WO2017054690A1

WO2017054690A1 - Method and device for detecting slow disk

Info

Publication number: WO2017054690A1
Application number: PCT/CN2016/100133
Authority: WO
Inventors: 熊睿之
Original assignee: 华为技术有限公司
Priority date: 2015-09-29
Filing date: 2016-09-26
Publication date: 2017-04-06
Also published as: CN106557389B; CN106557389A

Abstract

A method and a device for detecting a slow disk, relating to the technical field of computers, and being capable of solving the problem of low accuracy of slow disk detection. The method comprises: acquiring average service time, in a current pre-set period, for input/output (I/O) requests corresponding to disks of different types (101); acquiring, according to the relationship between the average service time for the I/O requests corresponding to the disks of different types and pre-set values, slow disk thresholds corresponding to the disks of different types in a next preset period (102); determining, according to the slow disk thresholds corresponding to disks of different types and the average service time for the I/O requests corresponding to each of the disks in the next pre-set period, a slow disk in the next pre-set period (103). The present solution is used for dynamically adjusting the slow disk thresholds.

Description

Slow disk detecting method and device

Technical field

The present invention relates to the field of computer technologies, and in particular, to a slow disk detection method and apparatus.

Background technique

A disk array system may include a plurality of Redundant Array of Independent Disks (RAID), and the RAID is a disk group (logic) formed by combining a plurality of independent hard disks (physical hard disks) in different manners. hard disk). By placing data on multiple hard drives, input and output operations can be balanced in a balanced manner, improving performance, providing higher storage performance and data backup technology than a single hard drive. At the same time, when storing data, the data is cut into a number of sections, which are stored on each hard disk.

For the disk array system, when the hard disk aging, head degradation, bad sectors of the hard disk, and the like occur in the system, the response speed of the hard disk is slowed down, and since the data is cut into multiple segments and stored on the respective hard disks, Therefore, the slow response of a hard disk will drag down the response speed of the entire system. Therefore, the input/output (I/O) response time of the hard disk needs to be monitored periodically or irregularly to respond to I/O. The time determines the slow disk, and then implements RAID reconfiguration, hard disk isolation, and the like for the slow disk.

At present, the method of detecting the slow disk is mostly to set a threshold. In a certain period of time or a plurality of identical time periods, if the average service time of the I/O request of the hard disk reaches the threshold, the hard disk is determined to be Slow disk. The average service time of the I/O request is the average of the n I/O response times sent in the time period. However, for different service types, the I/O request access pressure is different, and the same slow disk threshold cannot meet various service requirements. For different hard disk types, the I/O response time is also different, making the same slow disk. The threshold cannot adapt to different types of hard Disks and even hard disks have different degrees of aging after different service hours. If the same slow disk threshold is used, the real problematic hard disk cannot be accurately located. Therefore, the accuracy of this slow disk detection method is not accurate. low.

Summary of the invention

The embodiment of the invention provides a slow disk detection method and device, which can solve the problem of low accuracy of slow disk detection.

In a first aspect, a method for detecting a slow disk is provided, including:

Obtain an average service time of input and output I/O requests corresponding to different types of hard disks in the current preset period;

Obtaining a slow disk threshold corresponding to a different type of hard disk in the next preset period according to the relationship between the average service time of the I/O request corresponding to the type of the hard disk and the preset value;

The slow disk in the next preset period is determined according to the slow disk threshold corresponding to the different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period.

With reference to the first aspect, in the first possible implementation manner of the first aspect, the obtaining an average service time of the input/output I/O request corresponding to different types of hard disks in the current preset period includes:

Obtaining an average value of an average service time of an I/O request corresponding to each hard disk in the current preset period;

Obtaining a ratio of the number of I/O requests received by each hard disk in the current preset period to the total amount of I/O requests received by all the hard disks corresponding to each type of the hard disk;

Obtain an average service time of an I/O request corresponding to different types of hard disks in the current preset period according to the average value of each of the different types of hard disks and the ratio.

With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner of the first aspect, the obtaining an average service time of an I/O request corresponding to each hard disk in the current preset period The average value includes:

Obtaining a sum of response times of the I/O requests received by each of the hard disks in the current preset period;

Obtain an average service of the I/O request corresponding to each hard disk in the current preset period according to a ratio of the sum of the response times and the number of I/O requests corresponding to each hard disk in the current preset period. The average of the time.

In conjunction with the first possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the obtaining an average service time of an I/O request corresponding to each hard disk in the current preset period The average value includes:

Obtaining, by the sum of the average service time of the I/O request corresponding to each hard disk in the unit time within the current preset period;

And obtaining an average value of the average service time of the I/O request corresponding to each hard disk in the current preset period according to the ratio of the sum of the average service time and the number of the unit time.

In combination with the first possible implementation of the first aspect to the third possible implementation manner, in a fourth possible implementation manner of the first aspect, the method according to each of the different types of hard disks The average value and the ratio of the average service time for obtaining the I/O request corresponding to different types of hard disks in the current preset period include:

Weighting the average value of each of the different types of hard disks and the ratio to obtain an average service time of the I/O request corresponding to different types of hard disks in the current preset period;

If the average service time of the I/O request corresponding to any type of hard disk of different types of hard disks is represented as AvgT, then AvgT=X1*Z1+X2*Z2+...Xn*Zn, and X1 represents any of the types. The average value of the first hard disk in the hard disk, Z1 indicates the ratio corresponding to the first hard disk, X2 indicates the average value corresponding to the second hard disk in the hard disk of any type, and Z2 indicates the first The ratio corresponding to the two hard disks, Xn represents the average value corresponding to the nth hard disk in the hard disk of any type, and Zn represents the ratio corresponding to the nth hard disk.

With reference to the first aspect, in a fifth possible implementation manner of the first aspect, the average service time and the preset value of the I/O request according to different types of hard disks are The relationship between the slow disk thresholds corresponding to different types of hard disks in the next preset period is as follows:

According to the product of the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period is obtained.

With reference to the first aspect, in a sixth possible implementation manner of the first aspect, the method further includes:

During the first preset period after the storage array is powered on, the slow disks in different types of hard disks are determined according to the initial thresholds corresponding to different types of hard disks and the average service time of the I/O requests corresponding to each hard disk.

In a second aspect, a device for slow disk detection is provided, including:

The obtaining unit is configured to obtain an average service time of the input/output I/O request corresponding to different types of hard disks in the current preset period;

The obtaining unit is further configured to obtain, according to the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period;

And a determining unit, configured to determine, according to the slow disk threshold corresponding to the different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period, the slow disks in the next preset period.

With reference to the second aspect, in a first possible implementation manner of the second aspect, the acquiring unit is specifically configured to:

With reference to the first possible implementation of the second aspect, in a second possible implementation manner of the second aspect, the acquiring unit is specifically configured to:

Obtaining the I/O received by each of the hard disks in the current preset period The sum of the response times sought;

Obtain an average service of the I/O request corresponding to each hard disk in the current preset period according to a ratio of the sum of the response times to the number of I/O requests corresponding to each hard disk in the current preset period. The average of the time.

With reference to the first possible implementation of the second aspect, in a third possible implementation manner of the second aspect, the acquiring unit is specifically configured to:

The obtaining unit is specifically configured to:

With reference to the first possible implementation manner of the second aspect to the third possible implementation manner, in a fourth possible implementation manner of the second aspect, the acquiring unit is specifically configured to:

With reference to the second aspect, in a fifth possible implementation manner of the second aspect, the acquiring unit is specifically configured to:

In conjunction with the second aspect, in a sixth possible implementation of the second aspect, The determining unit is also used to:

The embodiment of the invention provides a slow disk detection method and device, which acquires an average service time of input and output I/O requests corresponding to different types of hard disks in a current preset period, and then according to I/O requests corresponding to different types of hard disks. The relationship between the average service time and the preset value is obtained by the slow disk threshold corresponding to different types of hard disks in the next preset period, and then according to the slow disk threshold corresponding to different types of hard disks and the I corresponding to each hard disk in the next preset period. The average service time of the /O request determines the slow disk in the next preset period. Therefore, the present invention can determine the slow disk of the next preset period according to the average service time of different types of hard disks in the current preset period. Threshold, which can make different slow disk thresholds for different types of hard disks due to different I/O service time, and also because I/O response time of different service types of hard disks is different, I/ of hard disks with different service years. The response time of O is also different, resulting in different average service time of the I/O of the hard disk, so that the difference of the slow disk threshold of the present invention is determined to be equivalent to the hard disk in the system. To compare, therefore, different types of differentiated slow disk thresholds of the present invention can also adapt to different service types and different service years of hard disks, and can solve slow disks according to different hard disk types, different service types, and different service years of hard disks. The problem of low accuracy is improved, and the accuracy of slow disk detection is improved.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only some of the present invention. For the embodiments, those skilled in the art can obtain other drawings according to the drawings without any creative work.

1 is a schematic flowchart of a slow disk detecting method according to an embodiment of the present invention;

2 is a schematic flowchart of a slow disk detecting method according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a slow disk detection apparatus according to an embodiment of the present invention.

FIG. 4 is a schematic diagram of a slow disk detection apparatus according to an embodiment of the present invention.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

The embodiment of the present invention provides a slow disk detection method. The execution body of the embodiment is a processing device for detecting a slow disk. The device may be implemented by using hardware and/or software. Preferably, the processing device may be configured to be loaded. On the array system of the hard disk, it is convenient to detect the hard disk in the array system, as shown in Figure 1, including:

101. Obtain an average service time of input and output I/O requests corresponding to different types of hard disks in the current preset period.

The device for slow disk detection may first determine the type of the hard disk in the array system. For example, the hard disk type may include Solid State Drives (SSD), Serial Attached Small Computer System Interface (SAS), and Near Line SAS (Near Line_SAS, NL_SAS) and Serial Advanced Technology Attachment (SATA), etc., and then obtain the average service time of I/O requests for each type of hard disk in the current preset period.

Generally speaking, for a single hard disk, the average service time of the I/O request refers to the average response time of multiple I/O requests received within 1 s of the single hard disk. The response time of the I/O request refers to the hard disk from the hard disk. The time until the end of the service is reached when the received I/O request arrives at the array system. After classifying the hard disks, according to the hard disk type, the average service time of the I/O requests of each type of hard disk in the current preset period is obtained. The preset period may be one hour or other values, which is not limited in this application.

102. Average service time and pre-payment according to I/O requests corresponding to different types of hard disks Set the value to obtain the slow disk threshold corresponding to different types of hard disks in the next preset period.

In the embodiment of the present invention, if the I/O response time of one or several hard disks is larger than the weighted average of the I/O response time of other hard disks in the array system, and the numerical value shows a multiple trend, then This hard disk or these hard disks is considered as a slow disk, that is, the concept of a slow disk is to make a horizontal comparison with all other hard disks.

Since the average service time of the I/O request of the hard disk is the average of the response time of the received I/O request per unit time, and the hard disk is classified, the determination of the slow disk threshold may be that different types of hard disks pass the weighting. The algorithm obtains the average service time of the I/O request corresponding to different types, that is, the weighted average value, so that when the average service time of the I/O request of the hard disk can be a multiple of the weighted average value of the hard disk of the type to which the hard disk belongs, Slow disk. Therefore, the slow disk threshold corresponding to the different types of hard disks in the next preset period may be obtained according to the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value. The product of the average service time of the /O request and the preset value, and obtains the slow disk threshold corresponding to different types of hard disks in the next preset period. The preset value here may be set to 10 according to experience, and may of course be other values, which is not limited in this application.

103. Determine, according to the slow disk threshold corresponding to different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period, the slow disks in the next preset period.

After the slow disk threshold of different types of hard disks according to the next preset period, in the next preset period, the average service time of the I/O request according to the fixed time of each hard disk may be slower than the hard disk of the type of the hard disk. The disk threshold is compared. If the average service time of the I/O request of any hard disk is greater than or equal to the slow disk threshold of the disk of the type to which it belongs, it can be determined that the hard disk is a slow disk. The certain time here refers to the average service time of the I/O request received by any hard disk in a certain period of time at a certain time in the next preset period. For example, the certain time is 5 minutes, and the preset period is one. hour.

The embodiment of the present invention provides a slow disk detection method, which obtains an average service time of input and output I/O requests corresponding to different types of hard disks in a current preset period, and then averages the I/O requests corresponding to different types of hard disks. The relationship between the service time and the preset value is used to obtain the slow disk threshold corresponding to different types of hard disks in the next preset period, and then according to different types. The slow disk threshold corresponding to the hard disk and the average service time of the I/O request corresponding to each hard disk in the next preset period determine the slow disk in the next preset period, and thus, the present invention can be different according to the current preset period. The average service time of the type of hard disk determines the slow disk threshold of the next preset period, so that different types of hard disks can be differentiated from the slow disk threshold due to the different service time of the I/O, and also the different services of the hard disk. The I/O response time of the type is different, and the response time of the I/O of the hard disk of different service years is also different, resulting in different average service time of the I/O of the hard disk, so that the difference of the slow disk threshold of the present invention is determined to be equivalent. Different types of differentiated slow disk thresholds of the present invention can also be adapted to hard disks of different service types and different service years, and can be hard disks according to different hard disk types, different service types, and different service years. The difference solves the problem of low accuracy of slow disk detection and improves the accuracy of slow disk detection.

The above embodiment will be specifically described below.

An embodiment of the present invention provides a slow disk detection method, as shown in FIG. 2, including:

201. Determine, in a first preset period after the storage array is powered on, a slow disk in a different type of hard disk according to an initial threshold corresponding to different types of hard disks and an average service time of an I/O request corresponding to each hard disk.

If the storage array is powered on, the slow disk threshold of different types of hard disks in the first preset period may be determined according to a preset initial threshold, which may be determined according to test data and current network service conditions, for example, a preset period, for example. Can be an hour or so. The average service time of the I/O request of any one of the hard disks may be an average of the response time of the I/O request of the hard disk for a period of time, and the time may be, for example, 5 minutes, that is, every 5 minutes. The average service time of the I/O request of any hard disk is compared with the initial threshold of the hard disk of the type to which it belongs. If it is greater than the initial threshold, it is determined to be a slow disk.

Then, the average service time of the input/output I/O request corresponding to the different types of hard disks in the current preset period can be obtained by the weighting calculation, which can be implemented by step 202 to step 204.

202. Obtain an average value of an average service time of an I/O request corresponding to each hard disk in the current preset period.

Specifically, when the critical time between the current preset period and the next preset period is obtained, The average value of the average service time of the I/O request corresponding to each hard disk in the pre-preset period, for example, in the first preset period and the critical time of the second preset period, the first preset period is acquired. The average of the average service time of the I/O requests for each hard disk.

For example, the sum of the response times of the I/O requests received by each hard disk in the current preset period may be obtained, and the number of I/O requests corresponding to each hard disk in the current preset period according to the sum of the response times The ratio of the average value Xn of the average service time of the I/O request corresponding to each hard disk in the current preset period is obtained.

That is, for a single hard disk, since the average service time of the I/O request is the average of the response times of the Z I/O requests received per unit time (for example, 1 s), then for a single hard disk, Counts the response time of M I/O requests received by a single hard disk in the current preset period (for example, one hour), and then adds the response time of each I/O request of M, adds the sum and divides by a single hard disk. The number of I/O requests received in the current preset period is the average value Xn of the average service time of the I/O requests corresponding to a single hard disk in the current preset period.

Optionally, the sum of the average service time of the I/O request corresponding to each hard disk in the current preset period is obtained, and the current preset is obtained according to the ratio of the average service time to the number of unit time. The average value Xn of the average service time of the I/O request corresponding to each hard disk in the cycle. For example, if the unit time is 1 s and the preset period is one hour, then for a single hard disk, the average service time of each I/O per 1 s can be added in one hour, and then divided by 3600 s to get the current preset. The average value Xn of the average service time of the I/O request corresponding to a single hard disk in the cycle.

203. Obtain a ratio of the number of I/O requests received by each hard disk in the current preset period to the total amount of I/O requests received by all the hard disks corresponding to each type of the hard disk.

Specifically, for a single hard disk, the number Yn of I/O requests received in the current preset period may be counted, and for each type of hard disk, the I received by all the hard disks in each type. The number of /O requests is added to obtain the total amount of I/O requests for each type of hard disk, and then, for one type of hard disk, the I/O received by the single hard disk in this type is received within the current preset period. The number of O requests Yn occupies the corresponding type of this type The ratio Zn of the total number of I/O requests received by the hard disk.

204. Obtain an average service time of an I/O request corresponding to different types of hard disks in the current preset period according to an average value and a ratio of each of the different types of hard disks.

Specifically, the average value and the ratio of each of the different types of hard disks may be weighted to obtain an average service time of the I/O request corresponding to different types of hard disks in the current preset period.

Wherein, if the average service time of the I/O request corresponding to any type of hard disk of different types of hard disks is expressed as AvgT, then AvgT=X ₁ *Z ₁ +X ₂ *Z ₂ +...X _n *Z _n , X ₁ represents an average value corresponding to the first hard disk of any one of the hard disks, Z ₁ represents a ratio corresponding to the first hard disk, and X ₂ represents a second hard disk of the hard disk of any type. Corresponding average value, Z ₂ represents the ratio corresponding to the second hard disk, X _n represents the average value corresponding to the nth hard disk in the hard disk of any type, and Z _n represents the corresponding nth hard disk. ratio.

Here, if a hard disk receives more I/O requests and its service is more busy, the hard disk with high I/O pressure has higher weight in the hard disk of its own type.

205. Acquire, according to the relationship between the average service time of the I/O request corresponding to the type of the hard disk and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period.

Specifically, according to the definition of the concept of the slow disk, the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value may be a product, and the preset value may be, for example, 10.

In the embodiment of the present invention, the determination of the preset value may also be determined according to the fault tolerance capability of the array system, and the higher the fault tolerance capability, the larger the preset value here. The preset period can be determined by the volatility of the I/O request of the array system. The greater the volatility, the longer the preset period can be set.

206. Determine a slow disk in the next preset period according to the slow disk threshold corresponding to different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period.

After determining the slow disk threshold corresponding to different types of hard disks according to the I/O request of the previous cycle, the slowest disk threshold corresponding to different types of hard disks may be determined in the next preset period. plate. Specifically, for any type of hard disk, the average service time of each I/O request of each hard disk may be based on any of the times Compare the slow disk thresholds of the type of hard disks, for example, calculate the average service time of a single hard disk every 5 minutes. If the average service time of the single hard disk determined by the 5 minutes is greater than or equal to the slow disk threshold of the hard disk of the type to which it belongs, then it is determined. A single hard disk is a slow disk.

207. Start slow disk data reconstruction and send alarm information.

If a hard disk is determined to be a slow disk, data of the slow disk can be obtained through data reconstruction, and the data of the slow disk is transferred to another idle disk, the slow disk is isolated, and an alarm message is sent to notify the administrator.

The embodiment of the present invention is to perform a horizontal comparison of the hard disks in the array system. Therefore, the response time of the I/O of the hard disk in the array system is different for different service types. Therefore, the slow disk detection mode in the embodiment of the present invention is different. It also adapts to different business types. Similarly, after the service life of the hard disk is increased, the hard disk in the array system is aged, and the response time of the I/O is slowed accordingly. In this embodiment of the present invention, the horizontal comparison method determines the slow disk and adapts different services. Years of hard drive.

The embodiment of the present invention provides a slow disk detection method, which obtains an average service time of input and output I/O requests corresponding to different types of hard disks in a current preset period, and then averages the I/O requests corresponding to different types of hard disks. The relationship between the service time and the preset value is used to obtain the slow disk threshold corresponding to different types of hard disks in the next preset period, and then according to the slow disk threshold corresponding to different types of hard disks and the I/O corresponding to each hard disk in the next preset period. The average service time of the request determines the slow disk in the next preset period. Therefore, the present invention can determine the slow disk threshold of the next preset period according to the average service time of different types of hard disks in the current preset period, so that Different types of hard disks have different slow disk thresholds due to different I/O service time, and I/O response time of hard disks with different service years due to different I/O response times of different service types of hard disks. Differently, the average service time of the I/O of the hard disk is different, so that the differential determination of the slow disk threshold of the present invention is equivalent to horizontally comparing the hard disks in the system. Therefore, different types of differentiated slow disk thresholds of the present invention can also be adapted to different service types and different service years of hard disks, and can solve slow disk detection accuracy according to different hard disk types, different service types, and different service years of hard disks. The low problem improves the accuracy of slow disk detection.

The embodiment of the present invention provides a device 3 for slow disk detection, as shown in FIG. 3, including:

The obtaining unit 301 is configured to obtain an average service time of the input/output I/O request corresponding to different types of hard disks in the current preset period;

The obtaining unit 301 is further configured to obtain, according to the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period;

The determining unit 302 is configured to determine a slow disk in the next preset period according to the slow disk threshold corresponding to the different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period.

Optionally, the obtaining unit 301 can be specifically configured to:

Obtain an average value of the average service time of the I/O request corresponding to each hard disk in the current preset period;

Obtain a ratio of the number of I/O requests received by each hard disk in the current preset period to the total amount of I/O requests received by all the hard disks corresponding to each type of the hard disk;

The average service time of I/O requests corresponding to different types of hard disks in the current preset period is obtained according to the average value and ratio of each hard disk in different types of hard disks.

Optionally, the obtaining unit 301 can be specifically configured to:

Obtaining the sum of response times of each hard disk corresponding to the received I/O request in the current preset period;

The average value of the average service time of the I/O request corresponding to each hard disk in the current preset period is obtained according to the ratio of the sum of the response times and the number of I/O requests corresponding to each hard disk in the current preset period.

Optionally, the obtaining unit 301 can be specifically configured to:

Obtaining the sum of the average service time of the I/O request corresponding to each hard disk in the unit time within the current preset period;

The average value of the average service time of the I/O request corresponding to each hard disk in the current preset period is obtained according to the ratio of the sum of the average service time and the number of unit time.

Optionally, the obtaining unit 301 can be specifically configured to:

Weighting the average value and ratio of each hard disk in different types of hard disks to obtain an average service time of I/O requests corresponding to different types of hard disks in the current preset period;

Optionally, the obtaining unit 301 can be specifically configured to:

Optionally, the determining unit 302 is further configured to:

During the first preset period after the storage array is powered on, the slow disks in different types of hard disks are determined according to the initial threshold corresponding to different types of hard disks and the average service time of the I/O requests corresponding to each hard disk.

The embodiment of the invention provides a device for detecting a slow disk, which obtains an average service time of an input/output I/O request corresponding to different types of hard disks in a current preset period, and then according to I/O requests corresponding to different types of hard disks. The relationship between the average service time and the preset value is used to obtain the slow disk threshold corresponding to different types of hard disks in the next preset period, and then according to the slow disk threshold corresponding to different types of hard disks and the I/ corresponding to each hard disk in the next preset period. The average service time of the O request determines the slow disk in the next preset period. Therefore, the present invention can determine the slow disk threshold of the next preset period according to the average service time of different types of hard disks in the current preset period. Different types of hard disks can be differentiated from the slow disk threshold due to the different service time of the I/O, and the I/O response time of the hard disk with different service years due to different I/O response times of different service types of the hard disk. The time is also different, resulting in different average service time of the I/O of the hard disk, so that the difference of the slow disk threshold of the present invention is determined to be equivalent to horizontal comparison of the hard disks in the system. Accordingly, the present invention is different types of slow disk difference threshold can be adapted to different service types and service life of the hard disk, the root can be According to different hard disk types, different service types, and different service years, the difference of hard disks solves the problem of low accuracy of slow disk detection, which improves the accuracy of slow disk detection.

The embodiment of the present invention provides a slow disk detecting device 4, as shown in FIG. 4, comprising: a memory 401, a processor 402, and a communication bus 403. The memory 401 is configured to store instructions and data, and the processor 402 is configured to execute the instruction for obtaining an average service time of input and output I/O requests corresponding to different types of hard disks in the current preset period; according to different types of hard disks. The relationship between the average service time of the corresponding I/O request and the preset value is obtained by the slow disk threshold corresponding to the different types of hard disks in the next preset period; according to the slow disk threshold corresponding to different types of hard disks and each of the next preset periods The average service time of the I/O request corresponding to each hard disk determines the slow disk in the next preset period. The data stored in the memory 401 includes an average service time of the input and output I/O requests corresponding to different types of hard disks in the current preset period, a preset value, and a slow disk threshold corresponding to different types of hard disks in the next preset period.

In the embodiment of the present invention, optionally, the processor 402 is configured to perform an average service time for obtaining input and output I/O requests corresponding to different types of hard disks in the current preset period, including:

The data stored by the memory 401 may also include the above average values and ratios.

In the embodiment of the present invention, optionally, the average value of the average service time of the processor 402 for performing the I/O request corresponding to each hard disk in the current preset period includes:

Obtain an average of the I/O requests corresponding to each hard disk in the current preset period according to the ratio of the sum of the response times and the number of I/O requests corresponding to each hard disk in the current preset period. The average value of the time.

In the embodiment of the present invention, the processor 402 is configured to perform an average of the I/O requests corresponding to different types of hard disks in the current preset period according to the average value and the ratio of each of the different types of hard disks. Service hours include:

In the embodiment of the present invention, the processor 402 is configured to perform, according to the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value, to obtain the hard disk corresponding to the different types of hard disks in the next preset period. The slow disk threshold includes:

In the embodiment of the present invention, optionally, the processor 402 is configured to execute the instruction, and may also be used to:

In the first preset period after the storage array is powered on, different types are determined according to the initial threshold corresponding to different types of hard disks and the average service time of I/O requests corresponding to each hard disk. A slow disk in a type of hard disk.

The embodiment of the invention provides a device for detecting a slow disk, which obtains an average service time of an input/output I/O request corresponding to different types of hard disks in a current preset period, and then according to I/O requests corresponding to different types of hard disks. The relationship between the average service time and the preset value is used to obtain the slow disk threshold corresponding to different types of hard disks in the next preset period, and then according to the slow disk threshold corresponding to different types of hard disks and the I/ corresponding to each hard disk in the next preset period. The average service time of the O request determines the slow disk in the next preset period. Therefore, the present invention can determine the slow disk threshold of the next preset period according to the average service time of different types of hard disks in the current preset period. Different types of hard disks can be differentiated from the slow disk threshold due to the different service time of the I/O, and the I/O response time of the hard disk with different service years due to different I/O response times of different service types of the hard disk. The time is also different, resulting in different average service time of the I/O of the hard disk, so that the difference of the slow disk threshold of the present invention is determined to be equivalent to horizontal comparison of the hard disks in the system. Therefore, different types of differentiated slow disk thresholds of the present invention can also adapt to different service types and different service years of hard disks, and can solve slow disk detection accuracy according to different hard disk types, different service types, and different service years of hard disks. The low problem improves the accuracy of slow disk detection.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, each functional unit in various embodiments of the present invention can be integrated into one process In the unit, each unit may be physically included separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

The above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium. The software functional units described above are stored in a storage medium and include instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform portions of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, and the program code can be stored. Medium.

It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not limited thereto; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that The technical solutions described in the foregoing embodiments are modified, or the equivalents of the technical features are replaced. The modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

A slow disk detecting method, comprising:

Obtain an average service time of input and output I/O requests corresponding to different types of hard disks in the current preset period;

Obtaining a slow disk threshold corresponding to a different type of hard disk in the next preset period according to the relationship between the average service time of the I/O request corresponding to the type of the hard disk and the preset value;

The slow disk in the next preset period is determined according to the slow disk threshold corresponding to the different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period.
The method according to claim 1, wherein the obtaining an average service time of the input/output I/O request corresponding to different types of hard disks in the current preset period comprises:

Obtaining an average value of an average service time of an I/O request corresponding to each hard disk in the current preset period;

Obtaining a ratio of the number of I/O requests received by each hard disk in the current preset period to the total amount of I/O requests received by all the hard disks corresponding to each type of the hard disk;

Obtain an average service time of an I/O request corresponding to different types of hard disks in the current preset period according to the average value of each of the different types of hard disks and the ratio.
The method according to claim 2, wherein the obtaining an average value of the average service time of the I/O request corresponding to each hard disk in the current preset period comprises:

Obtaining a sum of response times of the I/O requests received by each of the hard disks in the current preset period;

Obtain an average service of the I/O request corresponding to each hard disk in the current preset period according to a ratio of the sum of the response times to the number of I/O requests corresponding to each hard disk in the current preset period. The average of the time.
The method of claim 2 wherein said obtaining said The average of the average service time of the I/O requests for each hard disk in the previous preset period includes:

Obtaining, by the sum of the average service time of the I/O request corresponding to each hard disk in the unit time within the current preset period;

And obtaining an average value of the average service time of the I/O request corresponding to each hard disk in the current preset period according to the ratio of the sum of the average service time and the number of the unit time.
The method according to any one of claims 2-4, wherein the obtaining the different types of the current preset period according to the average value of each of the different types of hard disks and the ratio The average service time of the I/O request corresponding to the hard disk includes:

Weighting the average value of each of the different types of hard disks and the ratio to obtain an average service time of the I/O request corresponding to different types of hard disks in the current preset period;

Wherein, if the average service time of the I/O request corresponding to any type of hard disk of different types of hard disks is expressed as AvgT, then AvgT=X 1 *Z 1 +X 2 *Z 2 +...X n *Z n , X 1 represents an average value corresponding to the first hard disk of any one of the hard disks, Z 1 represents a ratio corresponding to the first hard disk, and X 2 represents a second hard disk of the hard disk of any type. Corresponding average value, Z 2 represents the ratio corresponding to the second hard disk, X n represents the average value corresponding to the nth hard disk in the hard disk of any type, and Z n represents the corresponding nth hard disk. ratio.
The method according to claim 1, wherein the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value is obtained, and the slow disk corresponding to the different types of hard disks in the next preset period is obtained. Thresholds include:

According to the product of the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period is obtained.
The method of claim 1 further comprising:

During the first preset period after the storage array is powered on, the slow disks in different types of hard disks are determined according to the initial thresholds corresponding to different types of hard disks and the average service time of the I/O requests corresponding to each hard disk.
A device for slow disk detection, comprising:

The obtaining unit is configured to obtain an average service time of the input/output I/O request corresponding to different types of hard disks in the current preset period;

The obtaining unit is further configured to obtain, according to the relationship between the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period;

And a determining unit, configured to determine, according to the slow disk threshold corresponding to the different types of hard disks and the average service time of the I/O request corresponding to each hard disk in the next preset period, the slow disks in the next preset period.
The device according to claim 8, wherein the obtaining unit is specifically configured to:

Obtaining an average value of an average service time of an I/O request corresponding to each hard disk in the current preset period;

Obtaining a ratio of the number of I/O requests received by each hard disk in the current preset period to the total amount of I/O requests received by all the hard disks corresponding to each type of the hard disk;

Obtain an average service time of an I/O request corresponding to different types of hard disks in the current preset period according to the average value of each of the different types of hard disks and the ratio.
The device according to claim 9, wherein the obtaining unit is specifically configured to:

Obtaining a sum of response times of the I/O requests received by each of the hard disks in the current preset period;

Obtain an average service of the I/O request corresponding to each hard disk in the current preset period according to a ratio of the sum of the response times to the number of I/O requests corresponding to each hard disk in the current preset period. The average of the time.
The device according to claim 9, wherein the obtaining unit is specifically configured to:

Obtaining, by the sum of the average service time of the I/O request corresponding to each hard disk in the unit time within the current preset period;

Obtaining according to the ratio of the sum of the average service time to the number of the unit time The average value of the average service time of the I/O request corresponding to each hard disk in the current preset period.
The device according to any one of claims 9-11, wherein the obtaining unit is specifically configured to:

Weighting the average value of each of the different types of hard disks and the ratio to obtain an average service time of the I/O request corresponding to different types of hard disks in the current preset period;

Wherein, if the average service time of the I/O request corresponding to any type of hard disk of different types of hard disks is expressed as AvgT, then AvgT=X 1 *Z 1 +X 2 *Z 2 +...X n *Z n , X 1 represents an average value corresponding to the first hard disk of any one of the hard disks, Z 1 represents a ratio corresponding to the first hard disk, and X 2 represents a second hard disk of the hard disk of any type. Corresponding average value, Z 2 represents the ratio corresponding to the second hard disk, X n represents the average value corresponding to the nth hard disk in the hard disk of any type, and Z n represents the corresponding nth hard disk. ratio.
The device according to claim 8, wherein the obtaining unit is specifically configured to:

According to the product of the average service time of the I/O request corresponding to the different types of hard disks and the preset value, the slow disk threshold corresponding to the different types of hard disks in the next preset period is obtained.
The device according to claim 8, wherein the determining unit is further configured to:

During the first preset period after the storage array is powered on, the slow disks in different types of hard disks are determined according to the initial thresholds corresponding to different types of hard disks and the average service time of the I/O requests corresponding to each hard disk.