CN116827831B - Load state detection system, method and computer readable storage medium - Google Patents
Load state detection system, method and computer readable storage medium Download PDFInfo
- Publication number
- CN116827831B CN116827831B CN202311101219.6A CN202311101219A CN116827831B CN 116827831 B CN116827831 B CN 116827831B CN 202311101219 A CN202311101219 A CN 202311101219A CN 116827831 B CN116827831 B CN 116827831B
- Authority
- CN
- China
- Prior art keywords
- physical machine
- request
- detection application
- preset
- response
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 175
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000004044 response Effects 0.000 claims description 113
- 238000012545 processing Methods 0.000 claims description 22
- 239000000523 sample Substances 0.000 claims description 22
- 230000008569 process Effects 0.000 description 7
- 238000004590 computer program Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Debugging And Monitoring (AREA)
Abstract
The embodiment of the application provides a load state detection system, a method and a computer readable storage medium, which are applied to a cloud platform, wherein the system comprises: the system comprises at least two physical machines, wherein the at least two physical machines comprise a first physical machine and a second physical machine, a first detection application is installed on the first physical machine, and a second detection application is installed on the second physical machine; the hardware resource use condition of the second physical machine can be adjusted according to the load adjustment request, and after the second physical machine completes adjustment, the first detection application and the second detection application are utilized to determine the load state of the second physical machine, so that the detection of the load capacity of the second physical machine is realized. And the load capacity of each physical machine can be obtained, and the load capacity of the cloud platform is determined based on the load capacity of each physical machine.
Description
Technical Field
The present application relates to the field of cloud platform technologies, and in particular, to a load state detection system, a load state detection method, and a computer readable storage medium.
Background
The cloud platform is a service based on hardware resources and software resources, provides computing, networking and storage functions, and is composed of a plurality of physical machines. Along with the rapid development of computer technology, the cloud platform is focused by more and more enterprises and users, and in order to adapt to the ever-expanding user scale, the load capacity of the cloud platform needs to be determined, so that the overall stability of the cloud platform is ensured.
Disclosure of Invention
Embodiments of the present application provide a load status detection system, method and computer readable storage medium to determine the load capacity of a cloud platform. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a load status detection system, applied to a cloud platform, where the system includes:
the system comprises at least two physical machines, wherein the at least two physical machines comprise a first physical machine and a second physical machine, a first detection application is installed on the first physical machine, and a second detection application is installed on the second physical machine;
the second detection application is used for determining the type of the hardware resource included on the second physical machine and sending the determined type information of the hardware resource to the first detection application;
The first detection application is used for sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine after receiving the type information of the hardware resource;
the second physical machine is used for receiving the load adjustment request, calling a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sending response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request;
the first detection application is further configured to send a first request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment;
the second detection application is further configured to send a second request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment.
Optionally, the hardware resource type includes a memory resource, a CPU resource, and a GPU resource, and the sending, to the second physical machine, a load adjustment request determined based on the type information of the hardware resource includes:
A load adjustment request comprising a preset memory utilization rate, a preset CPU utilization rate and a preset GPU utilization rate is sent to the second physical machine;
the calling the second detection application installed by the calling self adjusts the use condition of the hardware resource according to the load adjustment request, and the method comprises the following steps:
and calling a second detection application installed by the user to adjust the use condition of the memory according to the preset memory use rate, adjusting the use condition of the CPU according to the preset CPU use rate, and adjusting the use condition of the GPU according to the preset GPU use rate.
Optionally, after the first probe application sends a first request to the second physical machine, the first physical machine is further configured to record a sending time of the first request, and the second physical machine is further configured to record a receiving time of the first request, where the sending time of the first request and the receiving time of the first request are used to characterize a load state of the second physical machine.
Optionally, after receiving the first request, the second physical machine is further configured to respond to the first request to obtain a first response, send the first response to a first detection application, and determine a sending time of the first response; determining the processing duration of the first request based on the receiving time of the first request and the sending time of the first response;
The first detection application is used for receiving the first response and recording the receiving time of the first response, and the sending time of the first response, the processing time of the first request and the receiving time of the first response are used for representing the load state of the second physical machine.
Optionally, after the second probe application sends a second request to the second physical machine, the second probe application is configured to record a sending time of the second request, and the second physical machine is further configured to record a receiving time of the second request, where the sending time of the second request and the receiving time of the second request are used to characterize a load state of the second physical machine.
Optionally, after receiving the second request, the second physical machine is further configured to respond to the second request to obtain a second response, send the second response to a second detection application, and determine a sending time of the second response; determining the processing time of the second request based on the receiving time of the second request and the sending time of the second response;
the second detection application is configured to receive the second response, and record a receiving time of the second response, where the sending time of the second response, the processing duration of the second request, and the receiving time of the second response are used to characterize a load state of the second physical machine.
Optionally, the first detection application is further configured to select any one value from a preset memory usage rate, a preset CPU usage rate, and a preset GPU usage rate, and adjust the height according to a preset step length, and send a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any value reaches an adjustment upper limit, selecting any unselected value from preset memory utilization rate, preset CPU utilization rate and preset GPU utilization rate, adjusting the value according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any unselected value reaches the adjustment upper limit, selecting the unselected value from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, adjusting the height according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
In a second aspect, an embodiment of the present application further provides a load status detection method, applied to a load status detection system, where the system includes: at least two physical machines, the at least two physical machines including a first physical machine on which a first detection application is installed and a second physical machine on which a second detection application is installed, the method comprising:
The second detection application determines the type of the hardware resources included on the second physical machine and sends the determined type information of the hardware resources to the first detection application;
after receiving the type information of the hardware resources, the first detection application sends a load adjustment request determined based on the type information of the hardware resources to the second physical machine;
the second physical machine receives the load adjustment request, invokes a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sends response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request;
the first detection application sends a first request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment;
and the second detection application sends a second request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment.
Optionally, the hardware resource type includes a memory resource, a CPU resource, and a GPU resource, and the sending, to the second physical machine, a load adjustment request determined based on the type information of the hardware resource includes:
A load adjustment request comprising a preset memory utilization rate, a preset CPU utilization rate and a preset GPU utilization rate is sent to the second physical machine;
the calling the second detection application installed by the calling self adjusts the use condition of the hardware resource according to the load adjustment request, and the method comprises the following steps:
and calling a second detection application installed by the user to adjust the use condition of the memory according to the preset memory use rate, adjusting the use condition of the CPU according to the preset CPU use rate, and adjusting the use condition of the GPU according to the preset GPU use rate.
Optionally, after the first probe application sends a first request to the second physical machine, the first physical machine further records a sending time of the first request, and the second physical machine further records a receiving time of the first request, where the sending time of the first request and the receiving time of the first request are used to characterize a load state of the second physical machine.
Optionally, after receiving the first request, the second physical machine further responds to the first request to obtain a first response, sends the first response to a first detection application, and determines a sending time of the first response; determining the processing duration of the first request based on the receiving time of the first request and the sending time of the first response;
The first detection application receives the first response and records the receiving time of the first response, wherein the sending time of the first response, the processing time of the first request and the receiving time of the first response are used for representing the load state of the second physical machine.
Optionally, after the second probe application sends a second request to the second physical machine, the second probe application records a sending time of the second request, and the second physical machine further records a receiving time of the second request, where the sending time of the second request and the receiving time of the second request are used to characterize a load state of the second physical machine.
Optionally, after receiving the second request, the second physical machine further responds to the second request to obtain a second response, sends the second response to a second detection application, and determines a sending time of the second response; determining the processing time of the second request based on the receiving time of the second request and the sending time of the second response;
the second detection application receives the second response and records the receiving time of the second response, wherein the sending time of the second response, the processing time of the second request and the receiving time of the second response are used for representing the load state of a second physical machine.
Optionally, the first detection application further selects any value from a preset memory utilization rate, a preset CPU utilization rate and a preset GPU utilization rate to perform heightening according to a preset step length, and sends a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any value reaches an adjustment upper limit, selecting any unselected value from preset memory utilization rate, preset CPU utilization rate and preset GPU utilization rate, adjusting the value according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any unselected value reaches the adjustment upper limit, selecting the unselected value from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, adjusting the height according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
In a third aspect, embodiments of the present application further provide a computer readable storage medium having a computer program stored therein, which when executed by a processor implements any of the load state detection methods described above.
The embodiment of the application has the beneficial effects that:
the load state detection system provided by the embodiment of the application is applied to a cloud platform, and comprises the following components: the system comprises at least two physical machines, wherein the at least two physical machines comprise a first physical machine and a second physical machine, a first detection application is installed on the first physical machine, and a second detection application is installed on the second physical machine; the hardware resource use condition of the second physical machine can be adjusted according to the load adjustment request, and after the second physical machine completes adjustment, the first detection application and the second detection application are utilized to determine the load state of the second physical machine, so that the detection of the load capacity of the second physical machine is realized. And the load capacity of each physical machine can be obtained, and the load capacity of the cloud platform is determined based on the load capacity of each physical machine.
Of course, it is not necessary for any one product or method of practicing the application to achieve all of the advantages set forth above at the same time.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the application, and other embodiments may be obtained according to these drawings to those skilled in the art.
Fig. 1 is a schematic structural diagram of a load status detection system according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a load status detection method according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by the person skilled in the art based on the present application are included in the scope of protection of the present application.
Currently, the application range of the cloud platform is expanding continuously, and in order to adapt to large-scale use users and ranges, higher requirements are put on the stability of public clouds, private clouds and mixed clouds. How to ensure the normal running of the program under the condition of high load of the server is an urgent problem to be solved, so that the highest load index of the server needs to be known under the premise of ensuring the stability. In order to determine the load capacity of the cloud platform, the detection of the server stability can be performed through the following two schemes:
The first is to use the pressure testing software to perform strong pressure testing on certain services and observe the result after the pressure testing. For example, a user interface is created for a certain service, 10000 times are requested within 1 minute, how many times the observation result succeeds and how many times the observation result fails, and then the load capacity of the system service can be roughly estimated. This approach is merely a load estimation of the server service interface and is not representative of the load capacity of the server. Moreover, the service interface of the cloud platform is often loaded by a request, so 10000 requests can be equally divided into a plurality of servers according to the number of the servers. There is no way to know the load capacity of a particular server.
The second is that the cloud platform is provided with a superSales ratios such as CPU supersorder, disk supersorder. CPU supersale ratio 1:5, meaning N core CPU can supersale intoAnd (3) a core. However, this way of setting up on the platform means that the server capability of the whole platform is at the same time, but this is not true. For example, the same server responds differently under the same load, such as 95% of the CPU. For example, under the same load, some servers with high performance may not have exerted a machine value, but another server may have reached the katon limit. In addition, the overstock ratio is set to be too high, so that the platform is unstable and not available, and the overstock ratio is set to be too low, so that the platform resource is wasted greatly. It is therefore desirable to have the ability to probe each server as much as possible. The cloud platform can further physically divide servers with different load capacities, and stability is integrally improved.
To solve at least one of the above problems, in a first aspect, the present application provides a load status detection system applied to a cloud platform, referring to fig. 1, the system includes:
the system comprises at least two physical machines, wherein the at least two physical machines comprise a first physical machine and a second physical machine, a first detection application is installed on the first physical machine, and a second detection application is installed on the second physical machine.
The cloud platform is a service based on hardware resources and software resources, provides computing, networking and storage functions, and is composed of a plurality of physical machines, and the load capacity of the cloud platform depends on the sum of the load capacities of each physical machine included in the cloud platform. The load state detection system provided by the embodiment of the application can detect the load capacity of the cloud platform, and is applied to the cloud platform comprising at least two physical machines. And each physical machine is provided with a detection application for detecting the load capacity, wherein the first physical machine is provided with a first detection application, and the second physical machine is provided with a second detection application.
The first detection application is an internal detection application relative to the first physical machine and an external detection application relative to the second physical machine, so that the first detection application can detect the load state of the first physical machine and the load state of the second physical machine; the second detection application is an internal detection application with respect to the second physical machine and an external detection application with respect to the first physical machine, so that the second detection application can detect the load state of the first physical machine as well as the second physical machine.
The second detection application is used for determining the type of the hardware resource included on the second physical machine and sending the determined type information of the hardware resource to the first detection application.
In one example, before the step of determining, by the second probing application, the type of hardware resource included on the second physical machine, a process of establishing a communication connection between the first physical machine and the second physical machine is further required to determine that the probing of the load state is performed between the first physical machine and the second physical machine. Specifically, the communication connection establishment procedure may be a connection established through a handshake mechanism, including: the first physical machine sends a connection request to the second physical machine, and if the second physical machine accepts the connection request, the second physical machine sends a confirmation response; after the connection is established, data exchange between the first physical machine and the second physical machine may begin.
In the embodiment of the present application, the first physical machine detects the second physical machine first, which is taken as an example for explanation, however, the second physical machine may detect the first physical machine first, and the detection sequence is not limited by the present application. The application for detecting the load state of the second physical machine comprises a first detection application and a second detection application, wherein the first detection application is an external detection application and is deployed outside the second physical machine, and the second detection application is an internal detection application and is deployed inside the second physical machine.
The second probe application determines a type of hardware resource, such as a memory resource, a CPU resource, and a GPU resource, included in the second physical machine, and sends information about the type of hardware resource included in the second physical machine to the first probe application.
The first probe application is configured to send a load adjustment request to the second physical machine, the load adjustment request being determined based on the type information of the hardware resource, after receiving the type information of the hardware resource.
The first probing application, after receiving the type information of the hardware resources, knows which hardware resources are included on the second physical machine, and also knows which aspects of the second physical machine to probe for loads. The first detection application sends a load adjustment request determined based on the type information of the hardware resources to the second physical machine, so that the second physical machine can adjust the use condition of the hardware resources.
In one example, sending a load adjustment request to the second physical machine, the load adjustment request being determined based on the type information of the hardware resource, includes: and sending a load adjustment request containing the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate to the second physical machine. The preset memory usage, the preset CPU usage, and the preset GPU usage are all values that can be empirically set. For example, the preset memory utilization is 50%, the preset CPU utilization is 40%, and the preset GPU utilization is 50%. In one possible implementation, the first detection application may restore the state of the second physical machine according to the initial value, for example, the first detection application may cause the second physical machine to adjust the CPU utilization according to the initial value of 30% of the CPU utilization.
The second physical machine is used for receiving the load adjustment request, calling a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sending response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request.
After receiving the load adjustment request, the second physical machine may invoke the second detection application installed by the second physical machine to adjust the usage of the hardware resource according to the load adjustment request, and in one example, the second physical machine may invoke the second detection application installed by the second physical machine to adjust the usage of the memory according to a preset memory usage, adjust the usage of the CPU according to a preset CPU usage, and adjust the usage of the GPU according to a preset GPU usage.
When the second physical machine completes the adjustment of the hardware resource, the response information for the load adjustment request can be sent to the first detection application, and the first detection application can know that the second physical machine has completed the load adjustment after receiving the response information.
The first detection application is further configured to send a first request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment.
After the first probe application sends a first request to the second physical machine, the first physical machine is further configured to record a sending time of the first request, and the second physical machine is further configured to record a receiving time of the first request, where the sending time of the first request and the receiving time of the first request are used to characterize a load state of the second physical machine.
The first request may be a request made for a data acquisition service, a request made for a data update service, or a request made for a data storage service, which is not limited in the specific form of the request.
The time of transmission of the first request and the time of receipt of the first request are used to characterize the load state of the second physical machine, and in one example, a time difference between the time of transmission and the time of end may be calculated. If the time difference is 500ms as the critical value, that is, if the time difference between the time of sending the first request and the time of receiving the first request is not greater than 500ms, the load state of the second physical machine is considered to be good, and if the time difference exceeds 500ms, the load state of the second physical machine is considered to be poor.
In one example, after receiving the first request, the second physical machine is further configured to respond to the first request to obtain a first response, send the first response to a first detection application, and determine a sending time of the first response; and determining the processing duration of the first request based on the receiving time of the first request and the sending time of the first response.
And the second physical machine also records the sending time of the first response in the process of responding to the first request to obtain the first response, wherein the time difference between the receiving time of the first request and the sending time of the first response is the processing time of the second physical machine to the first request.
The first detection application is used for receiving the first response and recording the receiving time of the first response, and the sending time of the first response, the processing time of the first request and the receiving time of the first response are used for representing the load state of the second physical machine.
The determining the load state of the second physical machine according to the sending time of the first response and the receiving time of the first response may specifically be determining the load state of the second physical machine according to a time difference between the sending time of the first response and the first byte of the received first response data. If the time difference is 500ms as the critical value, that is, if the time difference between the time of sending the first response and the time of receiving the first byte of the first response data is not greater than 500ms, the load state of the second physical machine is considered to be good, and if the time difference exceeds 500ms, the load state of the second physical machine is considered to be poor.
The second detection application is further configured to send a second request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment.
After the second probe application sends a second request to the second physical machine, the second probe application is configured to record a sending time of the second request, and the second physical machine is further configured to record a receiving time of the second request, where the sending time of the second request and the receiving time of the second request are used to characterize a load state of the second physical machine.
The second request may be a request made for a data acquisition service, a request made for a data update service, or a request made for a data storage service, which is not limited in the specific form of the request.
The sending time of the second request and the receiving time of the second request are used to characterize the load state of the second physical machine, and in one example, a time difference between the sending time and the ending time may be calculated. If the time difference is 500ms as the critical value, that is, if the time difference between the time of sending the second request and the time of receiving the second request is not greater than 500ms, the load state of the second physical machine is considered to be good, and if the time difference exceeds 500ms, the load state of the second physical machine is considered to be poor.
In one example, after receiving the second request, the second physical machine is further configured to respond to the second request to obtain a second response, send the second response to a second detection application, and determine a sending time of the second response; and determining the processing duration of the second request based on the receiving time of the second request and the sending time of the second response.
And the second physical machine also records the sending time of the second response in the process of responding to the second request to obtain the second response, wherein the time difference between the receiving time of the second request and the sending time of the second response is the processing time of the second physical machine to the second request.
The second detection application is configured to receive the second response, and record a receiving time of the second response, where the sending time of the second response, the processing duration of the second request, and the receiving time of the second response are used to characterize a load state of the second physical machine.
The determining the load state of the second physical machine according to the sending time of the second response and the receiving time of the second response may specifically be determining the load state of the second physical machine according to a time difference between the sending time of the second response and the first byte of the received second response data. If the time difference is 500ms as the critical value, that is, if the time difference between the sending time of the second response and the time difference of receiving the first byte of the second response data is not greater than 500ms, the load state of the second physical machine is considered to be good, and if the time difference exceeds 500ms, the load state of the second physical machine is considered to be poor.
The load state detection system provided by the embodiment of the application is applied to the cloud platform, and the first detection application and the second detection application are respectively installed on the first physical machine and the second physical machine, so that the load states of the second physical machine under different load conditions are detected by using the first detection application and the second detection application, and the multi-dimensional comprehensive detection of the load capacity of the cloud platform can be realized.
In one example, the first detection application is further configured to select any one value from a preset memory usage rate, a preset CPU usage rate, and a preset GPU usage rate, and adjust the value according to a preset step size, and send a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
Taking the selection of the preset memory usage rate as an example for illustration, the preset step length is a value set empirically, for example, may be 1%, and for the preset memory usage rate of 50%, the preset step length is adjusted up by 1% each time, so as to obtain a set of data with the preset memory usage rate of 51%, the preset CPU usage rate of 40% and the preset GPU usage rate of 50%. And sending a load adjustment request containing a preset memory utilization rate of 51%, a preset CPU utilization rate of 40% and a preset GPU utilization rate of 50% to the second physical machine.
When any value reaches the adjustment upper limit, any unselected value is selected from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, the value is adjusted according to a preset step length, and a load adjustment request determined based on the type information of the hardware resource is sent to the second physical machine.
If the preset memory utilization rate has been adjusted to 100%, which means that the memory utilization rate has reached the adjustment upper limit, a value needs to be selected from the preset CPU utilization rate and the preset GPU utilization rate at this time, for example, the preset CPU utilization rate is selected, and then 40% of the preset CPU utilization rate can be adjusted according to a preset step length, so as to obtain a set of data with 50% of the preset memory utilization rate, 41% of the preset CPU utilization rate and 50% of the preset GPU utilization rate. The preset step length for increasing the preset CPU utilization may be the same as or different from the preset step length for increasing the preset memory utilization, which is not limited in the embodiment of the present application.
And sending a load adjustment request containing 50% of preset memory utilization rate, 41% of preset CPU utilization rate and 50% of preset GPU utilization rate to the second physical machine.
When any unselected value reaches the adjustment upper limit, selecting the unselected value from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, adjusting the height according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
If the preset CPU utilization rate is adjusted to 100%, the CPU utilization rate is up to the adjustment upper limit, and the preset GPU utilization rate is selected to be increased according to the preset step length, so that a group of data with the preset memory utilization rate of 50%, the preset CPU utilization rate of 40% and the preset GPU utilization rate of 51% is obtained. The preset step size for increasing the preset GPU utilization may be the same as or different from the preset step size for increasing the preset CPU utilization, which is not limited in the embodiment of the present application.
And sending a load adjustment request containing a preset memory utilization rate of 50%, a preset CPU utilization rate of 40% and a preset GPU utilization rate of 51% to the second physical machine.
The first detection application has the functions of increasing the memory utilization rate, the CPU utilization rate and the GPU utilization rate, and can simulate the read-write function of the magnetic disk.
In one example, the highest load and the lowest delay of the second physical machine can be obtained by data aggregation analysis of the second physical machine under the conditions of different memory utilization rates, CPU utilization rates and GPU utilization rates.
In a second aspect, an embodiment of the present application further provides a load status detection method, applied to a load status detection system, where the system includes: at least two physical machines, the at least two physical machines including a first physical machine on which a first detection application is installed and a second physical machine on which a second detection application is installed, the method comprising:
The second detection application determines the type of the hardware resources included on the second physical machine and sends the determined type information of the hardware resources to the first detection application;
after receiving the type information of the hardware resources, the first detection application sends a load adjustment request determined based on the type information of the hardware resources to the second physical machine;
the second physical machine receives the load adjustment request, invokes a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sends response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request;
the first detection application sends a first request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment;
and the second detection application sends a second request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment.
The following describes the flow of the load state detection method in detail with reference to fig. 2:
Firstly, deploying a detection application, specifically deploying a first detection application on a first physical machine and deploying a second detection application on a second physical machine; starting detection, accessing a second detection application through a first detection application, informing the second detection application to adjust the load S of the second physical machine, starting a round of detection, and detecting the load S+2 again after the round of detection is finished until the second physical machine generates larger time delay, and exiting the program; the first detection application informs the second detection application of increasing the memory utilization rate, the CPU utilization rate and the GPU utilization rate of the second physical machine, the second detection application correspondingly executes the steps of increasing the memory utilization rate, the CPU utilization rate and the GPU utilization rate of the second physical machine, after the second physical machine completes adjustment, the first detection application and the second detection application access the second physical machine, the first detection application accesses the second physical machine through accessing an API (application program interface), and the second detection application accesses the second physical machine through accessing the interface. In the process of detecting the first physical machine, the second detection application accesses the first detection application in an interface access mode, and notifies the first detection application to increase the memory utilization rate, the CPU utilization rate and the GPU utilization rate of the first physical machine.
In one example, after the first probe application sends a first request to the second physical machine, the first physical machine further records a sending time of the first request, and the second physical machine further records a receiving time of the first request, where the sending time of the first request and the receiving time of the first request are used to characterize a load state of the second physical machine.
In one example, after receiving the first request, the second physical machine further responds to the first request to obtain a first response, sends the first response to a first detection application, and determines a sending time of the first response; determining the processing duration of the first request based on the receiving time of the first request and the sending time of the first response;
the first detection application receives the first response and records the receiving time of the first response, wherein the sending time of the first response, the processing time of the first request and the receiving time of the first response are used for representing the load state of the second physical machine.
In one example, after the second probe application sends a second request to the second physical machine, the second probe application records a sending time of the second request, and the second physical machine further records a receiving time of the second request, where the sending time of the second request and the receiving time of the second request are used to characterize a load state of the second physical machine.
In one example, after receiving the second request, the second physical machine further responds to the second request to obtain a second response, sends the second response to a second detection application, and determines a sending time of the second response; determining the processing time of the second request based on the receiving time of the second request and the sending time of the second response;
the second detection application receives the second response and records the receiving time of the second response, wherein the sending time of the second response, the processing time of the second request and the receiving time of the second response are used for representing the load state of a second physical machine.
In one example, the first detection application further selects any one value from a preset memory usage rate, a preset CPU usage rate and a preset GPU usage rate to perform heightening according to a preset step length, and sends a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any value reaches an adjustment upper limit, selecting any unselected value from preset memory utilization rate, preset CPU utilization rate and preset GPU utilization rate, adjusting the value according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
When any unselected value reaches the adjustment upper limit, selecting the unselected value from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, adjusting the height according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
In yet another embodiment of the present application, a computer readable storage medium is provided, in which a computer program is stored, which when executed by a processor, implements the steps of any of the load condition detection methods described above.
In yet another embodiment of the present application, a computer program product containing instructions that, when run on a computer, cause the computer to perform the load condition detection method of any of the above embodiments is also provided.
In the above embodiments, it may be implemented in whole or in part 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 loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), etc.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the method embodiments, since they are substantially similar to the system embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.
Claims (9)
1. A load condition detection system for use with a cloud platform, the system comprising:
the system comprises at least two physical machines, wherein the at least two physical machines comprise a first physical machine and a second physical machine, a first detection application is installed on the first physical machine, and a second detection application is installed on the second physical machine;
the second detection application is used for determining the type of the hardware resource included on the second physical machine and sending the determined type information of the hardware resource to the first detection application;
the first detection application is used for sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine after receiving the type information of the hardware resource;
the second physical machine is used for receiving the load adjustment request, calling a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sending response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request;
The first detection application is further configured to send a first request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment;
the second detection application is further configured to send a second request to the second physical machine to detect a load state of the second physical machine after the second physical machine completes the adjustment.
2. The system of claim 1, wherein the hardware resource type includes a memory resource, a CPU resource, and a GPU resource, and wherein the sending the load adjustment request to the second physical machine, which is determined based on the type information of the hardware resource, includes:
a load adjustment request comprising a preset memory utilization rate, a preset CPU utilization rate and a preset GPU utilization rate is sent to the second physical machine;
the calling the second detection application installed by the calling self adjusts the use condition of the hardware resource according to the load adjustment request, and the method comprises the following steps:
and calling a second detection application installed by the user to adjust the use condition of the memory according to the preset memory use rate, adjusting the use condition of the CPU according to the preset CPU use rate, and adjusting the use condition of the GPU according to the preset GPU use rate.
3. The system of claim 1, wherein after the first probe application sends a first request to the second physical machine, the first physical machine is further configured to record a sending time of the first request, and the second physical machine is further configured to record a receiving time of the first request, where the sending time of the first request and the receiving time of the first request are used to characterize a load state of the second physical machine.
4. The system of claim 3, wherein the second physical machine is further configured to, after receiving the first request, respond to the first request to obtain a first response, send the first response to the first probe application, and determine a sending time of the first response; determining the processing duration of the first request based on the receiving time of the first request and the sending time of the first response;
the first detection application is used for receiving the first response and recording the receiving time of the first response, and the sending time of the first response, the processing time of the first request and the receiving time of the first response are used for representing the load state of the second physical machine.
5. The system of claim 1, wherein after the second probe application sends a second request to the second physical machine, the second probe application is configured to record a sending time of the second request, and the second physical machine is further configured to record a receiving time of the second request, where the sending time of the second request and the receiving time of the second request are used to characterize a load state of the second physical machine.
6. The system of claim 5, wherein the second physical machine is further configured to respond to the second request after receiving the second request, obtain a second response, send the second response to a second probe application, and determine a sending time of the second response; determining the processing time of the second request based on the receiving time of the second request and the sending time of the second response;
the second detection application is configured to receive the second response, and record a receiving time of the second response, where the sending time of the second response, the processing duration of the second request, and the receiving time of the second response are used to characterize a load state of the second physical machine.
7. The system according to claim 2, wherein the first detection application is further configured to select any one of a preset memory usage rate, a preset CPU usage rate, and a preset GPU usage rate to be increased according to a preset step size, and send a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
when any value reaches an adjustment upper limit, selecting any unselected value from preset memory utilization rate, preset CPU utilization rate and preset GPU utilization rate, adjusting the value according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine;
When any unselected value reaches the adjustment upper limit, selecting the unselected value from the preset memory utilization rate, the preset CPU utilization rate and the preset GPU utilization rate, adjusting the height according to a preset step length, and sending a load adjustment request determined based on the type information of the hardware resource to the second physical machine.
8. A load condition detection method, applied to a load condition detection system, the system comprising: at least two physical machines, the at least two physical machines including a first physical machine on which a first detection application is installed and a second physical machine on which a second detection application is installed, the method comprising:
the second detection application determines the type of the hardware resources included on the second physical machine and sends the determined type information of the hardware resources to the first detection application;
after receiving the type information of the hardware resources, the first detection application sends a load adjustment request determined based on the type information of the hardware resources to the second physical machine;
the second physical machine receives the load adjustment request, invokes a second detection application installed by the second physical machine to adjust the use condition of the hardware resource according to the load adjustment request, and after the adjustment is completed, sends response information of the load adjustment request to the first detection application, wherein the response information is used for informing the first detection application that the second physical machine has completed the adjustment according to the load adjustment request;
The first detection application sends a first request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment;
and the second detection application sends a second request to the second physical machine to detect the load state of the second physical machine after the second physical machine completes adjustment.
9. The method according to claim 8, wherein the hardware resource type includes a memory resource, a CPU resource, and a GPU resource, and the sending, to the second physical machine, a load adjustment request determined based on the type information of the hardware resource includes:
a load adjustment request comprising a preset memory utilization rate, a preset CPU utilization rate and a preset GPU utilization rate is sent to the second physical machine;
the calling the second detection application installed by the calling self adjusts the use condition of the hardware resource according to the load adjustment request, and the method comprises the following steps:
and calling a second detection application installed by the user to adjust the use condition of the memory according to the preset memory use rate, adjusting the use condition of the CPU according to the preset CPU use rate, and adjusting the use condition of the GPU according to the preset GPU use rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311101219.6A CN116827831B (en) | 2023-08-30 | 2023-08-30 | Load state detection system, method and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311101219.6A CN116827831B (en) | 2023-08-30 | 2023-08-30 | Load state detection system, method and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116827831A CN116827831A (en) | 2023-09-29 |
| CN116827831B true CN116827831B (en) | 2023-11-17 |
Family
ID=88118850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311101219.6A Active CN116827831B (en) | 2023-08-30 | 2023-08-30 | Load state detection system, method and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116827831B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114564313A (en) * | 2022-03-07 | 2022-05-31 | 中国农业银行股份有限公司 | Load adjustment method and device, electronic equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10002025B2 (en) * | 2014-08-29 | 2018-06-19 | Hitachi, Ltd. | Computer system and load leveling program |
| FR3091769B1 (en) * | 2019-01-15 | 2022-03-25 | Amadeus | A method and system for managing computing resources of a cloud computing platform |
-
2023
- 2023-08-30 CN CN202311101219.6A patent/CN116827831B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114564313A (en) * | 2022-03-07 | 2022-05-31 | 中国农业银行股份有限公司 | Load adjustment method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116827831A (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11537304B2 (en) | Data verification method and apparatus, and storage medium | |
| US20050038789A1 (en) | On demand node and server instance allocation and de-allocation | |
| CN102122303A (en) | Method for data migration, service system and sever equipment | |
| WO2016023442A1 (en) | Network request method, network fluctuation measurement method and apparatus, and terminal | |
| CN108509157A (en) | A kind of data balancing method and device applied to distributed file system | |
| AU2004266019A1 (en) | On demand node and server instance allocation and de-allocation | |
| CN109271172B (en) | Host performance expansion method and device of sweep cluster | |
| CN111562884B (en) | Data storage method and device and electronic equipment | |
| WO2021197392A1 (en) | Task queue generation | |
| CN103024798A (en) | Method and device for testing access point (AP) performance | |
| CN111562889A (en) | Data processing method, device, system and storage medium | |
| CN106375102A (en) | Service registration method, application method and correlation apparatus | |
| CN107395708A (en) | A kind of method and apparatus for handling download request | |
| CN116827831B (en) | Load state detection system, method and computer readable storage medium | |
| CN108200151B (en) | ISCSI Target load balancing method and device in distributed storage system | |
| CN114595167A (en) | Distributed cache system, method and device | |
| CN113852490A (en) | Method, device and equipment for realizing quality of service (QoS) of samba user in distributed storage system | |
| CN114253456A (en) | Cache load balancing method and device | |
| CN107066889A (en) | A kind of data access control method and system based on geographical location information | |
| US9185226B2 (en) | Voicemail server monitoring/reporting via aggregated data | |
| CN107689876A (en) | The distribution management method of metadata in distributed objects storage system | |
| US20230308503A1 (en) | File transmission method and apparatus, device, and storage medium | |
| TWI766387B (en) | Reverse proxy method and storage device with delay sensing and load balancing | |
| CN109471703B (en) | Cloud environment-based virtual machine secure migration method and device | |
| CN114374657A (en) | Data processing method and device |
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 |