CN111258705A - Method and device for detecting IO adjacent position interference of cloud hard disk input and output - Google Patents

Method and device for detecting IO adjacent position interference of cloud hard disk input and output Download PDF

Info

Publication number
CN111258705A
CN111258705A CN201811455270.6A CN201811455270A CN111258705A CN 111258705 A CN111258705 A CN 111258705A CN 201811455270 A CN201811455270 A CN 201811455270A CN 111258705 A CN111258705 A CN 111258705A
Authority
CN
China
Prior art keywords
virtual machine
request
hard disk
cloud hard
time difference
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201811455270.6A
Other languages
Chinese (zh)
Other versions
CN111258705B (en
Inventor
夏毅
孟宪杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201811455270.6A priority Critical patent/CN111258705B/en
Publication of CN111258705A publication Critical patent/CN111258705A/en
Application granted granted Critical
Publication of CN111258705B publication Critical patent/CN111258705B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45579—I/O management, e.g. providing access to device drivers or storage

Abstract

The application provides a method and a device for detecting IO adjacent-position interference of cloud hard disk input and output. According to the technical scheme, the parameters of the IO request are obtained through the virtual machine, the parameters comprise the block size of the IO request, the timestamp for sending the IO request and the timestamp for completing the IO request, and whether IO adjacent position interference occurs in the cloud hard disk of the virtual machine to which the parameters belong is judged according to the parameters. According to the technical scheme, IO adjacent position interference detection of the cloud hard disk of the virtual machine can be achieved through the virtual machine of the service layer, and the technical scheme is achieved on the virtual machine of the service layer, so that the technical scheme is independent of a cloud infrastructure layer.

Description

Method and device for detecting IO adjacent position interference of cloud hard disk input and output
Technical Field
The present application relates to the field of computers, and more particularly, to a method and an apparatus for detecting IO adjacent position interference of cloud hard disk input and output.
Background
Cloud computing is a new Information Technology (IT) infrastructure usage, delivery, and operation model. In a narrow cloud computing, namely, Infrastructure As A Service (IAAS), based on a virtualization technology, IT is configured to pool and manage IT infrastructures of a considerable scale, so that users can conveniently access and use the information technology through a network, and the system has the characteristics and advantages of rapid deployment, self-service use as required, charging as actual use and the like. Cloud computing can be classified into public clouds, private clouds, and hybrid clouds, depending on the provisioning mode.
In a public cloud environment, a physical storage system is managed in a centralized manner, virtualized into a cloud hard disk, and mounted on a virtual machine for use. The physical storage system of the public cloud is shared by a plurality of virtual machines, so that the cloud hard disk mounted on the virtual machine may be disturbed by the reading and writing of large-flow cloud hard disks of other virtual machines to cause performance degradation, and the borne application and service of the cloud hard disk are seriously affected.
Therefore, it is necessary to detect whether the cloud hard disk mounted on the virtual machine is subjected to proximity interference.
Disclosure of Invention
The application provides a method and a device for detecting IO adjacent-position interference of a cloud hard disk, which can realize IO adjacent-position interference detection of the cloud hard disk of a virtual machine in a service layer.
In a first aspect, the present application provides a method for detecting IO adjacent position interference of a cloud hard disk, where the method includes: the method comprises the steps that a first virtual machine obtains parameters of an IO request, wherein the parameters comprise the block size of the IO request, a timestamp for sending the IO request and a timestamp for finishing the IO request; and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the parameter.
In the above technical scheme, the first virtual machine can directly or indirectly express the read-write performance of the cloud hard disk by obtaining the block size of the IO request, the timestamp for sending the IO request and the timestamp for completing the IO request, so as to determine whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine. Therefore, the technical scheme can realize IO adjacent position interference detection of the cloud hard disk of the virtual machine through the virtual machine of the service layer.
In addition, the technical scheme is realized on a virtual machine of a business layer, so the technical scheme does not depend on a cloud infrastructure layer.
In some possible implementation manners, determining, by the first virtual machine, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to the parameter, includes: the first virtual machine determines a time difference between the timestamp of the send IO request and the timestamp of the complete IO request; the first virtual machine determines that the ratio of the time difference value to the block size of the IO request is unit sector processing delay; and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value.
According to the technical scheme, the unit sector processing time delay of the current IO request is determined according to the block size of the IO request, the time stamp of the IO request and the time stamp of the IO request, whether IO adjacent bit interference occurs in the cloud hard disk of the second virtual machine is judged according to the unit sector processing time delay of the current IO request, and the influence of the block size of the IO request on the IO request processing time delay at each time can be eliminated.
In some possible implementations, the unit sector processing delay reference value includes at least one of: the unit sector processing delay pre-estimated value is obtained by the first virtual machine according to the mapping relation among the block size of the IO request, the timestamp of the IO request and the unit sector processing delay pre-estimated value, the virtual machine set is composed of the second virtual machine and/or at least one third virtual machine, and the third virtual machine and the second virtual machine have the same application.
In the above technical solution, the accuracy of the detection result can be improved by comparing the unit sector processing delay with the historical minimum unit sector processing delay of the second virtual machine, the unit sector processing delay pre-estimated value, or the average value of the historical minimum unit sector processing delay of the at least one third virtual machine, or any combination thereof.
In some possible implementation manners, the determining, by the first virtual machine, whether IO adjacent-position interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to a deviation value of the unit sector processing delay with respect to the unit sector processing delay reference value includes: and when the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is greater than a first threshold value, the first virtual machine determines that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs.
In the above technical solution, a deviation value of the unit sector processing delay of the cloud hard disk of the second virtual machine with respect to the unit sector processing delay reference value is compared with a preset threshold, so that the detection accuracy can be flexibly adjusted by configuring the preset threshold.
In some possible implementation manners, when the deviation value of the unit sector processing delay with respect to the unit sector processing delay reference value is greater than a first threshold, the determining, by the first virtual machine, that IO adjacent bit interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs includes: and in a plurality of continuous monitoring periods, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs, wherein the deviation values of the unit sector processing delay relative to the unit sector processing delay reference value are all larger than the first threshold value.
In the above technical scheme, when the processing delay of the unit sector of the cloud hard disk of the second virtual machine continuously deviates from the unit sector processing delay reference value for many times and is far, the first virtual machine judges that the cloud hard disk of the second virtual machine has IO adjacent position interference, so that the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
In some possible implementation manners, determining, by the first virtual machine, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to the parameter, includes: the first virtual machine determines a time difference between the timestamp of the send IO request and the timestamp of the complete IO request; the first virtual machine determines a time difference pre-estimated value according to the block size of the IO request and the mapping relation between the block size of the IO request and the time difference; and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the time difference relative to the time difference estimated value.
In the technical scheme, only the time difference and the time difference estimated value between the timestamp for sending the IO request and the timestamp for completing the IO request need to be calculated, so that the calculation complexity can be reduced.
In some possible implementations, the determining, by the first virtual machine, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to a deviation value of the time difference from the time difference estimated value includes: and when the deviation value of the time difference relative to the time difference estimated value is larger than a second threshold value, the first virtual machine determines that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs.
In the above technical solution, a deviation value of the time difference of the cloud hard disk of the second virtual machine with respect to the time difference estimated value is compared with a preset threshold, so that the detection accuracy can be flexibly adjusted by configuring the preset threshold.
In some possible implementations, when the deviation value of the time difference from the time difference estimated value is greater than a second threshold, the determining, by the first virtual machine, that IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs includes: and determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs when the deviation value of the time difference relative to the time difference estimated value is greater than the second threshold value in a plurality of continuous monitoring periods.
In the above technical scheme, when the time difference of the cloud hard disk of the second virtual machine continuously deviates from the time difference estimated value for multiple times to a longer extent, the cloud hard disk of the second virtual machine is judged to have IO adjacent position interference, so that the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
In some possible implementations, when the unit sector processing latency is smaller than the historical unit sector processing latency, the method further includes: the first virtual machine updates historical minimum unit sector processing latency in a database.
In a second aspect, the present application provides an apparatus for detecting IO adjacent-position interference of cloud hard disk, where the apparatus includes: the acquisition module is used for acquiring parameters of the IO request, wherein the parameters comprise the block size of the IO request, a timestamp for sending the IO request and a timestamp for completing the IO request; and the processing module is used for determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the parameter.
In the above technical scheme, the first virtual machine can directly or indirectly express the read-write performance of the cloud hard disk by obtaining the block size of the IO request, the timestamp for sending the IO request and the timestamp for completing the IO request, so as to determine whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine. Therefore, the technical scheme can realize IO adjacent position interference detection of the cloud hard disk of the virtual machine through the virtual machine of the service layer.
In addition, the technical scheme is realized on a virtual machine of a business layer, so the technical scheme does not depend on a cloud infrastructure layer.
In some possible implementations, the processing module is specifically configured to: determining a time difference between the timestamp of the send IO request and the timestamp of the complete IO request; determining the ratio of the time difference to the block size of the IO request as unit sector processing delay; and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value.
According to the technical scheme, the unit sector processing time delay of the current IO request is determined according to the block size of the IO request, the time stamp of the IO request and the time stamp of the IO request, whether IO adjacent bit interference occurs in the cloud hard disk of the second virtual machine is judged according to the unit sector processing time delay of the current IO request, and the influence of the block size of the IO request on the IO request processing time delay at each time can be eliminated.
In some possible implementations, the unit sector processing delay reference value includes at least one of: the unit sector processing delay pre-estimated value is obtained by the first virtual machine according to the mapping relation among the block size of the IO request, the timestamp of the IO request and the unit sector processing delay pre-estimated value, the virtual machine set is composed of the second virtual machine and/or at least one third virtual machine, and the third virtual machine and the second virtual machine have the same application.
In the above technical solution, the accuracy of the detection result can be improved by comparing the unit sector processing delay with the historical minimum unit sector processing delay of the second virtual machine, the unit sector processing delay pre-estimated value, or the average value of the historical minimum unit sector processing delay of the at least one third virtual machine, or any combination thereof.
In some possible implementations, the processing module is specifically configured to: and when the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is larger than a first threshold value, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs.
In the above technical solution, a deviation value of the unit sector processing delay of the cloud hard disk of the second virtual machine with respect to the unit sector processing delay reference value is compared with a preset threshold, so that the detection accuracy can be flexibly adjusted by configuring the preset threshold.
In some possible implementations, the processing module is specifically configured to: and in a plurality of continuous monitoring periods, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs, wherein the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is greater than the first threshold value.
In the above technical scheme, when the processing delay of the unit sector of the cloud hard disk of the second virtual machine continuously deviates from the unit sector processing delay reference value for many times and is far, the first virtual machine judges that the cloud hard disk of the second virtual machine has IO adjacent position interference, so that the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
In some possible implementations, the processing module is specifically configured to: determining a time difference between the timestamp of the send IO request and the timestamp of the complete IO request; determining a time difference pre-estimated value according to the block size of the IO request and the mapping relation between the block size of the IO request and the time difference; and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the time difference relative to the time difference estimated value.
In the technical scheme, only the time difference and the time difference estimated value between the timestamp for sending the IO request and the timestamp for completing the IO request need to be calculated, so that the calculation complexity can be reduced.
In some possible implementations, the processing module is specifically configured to: and when the deviation value of the time difference relative to the time difference estimated value is greater than two threshold values, determining that IO adjacent position interference occurs on the cloud hard disk of the second virtual machine to which the parameter belongs.
In the above technical solution, a deviation value of the time difference of the cloud hard disk of the second virtual machine with respect to the time difference estimated value is compared with a preset threshold, so that the detection accuracy can be flexibly adjusted by configuring the preset threshold.
In some possible implementations, the processing module is specifically configured to: and determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs when the deviation value of the time difference relative to the time difference estimated value is greater than the second threshold value in a plurality of continuous monitoring periods.
In the above technical scheme, when the time difference of the cloud hard disk of the second virtual machine continuously deviates from the time difference estimated value for multiple times to a longer extent, the cloud hard disk of the second virtual machine is judged to have IO adjacent position interference, so that the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
In some possible implementations, the apparatus further includes: and the storage module is used for updating the historical minimum unit sector processing time delay in the database when the unit sector processing time delay is smaller than the historical unit sector processing time delay.
In a third aspect, the present application provides a device for detecting IO adjacent position interference in a cloud hard disk, where a virtual machine is deployed on the device, and the virtual machine may implement functions corresponding to each step in the method according to the first aspect, where the functions may be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more units or modules corresponding to the above functions.
In some possible implementations, the apparatus includes a processor configured to enable the apparatus to perform the corresponding functions in the method according to the first aspect. The apparatus may also include a memory, coupled to the processor, that retains program instructions and data necessary for the apparatus. Optionally, the apparatus further comprises a communication interface for supporting communication between the apparatus and other network elements.
In a fourth aspect, the present application provides a computer-readable storage medium, which includes instructions that, when executed by a processing module or a processor, cause an apparatus for detecting cloud hard disk input/output IO proximity interference to perform the method of the first aspect or any one of the implementation manners of the first aspect.
In a fifth aspect, the present application provides a chip, where instructions are stored, and when the instructions are executed on an apparatus for detecting IO adjacent position interference in a cloud hard disk, the chip is caused to perform the method according to the first aspect or any one of the implementation manners of the first aspect.
In a sixth aspect, the present application provides a computer program product, which when running on an apparatus for detecting cloud hard disk input/output IO proximity interference, causes the apparatus for detecting cloud hard disk input/output IO proximity interference to execute the method of the first aspect or any one of the implementation manners of the first aspect.
Drawings
Fig. 1 is a schematic diagram of IO proximity interference of a cloud hard disk.
Fig. 2 is a schematic flowchart of a method for detecting IO adjacent position interference of a cloud hard disk according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of an apparatus for detecting IO adjacent-position interference of a cloud hard disk according to an embodiment of the present application.
Fig. 4 is a schematic structural diagram of an apparatus for detecting IO proximity interference of a cloud disk according to another embodiment of the present application.
Fig. 5 is a schematic configuration diagram of a detection system according to an embodiment of the present application.
Detailed Description
In order to facilitate understanding of the technical solutions of the present application, first, concepts related to the present application are briefly introduced.
Virtualization technology: a plurality of independent operating systems are simultaneously operated on one physical hardware machine, the independently operated operating system resources come from the same bottom layer platform resource, and a virtual machine monitor at the bottom layer is responsible for the allocation of the system resources, the scheduling of the virtual machines, the communication among the virtual machines and the communication between the virtual machines and the outside. The virtualization technology redefines and divides IT resources by using a software method, and can realize dynamic allocation, flexible scheduling, cross-domain sharing and improvement of Information Technology (IT) resource utilization rate of the IT resources.
Cloud computing: is a novel IT infrastructure use, delivery and operation mode. Narrow cloud computing, that is, Infrastructure As A Service (IAAS), is based on a virtualization technology, and pools and manages IT infrastructures of a considerable scale, so that users can conveniently access and use the information technology through a network, and the system has the characteristics and advantages of rapid deployment, self-help on-demand use, charging according to actual use and the like. Cloud computing can be classified into public clouds, private clouds, and hybrid clouds according to the provisioning mode.
Cloud hard disk: in a cloud computing environment, a physical storage system is centrally managed and virtualized into a cloud hard disk, and the cloud hard disk is mounted on a virtual machine for use. In the aspect of evaluating the read-write performance, the cloud disk is the same as a common physical disk, and is measured by IO throughput (i.e., number of bytes per second) and Input Output Per Second (IOPS).
Virtual Machine (VM): the virtual device is simulated on the physical device through virtual machine software. For applications running in virtual machines, which operate as true physical devices, the virtual machines may have installed thereon a guest Operating System (OS) and applications, and may also have access to network resources.
Input Output (IO) adjacent bit interference of the cloud hard disk: when the virtual machine accesses data on the cloud hard disk, the virtual machine actually needs to finally access the remote shared physical storage hard disk through the network. On this path, there may be a resource conflict as shown in fig. 1, which eventually results in a reduction in the read-write performance of the cloud hard disk.
1. Network card of the physical machine: because a plurality of virtual machines may be deployed on one physical machine at the same time, each virtual machine may mount a plurality of cloud hard disks, and then the access of the cloud hard disks needs to pass through the network card of the physical machine, when the access flow of accessing the cloud hard disks exceeds the throughput performance of the network card of the physical machine, the access performance of the cloud hard disks is affected.
2. A data center network: data center network equipment needs to support read-write access of all virtual machines in a data center to a cloud hard disk, and when traffic is sudden (for example, virtual machine migration is performed in a large area), congestion of a data center network may be caused, and finally access performance of the cloud hard disk is affected.
3. Physical hard disk: one physical hard disk may be virtualized into a plurality of cloud hard disks and mounted to a plurality of virtual machines, so that when the virtual machines collectively perform read-write access on the cloud hard disks on the same physical hard disk, the access performance may also be reduced.
For a virtual machine, the performance of a cloud hard disk mounted on the virtual machine is influenced by the three possible performance influences caused by the reading and writing of large-flow cloud hard disks of other virtual machines, which are collectively referred to as IO adjacent position interference of the cloud hard disk in the application.
The cloud hard disk has reduced access performance due to IO adjacent position interference, and the loaded application and service of the cloud hard disk can be seriously influenced. Therefore, it is necessary to detect whether the cloud hard disk mounted on the virtual machine is subjected to proximity interference.
The embodiment of the application provides a method for detecting IO adjacent position interference of a cloud hard disk, when storage intensive application is operated on a virtual machine, the IO adjacent position interference detection of the cloud hard disk of the virtual machine can be realized on a service layer through the method, and then countermeasures and evasive measures are taken to avoid the influence of the countermeasures and the evasive measures on the application and the service.
Fig. 2 is a schematic flowchart of a method for detecting IO adjacent position interference of a cloud hard disk according to an embodiment of the present application. The method of fig. 2 may be applied in a public cloud, private cloud, or hybrid cloud environment, and executed by a business layer virtual machine. As shown in fig. 2, a method for detecting IO adjacent position interference of a cloud hard disk according to an embodiment of the present application may include at least some of the following contents.
At 210, the first virtual machine obtains parameters of the IO request, where the parameters include a block size of the IO request, a timestamp of sending the IO request, and a timestamp of completing the IO request.
In 220, the first virtual machine determines whether IO proximity interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the parameter.
The first virtual machine may be a control virtual machine which is deployed on a business layer and is specially used for detecting IO adjacent-position interference of the cloud hard disk, or may be a business virtual machine having a function of detecting IO adjacent-position interference of the cloud hard disk, which is not limited in this embodiment of the present application.
The second virtual machine may be any virtual machine that is mounted with a cloud hard disk and can acquire and report IO request parameters, and the specific type of the second virtual machine is not limited in the embodiment of the present application.
It should be understood that in some cases, the first virtual machine and the second virtual machine may be the same virtual machine.
In the above technical scheme, the first virtual machine can directly or indirectly express the read-write performance of the cloud hard disk by obtaining the block size of the IO request, the timestamp for sending the IO request and the timestamp for completing the IO request, so as to determine whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine. Therefore, the technical scheme can realize IO adjacent position interference detection of the cloud hard disk of the virtual machine through the virtual machine of the service layer.
In addition, the technical scheme is realized on a virtual machine of a business layer, so the technical scheme does not depend on a cloud infrastructure layer.
Optionally, the above technical solution may also be executed by a physical device.
Optionally, the IO parameter obtained by the first virtual machine may also be another parameter related to the performance of the cloud hard disk.
In some embodiments, the second virtual machine collects parameters of the IO request and reports the parameters to the first virtual machine, where the first virtual machine and the second virtual machine are different virtual machines. Optionally, the second virtual machine requests to acquire parameters for all the IO requests, and reports the acquired parameters. Optionally, the second virtual machine samples and collects parameters of the IO request, and reports the collected parameters, where the sampling interval may be configured, that is, whether IO adjacent-position interference occurs in the second virtual machine is periodically monitored.
In other embodiments, the first virtual machine collects parameters of its own IO request, and determines whether IO adjacent bit interference occurs in itself, where the first virtual machine and the second virtual machine are the same virtual machine. Optionally, the first virtual machine acquires parameters for all IO requests. Optionally, the first virtual machine samples and collects parameters of the IO request, and a sampling interval of the first virtual machine may be configurable.
Specifically, as one example, the second virtual machine or the first virtual machine may employ the blktrace tool of linux to collect parameters of the IO request. As another example, the second virtual machine or the first virtual machine may use a flexible ioster tool to collect parameters of the IO request.
The IO parameters acquired by the first virtual machine may include a block size of the IO request, a timestamp of sending the IO request, and a timestamp of completing the IO request. The block size of the IO request may represent the number of bytes of the IO request, and the unit may be the number of sectors; the timestamp for sending the IO request is the timestamp for sending the IO request to the cloud hard disk drive of the second virtual machine; the timestamp of the completion of the IO request is a timestamp of the completion of the data IO request.
And the first virtual machine determines whether IO adjacent bit interference occurs in the cloud hard disk of the second virtual machine to which the parameters belong according to the parameters of the IO request.
In some embodiments, the first virtual machine calculates a unit sector processing delay of the current IO request according to the block size of the IO request, the timestamp for sending the IO request, and the timestamp for completing the IO request, and determines whether IO adjacent-position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to a deviation value of the unit sector processing delay obtained by the calculation with respect to a unit sector processing delay reference value.
Specifically, the first virtual machine determines a time difference between a timestamp for sending the IO request and a timestamp for completing the IO request, that is, a time delay for completing the IO request by the cloud hard disk. For example, the time difference between the timestamp of the sending IO request and the timestamp of the completing IO request can be calculated using the following formula.
D2C_t＝Complete_ts-Driver_ts
D2C _ t is a time delay from sending to driving until completion of each IO request, that is, a time delay for completing an IO request by a cloud hard disk, Complete _ ts is a timestamp for completing an IO request, and Driver _ ts is a timestamp for sending an IO request.
Further, the first virtual machine determines that the ratio of the time difference value to the block size of the IO request is the unit sector processing delay. For example, the unit sector processing delay of the current IO request can be calculated using the following formula.
Delay_p_Sector＝D2C_t/IO_size
The Delay _ p _ Sector is a unit Sector processing time Delay, the D2C _ t is a time Delay for the cloud hard disk to complete the IO request, and the IO _ size is a block size of the IO request.
Optionally, the first virtual machine may calculate to obtain the unit sector processing delay according to an average value of block sizes of the multiple IO requests obtained within a preset time period and an average value of the multiple D2C _ t calculated within the preset time period.
Delay_p_Sector＝D2C_t’/IO_size’
The Delay _ p _ Sector is a unit Sector processing time Delay, the D2C _ t 'is an average value of time delays of the cloud hard disk to complete the IO request, and the IO _ size' is an average value of block sizes of the IO request.
Optionally, the first virtual machine may obtain a concentrated region (the concentrated region may be configured, for example, 90% duty ratio) according to an average value of block sizes of a plurality of IO requests obtained within a preset time period and an average value of a plurality of D2C _ t obtained by calculation within the preset time period, and calculate the unit sector processing delay. This avoids the effect of data introduced by the tool or other extreme.
In some embodiments, the unit sector processing latency reference value may be a minimum unit sector processing latency of the history of the second virtual machine recorded by the first virtual machine. The first virtual machine periodically monitors the unit sector processing time delay of the second virtual machine and records the historical minimum unit sector processing time delay. For example, the minimum unit sector processing delay can be recorded as follows.
Min_Delay_p_Sector＝Min(Delay_p_Sector1…n)
Wherein Min _ Delay _ p _ Sector is history unit Sector processing time Delay, and Delay _ p _ Sector1…nRepresenting the n unit sector processing delays.
In other embodiments, the unit sector processing delay reference value may be a unit sector processing delay estimate. The unit sector processing delay pre-estimated value may be obtained by the first virtual machine according to the block size of the IO request, the timestamp of the IO request, and the mapping relationship between the unit sector processing delay pre-estimated value and the timestamp of the IO request.
Alternatively, the mapping relationship may be fitted by a machine learning algorithm. Specifically, the following relationship may be trained by comprehensively using data collected on the distributed application virtual machine to determine the algorithm parameter w0、w1、w2、w3. And after the algorithm is trained, deploying the cloud hard disk to a first virtual machine, obtaining a unit sector processing delay pre-estimated value by using a fitted relation for IO _ size, Complete _ ts and Driver _ ts obtained each time, and further judging whether IO adjacent position interference occurs in the cloud hard disk of a second virtual machine according to a deviation value of the unit sector processing delay obtained through calculation relative to the unit sector processing delay pre-estimated value.
Delay_p_Sector＝w3*IO_size+w2*Complete_ts+w1*Driver_ts+w0
Wherein, w0、w1、w2、w3Are algorithm parameters.
The training of the machine learning algorithm may be performed in a development environment (for example, a private cloud or public cloud simulation environment in which no IO proximity interference occurs), or may be performed in a public cloud production environment (assuming that IO proximity interference in a public cloud environment is a small-probability event), and the embodiment of the present application is not particularly limited.
It should be understood that more parameter types, such as the number of cloud hard disks, the traffic of the storage network interface, and the like, may also be considered in the embodiments of the present application.
It should also be understood that the machine learning algorithm is only an example, and the mapping relationship may also be obtained by fitting other algorithms having the same function as the machine learning algorithm, for example, a depth algorithm, a neural network algorithm, and the like, and the embodiment of the present application is not particularly limited.
In other embodiments, the unit sector processing latency reference value may be an average value of historical minimum unit sector processing latencies of each virtual machine in a set of virtual machines, the set of virtual machines being composed of the second virtual machine and/or at least one third virtual machine, the third virtual machine having the same application as the second virtual machine. That is to say, the first virtual machine may record the historical minimum unit sector processing delay of each virtual machine to be detected, and when IO adjacent position interference detection is performed on a second virtual machine in the multiple virtual machines to be detected, the average value of the historical minimum unit sector processing delays of the virtual machines to be detected having the same application as the second virtual machine may be considered for comparison, and the average value of the historical minimum unit sector processing delays of the second virtual machine and at least one third virtual machine may also be considered for comparison.
In other embodiments, the unit sector processing delay reference value is any combination (e.g., pairwise combination, three-in-one combination, etc.) of the historical minimum unit sector processing delay of the second virtual machine, the unit sector processing delay pre-estimated value, and an average value of the historical minimum unit sector processing delay of each virtual machine in a set of virtual machines, the set of virtual machines being composed of the second virtual machine and/or at least one third virtual machine, the third virtual machine having the same application as the second virtual machine.
That is to say, the first virtual machine may determine whether IO adjacent-position interference occurs in the cloud hard disk of the second virtual machine according to any one of the historical minimum unit sector processing delay, the unit sector processing delay pre-estimated value, and the average value of the historical minimum unit sector processing delay of each virtual machine in the virtual machine set, or may determine whether IO adjacent-position interference occurs in the cloud hard disk of the second virtual machine according to any combination of the historical minimum unit sector processing delay, the unit sector processing delay pre-estimated value, and the average value of the historical minimum unit sector processing delay of each virtual machine in the virtual machine set.
In some embodiments, when the deviation value of the unit sector processing delay with respect to the unit sector processing delay reference value is greater than a first threshold (e.g., 50%, where the first threshold is configurable), the first virtual machine determines that IO proximity interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs. Therefore, the detection precision can be flexibly adjusted by configuring the preset threshold.
For example, when the unit sector processing delay exceeds 40% of the historical minimum unit sector processing delay in a rising manner, the first virtual machine determines that IO adjacent position interference occurs on the cloud hard disk of the second virtual machine.
In some embodiments, the first virtual machine periodically monitors the unit sector processing delay of the second virtual machine, and in a plurality of consecutive monitoring periods (for example, 3 monitoring periods, where the number of consecutive monitoring periods may be configured), the deviation values of the unit sector processing delay with respect to the unit sector processing delay reference value are all greater than the first threshold value, and it is determined that IO proximity interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs. Therefore, the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
In some embodiments, the first virtual machine determines the time difference estimated value according to a time difference between a time stamp for sending the IO request and a time stamp for completing the IO request (that is, a time delay for completing the IO request by the cloud hard disk), and according to a mapping relationship between a block size of the IO request and the time difference, and further determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine according to a deviation value of the calculated time difference with respect to the time difference estimated value. The calculation method of the time difference (i.e., the delay of the cloud hard disk completing the IO request) may refer to the above description, and is not described herein again.
Alternatively, the mapping relationship between the block size of the IO request and the time difference may be obtained by fitting a machine learning algorithm. Specifically, the following relationship may be trained by comprehensively using data collected on the distributed application virtual machine to determine the algorithm parameter w0、w1. After the algorithm is trained, deploying the algorithm to a first virtual machine, obtaining a time difference estimated value by using a fitted relation for the IO _ size obtained each time, and further judging whether IO adjacent position interference occurs in a cloud hard disk of a second virtual machine according to a deviation value of the time difference obtained through calculation relative to the time difference estimated value.
D2C_t＝w1*IO_size+w0
Wherein, w0、w1Are algorithm parameters.
The training of the machine learning algorithm may be performed in a development environment (for example, a private cloud or public cloud simulation environment in which no IO proximity interference occurs), or may be performed in a public cloud production environment (assuming that IO proximity interference in a public cloud environment is a small-probability event), and the embodiment of the present application is not particularly limited.
It should be understood that more parameter types, such as the number of cloud hard disks, the traffic of the storage network interface, and the like, may also be considered in the embodiments of the present application.
It should also be understood that the machine learning algorithm is only an example, and the mapping relationship may also be obtained by fitting other algorithms having the same function as the machine learning algorithm, for example, a depth algorithm, a neural network algorithm, and the like, and the embodiment of the present application is not particularly limited.
In some embodiments, when the deviation value of the time difference from the estimated time difference value is greater than a second threshold (e.g., 50%, the second threshold is configurable), the first virtual machine determines that IO neighbor interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs. Therefore, the detection precision can be flexibly adjusted by configuring the preset threshold.
For example, when the time difference exceeds the rise of the estimated time difference value by more than 30%, the first virtual machine determines that the cloud hard disk of the second virtual machine has IO proximity interference.
In some embodiments, the first virtual machine periodically monitors the time difference of the second virtual machine, and in a plurality of consecutive monitoring periods (for example, 3 monitoring periods, the number of consecutive monitoring periods may be configured), the deviation value of the time difference relative to the time difference estimated value is greater than two thresholds, and it is determined that IO proximity interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs. Therefore, the influence of accidental factors on the detection result can be avoided, and the accuracy of the detection result is improved.
Examples of the method for detecting IO adjacent position interference of a cloud hard disk provided by the present application are described above in detail. It is understood that, in order to implement the above functions, the apparatus for detecting IO adjacent position interference of a cloud hard disk includes a hardware structure and/or a software module corresponding to the execution of each function. Those of skill in the art would readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
According to the method, the device for detecting the IO adjacent position interference of the cloud hard disk can be divided into the functional units, for example, each function can be divided into the functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be realized in a hardware form, and can also be realized in a software functional unit form. It should be noted that the division of the units in the present application is schematic, and is only one division of logic functions, and there may be another division manner in actual implementation.
Fig. 3 is a schematic structural diagram of an apparatus for detecting IO adjacent-position interference of a cloud hard disk according to an embodiment of the present application. As shown in fig. 3, the apparatus 300 is deployed with a virtual machine 340, and the virtual machine 340 includes an obtaining module 310 and a processing module 320.
The obtaining module 310 is configured to obtain parameters of the IO request, where the parameters include a block size of the IO request, a timestamp of sending the IO request, and a timestamp of completing the IO request.
The processing module 320 is configured to determine, according to the parameter, whether IO adjacent-position interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs.
Optionally, the processing module 320 is specifically configured to determine a time difference between the timestamp of the sending IO request and the timestamp of the completing IO request; determining the ratio of the time difference to the block size of the IO request as unit sector processing delay; and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value.
Optionally, the unit sector processing delay reference value includes at least one of: the unit sector processing delay pre-estimated value is obtained by the first virtual machine according to the mapping relation among the block size of the IO request, the timestamp of the IO request and the unit sector processing delay pre-estimated value, the virtual machine set is composed of the second virtual machine and/or at least one third virtual machine, and the third virtual machine and the second virtual machine have the same application.
Optionally, the processing module 320 is specifically configured to determine that IO adjacent-position interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs when a deviation value of the unit sector processing delay with respect to the unit sector processing delay reference value is greater than a first threshold.
Optionally, the processing module 320 is specifically configured to determine that, in a plurality of continuous monitoring periods, all deviation values of the unit sector processing delay with respect to the unit sector processing delay reference value are greater than the first threshold, and IO adjacent position interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs.
Optionally, the processing module 320 is specifically configured to determine a time difference between the timestamp of the sending IO request and the timestamp of the completing IO request; determining a time difference pre-estimated value according to the block size of the IO request and the mapping relation between the block size of the IO request and the time difference; and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the time difference relative to the time difference estimated value.
Optionally, the processing module 320 is specifically configured to determine that IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs when a deviation value of the time difference from the time difference estimated value is greater than a second threshold.
Optionally, the processing module 320 is specifically configured to determine that IO adjacent bit interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs when all deviation values of the time difference with respect to the time difference estimated value are greater than the second threshold value in a plurality of continuous monitoring periods.
Optionally, virtual machine 340 also includes storage module 330.
The storage module 330 is configured to update the historical minimum unit sector processing delay in the database when the unit sector processing delay is smaller than the historical unit sector processing delay.
The specific functions and advantages of the obtaining module 310, the processing module 320 and the storing module 330 can refer to the method shown in fig. 2, and are not described herein again.
The acquisition module 310 may be a transceiver, a communication interface, or a processor. The storage module 330 may be a memory. The processing module 320 may be a processor or a controller, such as a Central Processing Unit (CPU), a general purpose processor, a Digital Signal Processor (DSP), an application-specific integrated circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor may also be a combination of computing functions, e.g., comprising one or more microprocessors, DSPs, and microprocessors, among others.
When the obtaining module is a transceiver, the processing module is a processor, and the storage module is a memory, the device for detecting the IO adjacent position interference of the cloud hard disk according to the present application may be the device shown in fig. 4.
As shown in fig. 4, the apparatus 400 may include a transceiver 410, a processor 420, and a memory 430.
Only one memory and processor are shown in fig. 4. In an actual control device product, there may be one or more processors and one or more memories. The memory may also be referred to as a storage medium or a storage device, etc. The memory may be provided independently of the processor, or may be integrated with the processor, which is not limited in this embodiment.
The transceiver 410, processor 420, and memory 430 communicate with each other via internal connection paths to transfer control and/or data signals.
In particular, the transceiver 410 is configured to obtain parameters of the IO request, where the parameters include a block size of the IO request, a timestamp of sending the IO request, and a timestamp of completing the IO request. The processor 420 is configured to determine, according to the parameter, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs.
The specific operation and advantages of the apparatus 400 can be seen from the description of the embodiment shown in fig. 2.
Fig. 5 is a schematic configuration diagram of a detection system according to an embodiment of the present application. Taking the case where the first virtual machine and the second virtual machine are different virtual machines as an example, fig. 5 is merely exemplary. As shown in fig. 5, virtual machine 540 may correspond to the first virtual machine above (acquisition module not shown); virtual machine 521, virtual machine 522, and virtual machine 531 may correspond to the second virtual machine above, and when the same application is deployed in virtual machine 521, virtual machine 522, and virtual machine 531, virtual machine 521, virtual machine 522, and virtual machine 531 may also correspond to the third virtual machine above.
Specifically, after the parameters of the IO request are collected by the collection module 523, the collection module 524, and the collection module 532, the virtual machine 521, the virtual machine 522, and the virtual machine 531 report the parameters to the virtual machine 540, where the reporting interval is configurable. Virtual machine 540 receives parameters of the IO request reported by virtual machine 521, virtual machine 522, and virtual machine 531, updates the database, determines and identifies IO adjacent-position interference, and performs alarm and subsequent corresponding processing (e.g., log and alarm; applying for a new virtual machine, replacing a virtual machine with IO adjacent-position interference after service deployment, etc.).
The specific operation and advantages of the detection system 500 can be seen from the description of the embodiment shown in fig. 2.
In the embodiments of the present application, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the inherent logic of the processes, and should not constitute any limitation to the implementation processes of the embodiments of the present application.
The steps of a method or algorithm described in connection with the disclosure herein may be embodied in hardware or in software instructions executed by a processor. The software instructions may be comprised of corresponding software modules that may be stored in Random Access Memory (RAM), flash memory, Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), registers, a hard disk, a removable disk, a compact disc read only memory (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an ASIC.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the procedures or functions described in the present application are generated in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable apparatus. The computer instructions may be stored on or transmitted over a computer-readable storage medium. The computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., Digital Versatile Disk (DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), etc.
The above-mentioned embodiments are further described in detail for the purpose of illustrating the present invention, and it should be understood that the above-mentioned embodiments are only for illustrating the present invention and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.

Claims (20)

1. A method for detecting IO adjacent-position interference of cloud hard disk input and output is characterized by comprising the following steps:
the method comprises the steps that a first virtual machine obtains parameters of an IO request, wherein the parameters comprise the block size of the IO request, a timestamp for sending the IO request and a timestamp for completing the IO request;
and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the parameter.
2. The method according to claim 1, wherein the determining, by the first virtual machine, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to the parameter includes:
the first virtual machine determines a time difference between the timestamp of the sending IO request and the timestamp of the completing IO request;
the first virtual machine determines that the ratio of the time difference value to the block size of the IO request is unit sector processing delay;
and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value.
3. The method of claim 2, wherein the unit sector processing delay reference value comprises at least one of: the unit sector processing delay pre-estimated value is obtained by the first virtual machine according to the mapping relation among the block size of the IO request, the timestamp of the IO request and the unit sector processing delay pre-estimated value, the virtual machine set is composed of the second virtual machine and/or at least one third virtual machine, and the third virtual machine and the second virtual machine have the same application.
4. The method according to claim 2 or 3, wherein the determining, by the first virtual machine, whether the IO adjacent-position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay with respect to the unit sector processing delay reference value includes:
and when the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is larger than a first threshold value, the first virtual machine determines that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs.
5. The method of claim 4, wherein when the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is greater than a first threshold value, the determining, by the first virtual machine, that the IO proximity interference occurs on a cloud hard disk of a second virtual machine to which the parameter belongs comprises:
and in a plurality of continuous monitoring periods, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs, wherein the deviation values of the unit sector processing delay relative to the unit sector processing delay reference value are all larger than the first threshold value.
6. The method according to claim 1, wherein the determining, by the first virtual machine, whether IO proximity interference occurs in a cloud hard disk of a second virtual machine to which the parameter belongs according to the parameter includes:
the first virtual machine determines a time difference between the timestamp of the sending IO request and the timestamp of the completing IO request;
the first virtual machine determines a time difference estimated value according to the block size of the IO request and the mapping relation between the block size of the IO request and the time difference;
and the first virtual machine determines whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the time difference relative to the time difference estimated value.
7. The method of claim 6, wherein the determining, by the first virtual machine, whether the IO proximity interference occurs to a cloud hard disk of a second virtual machine to which the parameter belongs according to the deviation value of the time difference from the time difference estimated value comprises:
and when the deviation value of the time difference relative to the time difference estimated value is larger than a second threshold value, the first virtual machine determines that IO adjacent position interference occurs on a cloud hard disk of a second virtual machine to which the parameter belongs.
8. The method of claim 7, wherein the determining, by the first virtual machine, that the IO proximity interference occurs on a cloud hard disk of a second virtual machine to which the parameter belongs when the deviation value of the time difference from the time difference estimated value is greater than a second threshold value comprises:
and determining IO adjacent position interference of a cloud hard disk of a second virtual machine to which the parameter belongs when the deviation values of the time difference relative to the time difference estimated value are all larger than the second threshold value in a plurality of continuous monitoring periods.
9. The method of any of claims 3 to 5, wherein when the unit sector processing latency is less than the historical unit sector processing latency, the method further comprises:
the first virtual machine updates historical minimum unit sector processing latency in a database.
10. An apparatus for detecting IO adjacent-position interference of cloud hard disk, wherein the apparatus is deployed with a virtual machine, and the virtual machine includes:
the system comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring parameters of an IO request, and the parameters comprise the block size of the IO request, a timestamp for sending the IO request and a timestamp for completing the IO request;
and the processing module is used for determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the parameter.
11. The apparatus of claim 10, wherein the processing module is specifically configured to:
determining a time difference between the timestamp of the transmit IO request and the timestamp of the complete IO request;
determining the ratio of the time difference to the block size of the IO request as unit sector processing time delay;
and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value.
12. The apparatus of claim 11, wherein the unit sector processing delay reference value comprises at least one of: the unit sector processing delay pre-estimated value is obtained by the first virtual machine according to the mapping relation among the block size of the IO request, the timestamp of the IO request and the unit sector processing delay pre-estimated value, the virtual machine set is composed of the second virtual machine and/or at least one third virtual machine, and the third virtual machine and the second virtual machine have the same application.
13. The apparatus according to claim 11 or 12, wherein the processing module is specifically configured to:
and when the deviation value of the unit sector processing delay relative to the unit sector processing delay reference value is larger than a first threshold value, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs.
14. The apparatus of claim 13, wherein the processing module is specifically configured to:
and in a plurality of continuous monitoring periods, determining that IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs, wherein the deviation values of the unit sector processing delay relative to the unit sector processing delay reference value are all larger than the first threshold value.
15. The apparatus of claim 10, wherein the processing module is specifically configured to:
determining a time difference between the timestamp of the transmit IO request and the timestamp of the complete IO request;
determining a time difference pre-estimated value according to the block size of the IO request and the mapping relation between the block size of the IO request and the time difference;
and determining whether IO adjacent position interference occurs in the cloud hard disk of the second virtual machine to which the parameter belongs according to the deviation value of the time difference relative to the time difference estimated value.
16. The apparatus of claim 15, wherein the processing module is specifically configured to:
and when the deviation value of the time difference relative to the time difference estimated value is larger than a second threshold value, determining that IO adjacent position interference occurs on a cloud hard disk of a second virtual machine to which the parameter belongs.
17. The apparatus of claim 16, wherein the processing module is specifically configured to:
and determining IO adjacent position interference of a cloud hard disk of a second virtual machine to which the parameter belongs when the deviation values of the time difference relative to the time difference estimated value are all larger than the second threshold value in a plurality of continuous monitoring periods.
18. The apparatus of any of claims 12 to 14, wherein the virtual machine further comprises:
and the storage module is used for updating the historical minimum unit sector processing time delay in the database when the unit sector processing time delay is smaller than the historical unit sector processing time delay.
19. An apparatus for detecting IO proximity interference of cloud hard disk, comprising a transceiver, a processor and a memory, and configured to perform the method according to any one of claims 1 to 9.
20. A computer-readable storage medium comprising instructions that, when executed by a processing module or a processor, cause an apparatus for detecting cloud hard disk input output, IO, proximity interference to perform the method of any of claims 1 to 9.
CN201811455270.6A 2018-11-30 2018-11-30 Method and device for detecting cloud hard disk input/output IO (input/output) ortho-position interference Active CN111258705B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811455270.6A CN111258705B (en) 2018-11-30 2018-11-30 Method and device for detecting cloud hard disk input/output IO (input/output) ortho-position interference

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811455270.6A CN111258705B (en) 2018-11-30 2018-11-30 Method and device for detecting cloud hard disk input/output IO (input/output) ortho-position interference

Publications (2)

Publication Number Publication Date
CN111258705A true CN111258705A (en) 2020-06-09
CN111258705B CN111258705B (en) 2023-12-08

Family

ID=70946514

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811455270.6A Active CN111258705B (en) 2018-11-30 2018-11-30 Method and device for detecting cloud hard disk input/output IO (input/output) ortho-position interference

Country Status (1)

Country Link
CN (1) CN111258705B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103389884A (en) * 2013-07-29 2013-11-13 华为技术有限公司 Method for processing input/output request, host, server and virtual machine
CN103516812A (en) * 2013-10-22 2014-01-15 浪潮电子信息产业股份有限公司 Method for accelerating cloud storage internal data transmission
US20150169341A1 (en) * 2013-12-16 2015-06-18 Vmware, Inc. Virtual machine data store queue allocation
CN104750547A (en) * 2013-12-31 2015-07-01 华为技术有限公司 Input-output (IO) request processing method and device of virtual machines
CN104995604A (en) * 2015-03-03 2015-10-21 华为技术有限公司 Resource allocation method of virtual machine and device thereof
CN107422989A (en) * 2017-07-27 2017-12-01 深圳市云舒网络技术有限公司 A kind of more copy read methods of Server SAN systems and storage architecture
CN108924221A (en) * 2018-06-29 2018-11-30 华为技术有限公司 The method and apparatus for distributing resource

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103389884A (en) * 2013-07-29 2013-11-13 华为技术有限公司 Method for processing input/output request, host, server and virtual machine
CN103516812A (en) * 2013-10-22 2014-01-15 浪潮电子信息产业股份有限公司 Method for accelerating cloud storage internal data transmission
US20150169341A1 (en) * 2013-12-16 2015-06-18 Vmware, Inc. Virtual machine data store queue allocation
CN104750547A (en) * 2013-12-31 2015-07-01 华为技术有限公司 Input-output (IO) request processing method and device of virtual machines
CN104995604A (en) * 2015-03-03 2015-10-21 华为技术有限公司 Resource allocation method of virtual machine and device thereof
CN107422989A (en) * 2017-07-27 2017-12-01 深圳市云舒网络技术有限公司 A kind of more copy read methods of Server SAN systems and storage architecture
CN108924221A (en) * 2018-06-29 2018-11-30 华为技术有限公司 The method and apparatus for distributing resource

Also Published As

Publication number Publication date
CN111258705B (en) 2023-12-08

Similar Documents

Publication Publication Date Title
US10175909B2 (en) Extent migration in multi-tier storage systems
KR102496598B1 (en) Non-volatile memory storage device capable of self-reporting performance capabilities
EP2515233A1 (en) Detecting and diagnosing misbehaving applications in virtualized computing systems
US11582130B2 (en) Performance monitoring in a distributed storage system
US9804881B2 (en) System and method for resizing a virtual desktop infrastructure using virtual desktop infrastructure monitoring tools
US20190340095A1 (en) Predicting performance of applications using machine learning systems
US11256595B2 (en) Predictive storage management system
US10069753B2 (en) Relationship-based resource-contention analysis system and method
US20140359624A1 (en) Determining a completion time of a job in a distributed network environment
US10430723B1 (en) Storage system with machine learning based skew prediction
US11556740B2 (en) Sensor triggered sound clip capturing for machine learning
WO2011059837A1 (en) Allocating storage memory based on future use estimates
TW201414257A (en) System and method for monitoring storage devices
US20140351504A1 (en) Apparatus and method for transferring data between storages having different access speeds
US11175829B2 (en) Automatic identification of workloads contributing to behavioral changes in storage systems using machine learning techniques
US11436123B2 (en) Application execution path tracing for inline performance analysis
US10552282B2 (en) On demand monitoring mechanism to identify root cause of operation problems
US20180300065A1 (en) Storage resource management employing end-to-end latency analytics
US20160077741A1 (en) Capturing demand on storage capacity and performance capability
US10289958B1 (en) Storage system with machine learning based skew prediction
CN111258705A (en) Method and device for detecting IO adjacent position interference of cloud hard disk input and output
US11416152B2 (en) Information processing device, information processing method, computer-readable storage medium, and information processing system
US10956084B2 (en) Drive utilization in multi-tiered systems with read-intensive flash
US20180329796A1 (en) Storage system performance models based on empirical component utilization
US20220011950A1 (en) Autonomous data protection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant