured performance value currently configured for the cloud disk is less than the performance peak value, determining multiple alternative preconfigured performance values ​​between the preconfigured performance value currently configured for the cloud disk and the performance peak value; for each alternative preconfigured performance value, determining, based on the historical usage data, an adjusted usage cost of the cloud disk during the historical time period when the preconfigured performance value is adjusted to the alternative preconfigured performance value; when the minimum adjusted usage cost is less than the actual usage cost of the cloud disk during the historical time period, determining the alternative preconfigured performance value corresponding to the minimum adjusted usage cost as the target preconfigured performance value.Optionally, determining multiple alternative preconfigured performance values ​​between the preconfigured performance value currently configured for the cloud disk and the performance peak includes: selecting multiple alternative preconfigured performance values ​​in sequence according to a preset interval between the preconfigured performance value currently configured for the cloud disk and the performance peak; the preset interval is determined based on the difference between the performance peak and the preconfigured performance value currently configured for the cloud disk. Optionally, the multiple alternative preconfigured performance values ​​include a first alternative preconfigured performance value; when determining, based on the historical usage data, that the preconfigured performance value is adjusted to the first alternative preconfigured performance value, the adjusted usage cost of the cloud disk within the historical time period includes: when determining, based on the first alternative preconfigured performance value and the length of the historical time period, the adjusted preconfigured read and write performance cost of the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first alternative preconfigured performance value; when determining, based on the historical usage data, the adjusted number of burst reads and writes of the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first alternative preconfigured performance value; when determining, based on the adjusted burst read and write number, the adjusted burst read and write performance cost of the cloud disk within the historical time period; when determining, based on the adjusted preconfigured read and write performance cost and the adjusted burst read and write performance cost, the adjusted usage cost of the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first alternative preconfigured performance value. Optionally, determining, based on the historical usage data, the adjusted number of burst reads and writes of the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first alternative preconfigured performance value includes: determining, based on the historical usage data, the actual number of burst reads and writes and the proportion of actual burst duration of each time interval of the cloud disk in multiple time intervals; the multiple time intervals are obtained by dividing the historical time period; estimating, based on the difference between the preconfigured performance value currently configured for the cloud disk and the first alternative preconfigured performance value, the duration of each time interval, and the proportion of burst duration of the cloud disk in each time interval, the amount of burst reduction of the cloud disk in each time interval when the preconfigured performance value is adjusted to the first alternative preconfigured performance value; and determining, based on the actual number of burst reads and writes and the burst reduction of each time interval of the cloud disk in the multiple time intervals, the adjusted number of burst reads and writes of the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first alternative preconfigured performance value.Optionally, the method further includes: when the minimum adjusted usage cost is greater than or equal to the actual usage cost of the cloud disk during the historical time period, not sending cost optimization recommendation information to the user. Optionally, based on the historical usage data, predicting a target preconfigured performance value that satisfies a preset cost optimization condition for a future time period, further including: if the currently configured preconfigured performance value of the cloud disk is greater than the peak performance value, determining the peak performance value as the target preconfigured performance value that satisfies the preset cost optimization condition. Optionally, the preconfigured read / write performance includes preconfigured IOPS performance; and the burst read / write performance includes burst IOPS performance. A second aspect of the present disclosure provides a data processing method, comprising: obtaining historical usage data of a cloud disk; the cloud disk is configured with preconfigured read and write performance and burst read and write performance; the preconfigured read and write performance is billed based on the preconfigured performance value and billing duration; the burst read and write performance is billed based on the number of burst reads and writes of the cloud disk; the burst read and write number refers to the number of reads and writes involved in the portion where the actual used performance value exceeds the preconfigured performance value; and based on the historical usage data, predicting a target preconfigured performance value that meets a preset cost optimization condition. Optionally, predicting a target preconfigured performance value that satisfies a preset cost optimization condition based on the historical usage data includes: determining, based on the historical usage data, a peak performance value actually used by the cloud disk during a historical time period; when the preconfigured performance value currently configured for the cloud disk is less than the peak performance value, determining multiple alternative preconfigured performance values ​​between the preconfigured performance value currently configured for the cloud disk and the peak performance value; determining, for each alternative preconfigured performance value, based on the historical usage data, an adjusted usage cost of the cloud disk during the historical time period when the preconfigured performance value is adjusted to the alternative preconfigured performance value; and when the minimum adjusted usage cost is less than the actual usage cost of the cloud disk during the historical time period, determining the alternative preconfigured performance value corresponding to the minimum adjusted usage cost as the target preconfigured performance value. A third aspect of the present disclosure provides an electronic device. The electronic device includes: a memory and a processor, wherein the memory is configured to store a program; and the processor is coupled to the memory and configured to execute the program stored in the memory to implement any of the methods described above. According to a fourth aspect of the present disclosure, a computer-readable storage medium storing a computer program is provided, wherein the computer program, when executed by a computer, can implement any of the above-mentioned methods.In the technical solutions provided by the embodiments of the present disclosure, historical cloud disk usage data is obtained. Based on this historical cloud disk usage data, reasonable pre-configured performance values ​​for future time periods can be relatively accurately predicted and recommended to users. In this way, users can configure the pre-configured read and write performance of the cloud disk according to the recommended pre-configured performance values, thereby optimizing the user's cloud disk usage costs. In the technical solutions provided by the embodiments of the present disclosure, historical cloud disk usage data is obtained. Based on this historical cloud disk usage data, reasonable pre-configured performance values ​​for future time periods can be relatively accurately predicted. In this way, configuring the pre-configured read and write performance of the cloud disk according to the recommended pre-configured performance values ​​can optimize the user's cloud disk usage costs. BRIEF DESCRIPTION OF THE DRAWINGS To more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required in the embodiments or the prior art descriptions. Obviously, the drawings described below represent some embodiments of the present disclosure. Those skilled in the art can derive other drawings based on these drawings without inventive effort. Figure 1 is a flowchart of an information recommendation method provided in an embodiment of the present disclosure; Figure 2 is an example graph of a performance value curve required for cloud disk operation provided in an embodiment of the present disclosure; Figure 3 is an interactive diagram of the information recommendation method provided in an embodiment of the present disclosure; Figure 4 is a flowchart of a data processing method provided in an embodiment of the present disclosure; and Figure 5 is a block diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS To help those skilled in the art better understand the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present disclosure, and not all of them. All other embodiments derived by those skilled in the art based on the embodiments of the present disclosure without inventive effort are within the scope of protection of the present disclosure. In addition, some processes described in the specification, claims, and drawings of the present disclosure include multiple operations that appear in a specific order. These operations may be executed out of the order in which they appear or in parallel. Operation numbers, such as 101 and 102, are merely used to distinguish between different operations and do not represent any order of execution. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the terms "first" and "second" herein are used to distinguish between different messages, devices, modules, and the like, and do not imply a sequential order or limit the "first" and "second" to different types.It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, storage, and display) involved in this disclosure are all authorized by the user or fully authorized by all parties. The collection, use, and processing of relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or deny. First, the terms used in the embodiments of this disclosure are explained. It should be understood that this explanation is for a clearer understanding of the embodiments of this disclosure and does not necessarily constitute a limitation of the embodiments of this disclosure. IOPS (Input / Output Operations Per Second), that is, the number of read / write (IO) requests that can be processed per second, is one of the main values ​​​​for measuring storage performance. Preconfigured performance (i.e., preconfigured read and write performance): This refers to the performance preconfigured for the cloud disk. In the embodiments of this disclosure, preconfigured performance is the sum of the baseline performance provided by the cloud disk and the additional performance customized by the user on top of the baseline performance. The baseline performance provided by the cloud disk can be understood as the inherent performance of the cloud disk; this performance is fixed and cannot be customized by the user. Burst performance (i.e., burst read and write performance): This performance automatically increases as demand increases. The information recommendation method provided by the embodiments of this disclosure will be described below with reference to Figure 1. The execution entity of this method can be a server. The server can be a regular server, a cloud server, or a virtual server, and this is not specifically limited in this embodiment. As shown in Figure 1, the method includes:

[0004] 101. Obtain historical usage data for a cloud disk. The cloud disk is configured with preconfigured performance and burst performance. The preconfigured performance is billed based on the preconfigured performance value and billing duration. The burst performance is billed based on the number of burst reads and writes on the cloud disk. The burst read and write number refers to the number of reads and writes resulting from the portion of the actual performance value exceeding the preconfigured performance value. It should be noted that the performance value herein refers to the read and write performance value.

[0005] 102. Predict, based on the historical usage data, a target preconfigured performance value that meets a preset cost optimization condition within a future time period.

[0006] 103. Send cost optimization recommendation information to the user based on the target preconfigured performance value. In the above 101, in one example, the preconfigured performance may be preconfigured IOPS performance, and the burst performance may be burst IOPS performance. In another example, the preconfigured performance may be preconfigured throughput performance, and the burst performance may be burst throughput performance. Generally speaking, IOPS performance and throughput performance are related. Therefore, in actual applications, only one of them needs to be selected for billing. The historical usage data of the cloud disk may include: historical usage data of the cloud disk within a historical time period. The start and end points of the historical time period can be set according to actual needs and are not specifically limited in this embodiment of the present disclosure. Generally speaking, the closer the historical time period is to the current time period, the more the cloud disk usage data within that historical time period can reflect the future cloud disk usage. Therefore, the historical time period may be a recent historical period, such as the last month, the last week, the last three days, etc. In one example, historical usage data may include historical usage records, such as the actual performance value of the cloud disk used in each sub-time interval (e.g., per second), such as the actual IOPS used per second. Specifically, during the operation of the cloud disk, the current performance value of the cloud disk may be recorded at each sub-time interval. In another example, historical usage data may include processed data from the historical usage records. The number of burst reads and writes of the cloud disk in each time interval (e.g., per hour) during the historical period may be determined based on the historical usage records of the cloud disk and the pre-configured performance value currently configured for the cloud disk. The time interval is greater than the sub-time interval; specifically, the time interval is an integer multiple of the sub-time interval. The number of burst reads and writes of the cloud disk in each time interval is the sum of the number of burst reads and writes of the cloud disk in each sub-time interval within the time interval. The number of burst reads and writes of the cloud disk in a sub-time interval is the product of the burst performance value of the cloud disk in the sub-time interval and the duration of the sub-time interval. The burst performance value of the cloud disk in the sub-time interval is the portion by which the actual performance value of the cloud disk in the sub-time interval exceeds the pre-configured performance value currently configured for the cloud disk. It should be noted that if the actual performance value used by the cloud disk in a sub-interval does not exceed the pre-configured performance value of the cloud disk, the burst performance value of the cloud disk in that sub-interval is zero. For easier understanding, the following is an explanation with reference to Figure 2: Figure 2 shows the performance value curve used by the cloud disk. The horizontal axis represents time in seconds, and the vertical axis represents the performance value. The number of burst reads and writes per hour (i.e., time interval) of the cloud disk is the sum of the areas of regions H1 and H2 in Figure 2.For example, the vertical axis in FIG2 represents IOPS, and the pre-configured performance value is the pre-configured IOPS value. Furthermore, the peak performance value actually used by the cloud disk during a historical time period can be determined based on the historical usage record data of the cloud disk. The peak performance value is the maximum performance value actually used by the cloud disk during the historical time period. For example, the historical usage record data of the cloud disk includes: the actual performance value of the cloud disk used in the first second, the actual performance value of the cloud disk used in the second, ..., and the actual performance value of the cloud disk used in the nth second; where, , , ..., . Therefore, the actual performance peak of the cloud disk during the historical time period is y, where j is one of 1, 2, ..., and n. Optionally, the burst duration ratio of the cloud disk in each time interval during the historical time period can be determined based on the cloud disk's historical usage records. Exemplarily, the hourly burst duration ratio of the cloud disk refers to the ratio h / H between the cloud disk's burst duration h in that hour and the hour's duration H (i.e., 1 hour). Exemplarily, historical usage data may include: the number of burst reads and writes of the cloud disk in each time interval during the historical time period, the actual performance peak of the cloud disk during the historical time period, and the burst duration ratio of the cloud disk in each time interval during the historical time period. In one example, the preconfigured performance cost (i.e., the preconfigured read and write performance cost) is determined based on the preconfigured performance value and the billing duration. In actual applications, the preconfigured performance value is the sum of the basic performance value and the additional performance value customized by the user for the cloud disk. Therefore, the usage cost corresponding to the preconfigured performance is the sum of the basic performance cost and the additional performance cost. The basic performance cost is related to the billing duration and is unrelated to the preconfigured performance value. The additional performance cost is related to both the preconfigured performance value and the billing duration, specifically, the user-defined additional performance value for the cloud disk and the billing duration. In other words, the additional performance cost is determined based on the user-defined additional performance value for the cloud disk and the billing duration. Specifically, the additional performance cost can be determined as the product of the additional performance price, the additional performance value, and the billing duration. For example, the preconfigured performance cost of a cloud disk during a historical time period is the sum of the basic performance cost of the cloud disk during the historical time period and the additional performance cost of the cloud disk during the historical time period. The basic performance cost of the cloud disk during the historical time period is the product of the basic performance price and the duration of the historical time period (i.e., the billing duration). The additional performance cost of the cloud disk during the historical time period is the product of the additional performance price, the additional performance value, and the duration of the historical time period. The burst performance cost (also known as the burst read / write performance cost) is determined based on the burst performance price and the number of burst reads and writes on the cloud disk. Specifically, the product of the burst performance price and the number of burst reads and writes on the cloud disk can be used as the burst performance cost. For example, the burst performance cost of a cloud disk during a historical time period is the product of the burst performance price and the number of burst reads and writes on the cloud disk during that time period. The number of burst reads and writes on the cloud disk during that time period refers to the sum of the number of burst reads and writes on the cloud disk per sub-time interval (e.g., per second) or the sum of the number of burst reads and writes on the cloud disk per time interval (e.g., per hour) during that time period.The actual usage cost of a cloud disk during a historical period is the sum of the actual preconfigured performance cost of the cloud disk during that period and the actual burst performance cost of the cloud disk during that period. In step 102 above, the target usage cost of the cloud disk during that period is calculated based on historical usage data and the target preconfigured performance value. The target usage cost of the cloud disk during that period is less than the actual usage cost of the cloud disk during that period. The actual usage cost of the cloud disk during that period is calculated based on historical usage data and the preconfigured performance value currently configured for the cloud disk. In other words, step 102 seeks a target preconfigured performance value that can reduce the cost of cloud disk usage. The specific search method will be described in detail in the following embodiments. In step 103 above, in one example, the target preconfigured performance value can be recommended to cloud disk users. In another example, cost optimization recommendation information can be generated based on the difference between the target preconfigured performance value and the basic performance value (i.e., the target additional performance value) and sent to cloud disk users. For example, a recommendation message can be displayed on the user's cloud disk usage interface: "Cloud disk ***, you can optimize costs by configuring 5066 additional IOPS. Based on your usage over the past seven days, your costs are expected to decrease by 50%." "***" can be the name or number of the cloud disk. In actual applications, in addition to sending cost optimization recommendations to users through the cloud disk usage interface, other methods such as email and text messages can also be used, which are not specifically limited in this embodiment. The technical solution provided in this embodiment obtains historical cloud disk usage data; based on this historical cloud disk usage data, a reasonable pre-configured performance value for a future time period can be relatively accurately predicted and recommended to the user. In this way, the user can configure the cloud disk's pre-configured performance according to the recommended pre-configured performance value, thereby optimizing the user's cloud disk usage costs. In one feasible solution, the step "predicting a target pre-configured performance value that meets the preset cost optimization conditions based on the historical usage data" in step 102 above can be implemented using the following steps:

[0007] 1021. Determine, based on the historical usage data, a performance peak value of the cloud disk actually used during a historical period.

[0008] 1022. Determine multiple candidate pre-configured performance values ​​between the pre-configured performance value currently configured for the cloud disk and the performance peak value.

[0009] 1023. For each candidate pre-configured performance value, determine, based on the historical usage data, an adjusted usage cost of the cloud disk during the historical time period when the pre-configured performance value is adjusted to the candidate pre-configured performance value.

[0010] 1024. When the minimum adjusted usage cost is less than the actual usage cost of the cloud disk during the historical time period, determine the alternative preconfigured performance value corresponding to the minimum adjusted usage cost as the target preconfigured performance value. Regarding 1021 above, the specific method for determining the actual peak performance of the cloud disk during the historical time period can be found in the corresponding content of the above embodiments and will not be repeated here. Regarding 1022 above, in one specific example, multiple alternative preconfigured performance values ​​can be selected sequentially at a preset interval between the currently configured preconfigured performance value of the cloud disk and the performance peak. The preset interval can be preconfigured or determined based on the difference between the performance peak and the currently configured preconfigured performance value of the cloud disk. If the preset interval is preconfigured, its size can be set based on practical experience and is not specifically limited in this embodiment. If the preset interval is determined based on the difference between the performance peak and the currently configured preconfigured performance value of the cloud disk, a preset number of alternative preconfigured performance values ​​can be configured in advance, with the preset interval being the ratio of the difference to the preset number. That is, the number of alternative pre-configured performance values ​​is negatively correlated with the size of the preset spacing, that is, the smaller the preset spacing, the greater the number of alternative pre-configured performance values. In actual applications, the more alternative pre-configured performance values ​​there are, the greater the amount of calculation involved in this solution, that is, the more computing resources are occupied, and the usage cost corresponding to the target pre-configured performance value finally determined is closer to the minimum value, that is, the more accurate it is. Therefore, in actual applications, the size of the preset spacing can be flexibly configured according to actual needs. In the above 1023, the multiple alternative pre-configured performance values ​​include a first alternative pre-configured performance value; the first alternative pre-configured performance value refers to any one of the multiple alternative pre-configured performance values. In the above 1023, "when determining to adjust the pre-configured performance value to the first alternative pre-configured performance value based on the historical usage data, the adjusted usage cost of the cloud disk in the historical time period" can be implemented by the following steps:

[0011] 511. Determine, based on the first candidate preconfigured performance value and the length of the historical time period, an adjusted preconfigured performance cost (i.e., adjusted preconfigured read and write performance cost) of the cloud disk during the historical time period when the preconfigured performance value is adjusted to the first candidate preconfigured performance value.

[0012] 512. When determining, based on the historical usage data, to adjust the preconfigured performance value to the first candidate preconfigured performance value, the adjusted number of burst reads and writes of the cloud disk within the historical time period.

[0013] 513. Determine an adjusted burst performance cost (ie, an adjusted burst read / write performance cost) of the cloud disk in the historical time period according to the adjusted burst read / write number.

[0014] S14. Determine, based on the adjusted preconfigured performance cost and the adjusted burst performance cost, the adjusted usage cost of the cloud disk during the historical time period when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. In S11 above, the adjusted preconfigured performance cost of the cloud disk during the historical time period is determined based on the first candidate preconfigured performance value and the length of the historical time period. The difference between the first candidate preconfigured performance value and the basic performance value is the adjusted additional performance value. The adjusted preconfigured performance cost of the cloud disk during the historical time period is the sum of the basic performance cost of the cloud disk during the historical time period and the adjusted additional performance cost of the cloud disk during the historical time period. The adjusted additional performance cost of the cloud disk during the historical time period is the product of the adjusted additional performance value, the additional performance unit price, and the length of the historical time period. In S12 above, the adjusted burst read and write times of the cloud disk during the historical time period are determined based on the historical usage data when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. In one example, the calculation method of the adjusted burst read / write times can refer to the corresponding contents in the above embodiments. In another example, in order to reduce the amount of calculation, the following method can be used to calculate the adjusted burst read / write times:

[0015] S121. Determine, based on the historical usage data, the actual number of burst reads and writes and the actual proportion of burst duration in each of multiple time intervals for the cloud disk. The multiple time intervals are obtained by dividing the historical time periods.

[0016] S122. Based on the difference between the preconfigured performance value currently configured for the cloud disk and the first alternative preconfigured performance value, the duration of each time interval, and the proportion of the burst duration of the cloud disk in each time interval, estimate the reduction in bursts of the cloud disk in each time interval when the preconfigured performance value is adjusted to the first alternative preconfigured performance value (that is, the number of burst reads and writes of the cloud disk reduced in each time interval when the preconfigured performance value is adjusted to the first alternative preconfigured performance value).

[0017] S123: Determine, based on the actual burst read and write counts and burst reductions for each of the multiple time intervals, the adjusted burst read and write counts for the cloud disk within the historical time period when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. In S121 above, the calculation method for the actual burst read and write counts and the actual burst duration ratio for each of the multiple time intervals for the cloud disk can be found in the corresponding content of the above embodiments. The actual burst read and write counts and the actual burst duration ratio for each of the multiple time intervals for the cloud disk are determined based on the preconfigured performance value currently configured for the cloud disk. In S122 above, the multiple time intervals include the first time interval, which refers to any one of the multiple time intervals. The burst reduction ratio for the cloud disk in the first time interval when the preconfigured performance value is adjusted to the first candidate preconfigured performance value is determined by multiplying the difference between the currently configured preconfigured performance value for the cloud disk and the first candidate preconfigured performance value, the duration of the first time interval, and the burst duration ratio for the cloud disk in the first time interval. Here, the burst reduction amount is corrected using the burst duration ratio to ensure that the final estimated burst reduction amount is close to the actual burst reduction amount, thereby improving the accuracy of the estimated adjusted burst read / write times for the cloud disk during the historical time period. In S123 above, the difference between the cloud disk's burst read / write times in the first time interval and the burst reduction amount in the first time interval is determined as the adjusted burst read / write times for the cloud disk during the first time interval when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. The sum of the adjusted burst read / write times for each of multiple time intervals is determined as the adjusted burst read / write times for the cloud disk during the historical time period when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. In S13 above, the product of the adjusted burst read / write times and the burst performance unit price is determined as the adjusted burst performance cost for the cloud disk during the historical time period. In S14 above, the sum of the adjusted preconfigured performance cost of the cloud disk during the historical time period and the adjusted burst performance cost of the cloud disk during the historical time period is determined as the adjusted usage cost of the cloud disk during the historical time period when the preconfigured performance value is adjusted to the first candidate preconfigured performance value. The adjusted usage cost of the cloud disk during the historical time period when the preconfigured performance value is adjusted to another candidate preconfigured performance value can be determined in the same manner as above.Taking IOPS as an example, the following formulas (1) and (3) can be used to calculate the actual usage cost of the cloud disk in the historical time period, and the following formulas (2), (4), (5) and (6) can be used to calculate the adjusted usage cost of the cloud disk in the historical time period:.

[0018] E o = e + EQ (1) in,

[0019] £*o = % * p * T + a * bur ⑶ When pre-configuring IOPS values, the adjusted cost of the cloud disk within a historical week. e is the basic IOPS cost, which remains unchanged before and after the adjustment. Since only the relationship with the can is needed later, to reduce computational complexity, only formulas (3) to (6) can be calculated. The relationship between and the can is also the relationship between Eo and the can.

[0020] % is the additional IOPS value currently configured for the cloud disk, p is the additional IOPS price, T is the length of a historical week, a is the burst I / O volume (that is, the number of burst reads and writes) of the cloud disk in the historical week (that is, the historical time period); bur is the burst IOPS price (that is, the burst performance price); IOPS max The actual IOPS peak value (also known as performance peak) of the cloud disk in the past week; base is the basic IOPS value of the cloud disk; IOPSmax-base-x is the IOPS peak value and the current configuration of the cloud disk. The difference between the IOPS value (i.e., the alternative pre-configured performance value) and the pre-configured IOPS value (i.e., the pre-configured performance value) currently configured for the cloud disk; q is the adjusted burst I / O volume (adjusted burst read and write times) of the cloud disk in the historical week when the pre-configured IOPS value is adjusted to the i-th alternative pre-configured IOPS value; m is the total number of hours in the historical week (i.e., the total number of time intervals); % is the burst I / O volume of the cloud disk in the j-th hour; ratiOj is the percentage of the duration of the burst on the cloud disk in the j-th hour; t is the duration of an hour; n £ * t * ratiOj can be understood as the reduction in burst I / O of the cloud disk in the jth hour (i.e., the burst reduction amount) when the pre-configured IOPS value is adjusted to the i-th alternative pre-configured IOPS value; A- - n £* t * ratiOj can be understood as the adjusted burst I / O volume of the cloud disk in the jth hour when the preconfigured IOPS value is adjusted to the i-th alternative preconfigured IOPS value. It should be noted that in the above example, the difference between the IOPS peak and the preconfigured performance value currently configured for the cloud disk is divided into ten equal parts to obtain the preset interval. In the above 1024, when the minimum adjusted usage cost is less than the actual usage cost of the cloud disk during the historical time period, the alternative preconfigured performance value corresponding to the minimum adjusted usage cost is determined as the target preconfigured performance value. Optionally, the above method may also include:

[0021] 104. When the minimum adjusted usage cost is greater than or equal to the actual usage cost of the cloud disk during the historical time period, no cost optimization recommendation information is sent to the user. That is, if cost optimization is not achieved, no optimization recommendations need to be sent to the user. In actual applications, three situations may exist: the first is that the pre-configured performance value of the cloud disk is less than the peak performance value; the second is that the pre-configured performance value of the cloud disk is equal to the peak performance value; and the third is that the pre-configured performance value of the cloud disk is greater than the peak performance value. If the first situation occurs, the step of "determining multiple candidate pre-configured performance values ​​between the pre-configured performance value of the cloud disk and the peak performance value" in step 1022 above may be performed. That is, if the first situation occurs, steps 1022-1024 above are performed. If the second situation occurs, no steps are performed. If the third situation occurs, the peak performance value is determined as the target pre-configured performance value that meets the preset cost optimization condition. That is, the pre-configured performance value of the cloud disk is reduced to the peak performance value. The information recommendation method provided by the embodiment of the present disclosure will be described below in conjunction with FIG3 : The system architecture involved in the information recommendation method includes: a task scheduling platform 31, a cloud disk management and control service 32, a cloud disk metadata repository 33, a cloud disk hourly data statistics service 34, a cloud disk second-level data statistics service 35, a centralized service 36, and a cloud disk management and control service message module 321. The cloud disk management and control service message module 321 can be a functional module in the cloud disk management and control service 32. As shown in FIG3 , the information recommendation method includes the following steps:

[0022] 1. Triggering cost optimization offline analysis on a weekly basis. For example, the task scheduling platform 31 may send a cost optimization analysis instruction to the cloud disk management and control service 32 at 2:00 AM every Sunday.

[0023] 2. Pull cloud disk metadata. Cloud disk metadata may include: cloud disk name, current pre-configured performance value of the cloud disk, and burst performance activation status. Cloud disk management service 32 pulls cloud disk metadata from cloud disk metadata repository 33.

[0024] 3. Pull cloud disk metering data. The cloud disk management and control service 32 pulls cloud disk metering data from the cloud disk hourly data statistics service 34. This cloud disk metering data (i.e., historical cloud disk usage data) may include the number of burst reads and writes per hour for the cloud disk over the past week, also known as burst I / O volume. This number is pre-calculated by the cloud disk hourly data statistics service 34. This data can be directly obtained during offline cost optimization analysis, shortening the time required for cost optimization. 4. Pull cloud disk weekly performance peaks. The cloud disk management and control service 32 pulls cloud disk weekly performance peaks, such as the peak IOPS for the past week, from the cloud disk second-level data statistics service 35.

[0025] 5. Calculate the recommended configuration and generate an optimization event. The process of calculating the recommended configuration is also the process of finding the target pre-configured performance value that meets the preset cost optimization conditions. The specific implementation method can be found in the corresponding content of the above embodiments.

[0026] 6. Cost optimization event storage. The cloud disk management service 32 stores the optimization event (i.e., recommended information related to cost optimization) in the centralized service 36.

[0027] 7. Trigger the sending of optimization events on a weekly basis. The task scheduling platform can send a notification instruction to the cloud disk management service 32 at 8:00 am every Monday.

[0028] 8. Weekly Notification to Users. The cloud disk management and control service 32 can trigger the cloud disk management and control service message module 321 to send optimization events stored in the centralized service 36 to the corresponding users. Figure 4 shows a flow chart of the data processing method provided in an embodiment of the present disclosure. The execution entity of this method can be a server. The server can be a regular server, a cloud server, or a virtual server, and this embodiment of the present disclosure does not specifically limit this.

[0029] 401. Obtain historical usage data of a cloud disk. The cloud disk is configured with pre-configured performance and burst performance. The pre-configured performance is charged based on the pre-configured performance value and billing duration. The burst performance is charged based on the number of burst reads and writes of the cloud disk.

[0030] 402. Predict a target preconfigured performance value that satisfies preset cost optimization conditions based on the historical usage data. The specific implementation of steps 401 and 402 can be found in the corresponding sections of the aforementioned embodiments and will not be repeated here. In the embodiments of the present disclosure, after obtaining the target preconfigured performance value, cost optimization recommendations can be sent to the user based on the target preconfigured performance value, or the preconfigured performance value of the cloud disk can be automatically modified based on the target preconfigured performance value. It should be noted that automatically modifying the preconfigured performance value of the cloud disk based on the target preconfigured performance value is performed only if the user has previously granted the cloud vendor permission to automatically optimize the preconfigured performance value. In the technical solution provided by the embodiments of the present disclosure, historical cloud disk usage data is obtained; based on this historical cloud disk usage data, reasonable preconfigured performance values ​​for a future time period can be relatively accurately predicted. Thus, configuring the preconfigured performance of the cloud disk according to the recommended preconfigured performance value can optimize the user's cost of using the cloud disk. It should be noted that any steps not fully described in the method provided by the embodiments of the present disclosure can be found in the corresponding sections of the aforementioned embodiments and will not be repeated here. In addition to the above steps, the method provided in the embodiments of the present disclosure may also include some or all of the other steps in the above embodiments. For details, please refer to the corresponding content of the above embodiments and will not be repeated here. Figure 5 shows a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure. As shown in Figure 5, the electronic device includes a memory 1101 and a processor 1102. The memory 1101 can be configured to store various other data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device. The memory 1101 may be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), electrical programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.The memory 1101 is configured to store programs; the processor 1102, coupled to the memory 1101, is configured to execute the programs stored in the memory 1101 to implement the methods provided in the aforementioned method embodiments. Furthermore, as shown in FIG5 , the electronic device further includes: a communication component 1103, a display 1104, a power supply component 1105, an audio component 1106, and other components. FIG5 illustrates only some components, which does not imply that the electronic device includes only the components shown in FIG5 . Accordingly, embodiments of the present disclosure further provide a computer-readable storage medium storing a computer program. When executed by a computer, the computer program can implement the steps or functions of the methods provided in the aforementioned method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of these modules may be selected to achieve the objectives of the present embodiments according to actual needs. A person of ordinary skill in the art can understand and implement the present invention without inventive effort. Through the above description of the embodiments, a person of ordinary skill in the art can clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a ROM (Read Only Memory) / RAM (Random Access Memory), a magnetic disk, or an optical disk, and includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or portions thereof. Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present disclosure and are not intended to limit the present disclosure. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they may modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. However, such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present disclosure.

Claims

Claims 1. An information recommendation method, comprising: Obtain historical usage data of the cloud disk; the cloud disk is configured with pre-configured read / write performance and burst read / write performance; The pre-configured read / write performance is billed according to the pre-configured performance value and the billing duration; the burst read / write performance is billed according to the number of burst read / write operations of the cloud disk; the number of burst read / write operations refers to the number of read / write operations involved in the part where the actually used performance value exceeds the pre-configured performance value; according to the historical usage data, predict the target pre-configured performance value that meets the preset cost optimization condition; according to the target pre-configured performance value, send recommendation information about cost optimization to the user.

2. The method according to claim 1, wherein Predict the target pre-configured performance value that meets the preset cost optimization condition according to the historical usage data, including: according to the historical usage data, determine the actual performance peak used by the cloud disk during the historical time period; when the pre-configured performance value currently configured for the cloud disk is less than the performance peak, determine multiple alternative pre-configured performance values between the pre-configured performance value currently configured for the cloud disk and the performance peak; for each alternative pre-configured performance value, according to the historical usage data, determine the adjusted usage cost of the cloud disk during the historical time period when the pre-configured performance value is adjusted to this alternative pre-configured performance value; when the minimum adjusted usage cost is less than the actual usage cost of the cloud disk during the historical time period, determine the alternative pre-configured performance value corresponding to the minimum adjusted usage cost as the target pre-configured performance value.

3. The method according to claim 2, wherein, Determine multiple alternative pre-configured performance values between the pre-configured performance value currently configured for the cloud disk and the performance peak, including: between the pre-configured performance value currently configured for the cloud disk and the performance peak, sequentially select multiple alternative pre-configured performance values according to a preset interval; the preset interval is determined according to the difference between the performance peak and the pre-configured performance value currently configured for the cloud disk.

4. The method according to claim 2 or 3, wherein The multiple alternative preconfigured performance values include a first alternative preconfigured performance value; according to the historical usage data, when determining to adjust the preconfigured performance value to the first alternative preconfigured performance value, the adjusted usage cost of the cloud disk during the historical time period includes: according to the first alternative preconfigured performance value and the duration of the historical time period, determining the adjusted preconfigured read and write performance cost of the cloud disk during the historical time period when adjusting the preconfigured performance value to the first alternative preconfigured performance value; according to the historical usage data, determining the adjusted burst read and write times of the cloud disk during the historical time period when adjusting the preconfigured performance value to the first alternative preconfigured performance value; according to the adjusted burst read and write times, determining the adjusted burst read and write performance cost of the cloud disk during the historical time period; according to the adjusted preconfigured read and write performance cost and the adjusted burst read and write performance cost, determining the adjusted usage cost of the cloud disk during the historical time period when adjusting the preconfigured performance value to the first alternative preconfigured performance value. When determining to adjust the preconfigured performance value to the first alternative preconfigured performance value according to the historical usage data, the adjusted burst read and write times of the cloud disk during the historical time period include: according to the historical usage data, determining the actual burst read and write times and the proportion of the actual burst duration in each of the multiple time intervals of the cloud disk; the multiple time intervals are obtained by dividing the historical time period; according to the difference between the preconfigured performance value currently configured for the cloud disk and the first alternative preconfigured performance value, the duration of each time interval, and the proportion of the burst duration of the cloud disk in each time interval, estimating the burst reduction amount of the cloud disk in each time interval when adjusting the preconfigured performance value to the first alternative preconfigured performance value; according to the actual burst read and write times and the burst reduction amount of the cloud disk in each of the multiple time intervals, determining the adjusted burst read and write times of the cloud disk during the historical time period when adjusting the preconfigured performance value to the first alternative preconfigured performance value.

5. The method according to claim 4, wherein Also includes:

6. The method according to claim 2 or 3, wherein When the minimum adjusted usage cost is greater than or equal to the actual usage cost of the cloud disk during the historical time period, no recommendation information regarding cost optimization is sent to the user. According to the historical usage data, predicting the target preconfigured performance value that meets the preset cost optimization conditions in the future time period further includes: if the preconfigured performance value currently configured for the cloud disk is greater than the performance peak value, determining the performance peak value as the target preconfigured performance value that meets the preset cost optimization conditions.

7. The method according to claim 2 or 3, wherein The preconfigured read and write performance includes: preconfigured IOPS performance; the burst read and write performance includes: burst IOPS performance.

8. The method according to claim 2 or 3, wherein ​ 9. The method according to claim 2 or 3, wherein The preconfigured read / write performance includes: preconfigured throughput performance; the burst read / write performance includes: burst throughput performance.

10. The method according to claim 1, wherein The historical usage data includes at least one of the following: historical usage record data, data obtained by processing the historical usage record data.

11. The method according to claim 5, wherein The actual burst read / write times and the proportion of the actual burst duration are determined based on the preconfigured performance value of the current configuration of the cloud disk.

12. A data processing method, comprising: Obtain the historical usage data of the cloud disk; the cloud disk is configured with preconfigured read / write performance and burst read / write performance; The preconfigured read / write performance is billed according to the preconfigured performance value and the billing duration; the burst read / write performance is billed according to the burst read / write times of the cloud disk; the burst read / write times refer to the read / write times involved in the part where the actually used performance value exceeds the preconfigured performance value; according to the historical usage data, predict the target preconfigured performance value that meets the preset cost optimization condition.

13. The method according to claim 12, wherein, According to the historical usage data, predict to meet the preset cost 16 optimization condition of the target preconfigured performance value, including: according to the historical usage data, determine the performance peak actually used by the cloud disk in the historical time period; when the preconfigured performance value of the current configuration of the cloud disk is less than the performance peak, determine multiple alternative preconfigured performance values between the preconfigured performance value of the current configuration of the cloud disk and the performance peak; for each alternative preconfigured performance value, according to the historical usage data, determine the adjusted usage cost of the cloud disk in the historical time period when the preconfigured performance value is adjusted to this alternative preconfigured performance value; when the smallest adjusted usage cost is less than the actual usage cost of the cloud disk in the historical time period, determine the alternative preconfigured performance value corresponding to the smallest adjusted usage cost as the target preconfigured performance value.

14. An electronic device, comprising: A memory and a processor, wherein the memory is used to store a program; the processor is coupled to the memory and is used to execute the program stored in the memory to implement the method according to any one of claims 1 to 12.

15. A computer-readable storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the method according to any one of claims 1 to 12.

Images