CN114154678A - A large-scale simulation resource adjustment method, device, equipment and storage medium - Google Patents
A large-scale simulation resource adjustment method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114154678A CN114154678A CN202111232829.0A CN202111232829A CN114154678A CN 114154678 A CN114154678 A CN 114154678A CN 202111232829 A CN202111232829 A CN 202111232829A CN 114154678 A CN114154678 A CN 114154678A
- Authority
- CN
- China
- Prior art keywords
- simulation
- resource
- historical
- scale simulation
- scale
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/219—Managing data history or versioning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本申请公开了一种大规模仿真资源调整方法、装置、设备及存储介质，该方法包括：S1:获取历史大规模仿真的资源需求历史信息；S2:根据资源需求历史信息生成当前大规模仿真的资源需求预测信息；S3:获取当前大规模仿真的资源需求实时信息；S4:根据资源需求预测信息和资源需求实时信息生成下一次大规模仿真的机器需求。本申请可以有效地预估下一次大规模仿真运行时所占用的资源，从而优化资源的使用效率，实现节省大规模仿真成本的目的。
The present application discloses a large-scale simulation resource adjustment method, device, equipment and storage medium. The method includes: S1: obtaining historical resource demand information of historical large-scale simulation; S2: generating current large-scale simulation data according to the resource demand historical information Resource demand prediction information; S3: Obtain real-time resource demand information of the current large-scale simulation; S4: Generate machine demand for the next large-scale simulation according to the resource demand forecast information and the real-time information of resource demand. The present application can effectively estimate the resources occupied by the next large-scale simulation operation, thereby optimizing the utilization efficiency of resources and realizing the purpose of saving large-scale simulation costs.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a large-scale simulation resource adjusting method, device, equipment and storage medium.
Background
Simulation is a frequently used technology in the field of automatic driving, and can improve the development efficiency and shorten the development time in the development of an automatic driving stack. Compared with the existing road test of the automatic driving safety personnel, which is widely applied, the resource, time and expense used by the test can be saved. But a small number of simulation scenarios may introduce bias and may not be sufficient to verify the functionality of the autopilot software stack. In an actual development process, a large-scale simulation (i.e., a simulation process with thousands of simulation embodiments) is required to perform the verification of the autopilot software function. Because the number of verification requirements is fluctuating, the verification requirements at working hours are more vigorous in normal working days, but the verification requirements at other times are more atrophic. The conventional method of using reserved resources causes a great amount of waste of resources in idle time. Worse, if the simulation demand fluctuation of the ordinary time is larger, the resource waste at the idle time is larger.
Disclosure of Invention
Therefore, the technical problem solved by the embodiments of the present application is to provide a method, an apparatus, a device and a storage medium for adjusting large-scale simulation resources, which can effectively estimate the resources occupied by the next large-scale simulation operation, thereby optimizing the utilization efficiency of the resources and achieving the purpose of saving the large-scale simulation cost.
In order to solve the technical problem, the technical scheme adopted by the application specifically comprises the following steps:
in one aspect, an embodiment of the present application provides a method for adjusting large-scale simulation resources, including:
s1, acquiring historical information of resource requirements of historical large-scale simulation;
s2, generating resource demand prediction information of the current large-scale simulation according to the resource demand historical information;
s3, acquiring real-time information of resource requirements of current large-scale simulation;
and S4, generating the machine requirement of the next large-scale simulation according to the resource requirement prediction information and the resource requirement real-time information.
Further, the resource demand history information comprises historical simulation time required by historical large-scale simulation and historical summarization of each resource required by the historical large-scale simulation; the S1 includes:
s11, collecting historical simulation time needed by historical large-scale simulation;
s12, collecting historical summaries of all resources required by the historical large-scale simulation;
and S13, storing historical simulation time and historical summaries of various resources.
Preferably, the resource demand prediction information includes prediction simulation time required by the current large-scale simulation and prediction summary of each resource required by the current large-scale simulation; the S2 includes:
s21, calculating the prediction simulation time required by the current large-scale simulation according to the historical simulation time required by the historical large-scale simulation;
s22, calculating the forecasting summary of each resource required by the current large-scale simulation according to the resource type and the historical summary of each resource required by the historical large-scale simulation;
and S23, storing the predicted simulation time and the prediction summary of each resource.
More preferably, the real-time information of resource demand includes real-time simulation time required by the current large-scale simulation and real-time summary of each resource required by the current large-scale simulation; the S3 includes:
s31, collecting the real-time simulation time needed by the current large-scale simulation;
s32, collecting the real-time summary of each resource needed by the current large-scale simulation;
and S33, storing the real-time simulation time and the real-time summary of each resource.
Further, the method for adjusting large-scale simulation resources according to the embodiment of the present application further includes S5, where S5 includes:
and optimizing the machine requirement of next large-scale simulation.
Preferably, the optimizing the machine requirements of the next large-scale simulation includes:
and S51, generating at least two different machine combination requests according to the machine requirements.
More preferably, the optimizing the machine requirements of the next large-scale simulation further includes:
and S52, selecting a proper machine combination request according to the running condition of the simulation system.
More preferably, the method for adjusting large-scale simulation resources according to the embodiment of the present application further includes S6, where S6 includes:
a machine demand is issued to the public cloud system.
In another aspect, an embodiment of the present application provides a large-scale simulation resource adjustment apparatus, including:
the first acquisition module is used for acquiring historical resource demand history information of historical large-scale simulation;
the first processing module is used for generating resource demand prediction information of current large-scale simulation according to the resource demand historical information;
the second acquisition module is used for acquiring the real-time information of the resource demand of the current large-scale simulation;
and the second processing module is used for generating the machine requirement of the next large-scale simulation according to the resource requirement prediction information and the resource requirement real-time information.
In another aspect, an embodiment of the present application provides an apparatus, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the steps of the large-scale simulation resource adjustment method according to any one of the above methods when executing the computer program.
In another aspect, an embodiment of the present application provides a storage medium, where a computer program is stored in the storage medium, and the computer program is executed by a processor to perform the steps of the method for adjusting large-scale simulation resources according to any one of the above methods.
In summary, compared with the prior art, the beneficial effects brought by the technical scheme provided by the embodiment of the present application at least include:
1. according to the embodiment of the application, the resource demand historical information of historical large-scale simulation is obtained, the resource demand prediction information of the current large-scale simulation is generated according to the resource demand historical information, the resource demand real-time information of the current large-scale simulation is obtained, and the machine demand of the next large-scale simulation is generated according to the resource demand prediction information and the resource demand real-time information, so that the resource occupied by the next large-scale simulation operation can be effectively estimated, the use efficiency of the resource is optimized, and the purpose of saving the large-scale simulation cost is achieved.
2. According to the embodiment of the application, the step of optimizing the machine requirement of next large-scale simulation is set, so that the use condition of the machine is further optimized, the use efficiency of the machine is improved, and the iteration cost of the automatic driving software stack is reduced.
3. According to the method and the device, the step of generating at least two different machine combination requests according to the machine requirements is set, so that the machine requirements of next large-scale simulation are optimized, and the selection of the machine requirements is more diversified and intelligent.
4. The embodiment of the application improves the use flexibility of the simulation system by setting the step of selecting the appropriate machine combination request according to the running condition of the simulation system.
Drawings
Fig. 1 is a schematic flowchart of a large-scale simulation resource adjustment method according to a first exemplary embodiment of the present application.
FIG. 2 is a flowchart illustrating a method for generating resource demand forecast information for current large-scale simulation according to resource demand history information in FIG. 1.
Fig. 3 is a schematic flowchart of a large-scale simulation resource adjustment method according to an eighth exemplary embodiment of the present application.
Fig. 4 is a schematic structural diagram of a large-scale simulation resource adjustment apparatus according to a twelfth exemplary embodiment of the present application.
Fig. 5 is a schematic structural diagram of an apparatus according to a thirteenth exemplary embodiment of the present application.
Detailed Description
The present embodiment is only for explaining the present application, and it is not limited to the present application, and those skilled in the art can make modifications of the present embodiment without inventive contribution as needed after reading the present specification, but all of them are protected by patent law within the scope of the claims of the present application.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "comprises," "comprising," or any other variation thereof, in the description and claims of this application, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the embodiments of the present application, words such as "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "e.g.," is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.
In the embodiments of the present application, the terms "first", "second", and the like are used for distinguishing identical items or similar items having substantially the same functions, and it should be understood that the terms "first", "second", and "N" have no logical or temporal dependency, and do not limit the number or execution order.
In the embodiments of the present application, the term "at least two" means two or more, and the meaning of "a plurality" means three or more.
The embodiments of the present application will be described in further detail with reference to the drawings attached hereto.
Fig. 1 is a large-scale simulation resource adjustment method provided in a first exemplary embodiment of the present application, and main steps of the method are described as follows:
s1, acquiring historical information of resource requirements of historical large-scale simulation;
s2, generating resource demand prediction information of the current large-scale simulation according to the resource demand historical information;
s3, acquiring real-time information of resource requirements of current large-scale simulation;
and S4, generating the machine requirement of the next large-scale simulation according to the resource requirement prediction information and the resource requirement real-time information.
According to the first exemplary embodiment of the application, resource demand history information of historical large-scale simulation is firstly obtained, resource demand prediction information of current large-scale simulation is generated according to the resource demand history information, resource demand real-time information of current large-scale simulation is then obtained, and finally machine demand of next large-scale simulation is generated according to the resource demand prediction information and the resource demand real-time information, so that resources occupied by next large-scale simulation operation can be effectively estimated, the use efficiency of the resources is optimized, and the purpose of saving large-scale simulation cost is achieved.
It should be noted that: the execution sequence of S1 to S4: the sequence described in this embodiment may be executed, i.e., S1 is executed first, then S2 is executed, then S3 is executed, and finally S4 is executed; in other embodiments, S1 and S3 may be performed in parallel, and then S2 and then S4 may be performed, or S1, then S3, then S2 and finally S4 may be performed first, or S3, then S1, then S2 and finally S4 may be performed first.
The second exemplary embodiment of the present application provides a large-scale simulation resource adjustment method, which is further improved on the basis of the first exemplary embodiment shown in fig. 1, and the specific improvements are as follows:
the resource demand history information comprises history simulation time required by history large-scale simulation and history summary of each resource required by history large-scale simulation; the S1 includes:
s11, collecting historical simulation time needed by historical large-scale simulation;
s12, collecting historical summaries of all resources required by the historical large-scale simulation;
and S13, storing historical simulation time and historical summaries of various resources.
The collection of resource demand history information required by historical large-scale simulation can be realized through the second exemplary embodiment of the application.
It should be noted that, the collecting of the history summaries of the resources required by the historical large-scale simulation means that the historical resource quantities required by each historical simulation instance related to the resource are respectively summed in the historical large-scale simulation according to the type of the resource to obtain the history summary corresponding to the resources.
It should be noted that the specific step of S1 is implemented by pcpuomethesus monitoring software.
The third exemplary embodiment of the present application provides a large-scale simulation resource adjustment method, which is further improved on the basis of the second exemplary embodiment of the present application, as shown in fig. 2, specifically improved as follows:
the resource demand prediction information comprises prediction simulation time required by current large-scale simulation and prediction summary of each resource required by the current large-scale simulation; the S2 includes:
s21, calculating the prediction simulation time required by the current large-scale simulation according to the historical simulation time required by the historical large-scale simulation;
s22, calculating the forecasting summary of each resource required by the current large-scale simulation according to the resource type and the historical summary of each resource required by the historical large-scale simulation;
and S23, storing the predicted simulation time and the prediction summary of each resource.
The third exemplary embodiment of the present application can realize the collection of the resource demand prediction information required by the current large-scale simulation.
The fourth exemplary embodiment of the present application provides a large-scale simulation resource adjustment method, which is further improved on the basis of the third exemplary embodiment of the present application, and the specific improvements are as follows:
the S21 includes:
the predicted simulation time required by the current large-scale simulation is defined to be a certain multiple of the historical simulation time required by the historical large-scale simulation.
For example, the predicted simulation time required by the current large-scale simulation is defined to be a times of the historical simulation time required by the past historical large-scale simulation, wherein a is a non-zero natural number, and a user can set a specific numerical value of a according to actual needs. Or setting the historical simulation time required by the historical large-scale simulation as the average of the historical simulation time required by the historical large-scale simulation of the past B times, and then defining the predicted simulation time required by the current large-scale simulation as A times of the average of the historical simulation time required by the historical large-scale simulation of the past B times. Similarly, A and B are non-zero natural numbers, and the user can set the specific numerical values of A and B according to the actual situation.
In order to accurately and exhaustively calculate the forecast summary of each resource required by the current large-scale simulation, the sixth exemplary embodiment of the present application provides a large-scale simulation resource adjustment method, which is further improved on the basis of the third exemplary embodiment of the present application, and the specific improvements are as follows:
the S22 includes:
s221, defining the predicted quantity of a certain resource required by the current simulation example as a certain multiple of the historical quantity of the corresponding resource required by the corresponding historical simulation example in the past;
s222, traversing all resources required by the current simulation example, and repeating S22 to obtain the predicted quantity of all resources required by the current simulation example;
and S223, traversing all current simulation examples of the current large-scale simulation, and respectively and correspondingly summing the same resources of all the current simulation examples in the current large-scale simulation according to the types of the resources to obtain the prediction summary of all the resources required by the current large-scale simulation.
The following example describes the specific steps of S22 in detail:
for the first resource required by the first current simulation example, the predicted quantity is C times of the historical quantity of the first resource required by the corresponding historical simulation example in the past, wherein C is a non-zero natural number, and a user can set the specific numerical value of C according to the actual situation.
It should be noted that: the C times of the historical quantity of the first resource required by the past corresponding historical simulation example can be set as C times of the historical quantity of the first resource required by the past corresponding historical simulation embodiment, or can be set as C times of the average historical quantity of the first resource required by the past corresponding historical simulation embodiments for multiple times.
For the second resource required by the first current simulation example, the predicted quantity is D times of the historical quantity of the second resource required by the corresponding historical simulation example in the past, wherein D is a non-zero natural number, and a user can set the specific numerical value of D according to the actual situation.
It should be noted that: the D times of the historical quantity of the second resource required by the past corresponding historical simulation example can be set to be D times of the historical quantity of the second resource required by the past corresponding historical simulation embodiment for one time, and also can be set to be D times of the average historical quantity of the second resource required by the past corresponding historical simulation embodiments for multiple times.
By analogy, traversing each resource required by the first current simulation instance until the last resource required by the first current simulation instance (namely the Nth resource required by the first current simulation instance) is predicted, wherein the predicted quantity of the resource is E times of the historical quantity of the Nth resource required by the corresponding historical simulation instance in the past, both E and N are non-zero natural numbers, and a user can set a specific numerical value of E according to actual needs.
It should be noted that: the E times of the historical quantity of the nth resource required by the historical simulation example corresponding to the past may be set to be E times of the historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past, or may be set to be E times of the average historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past for multiple times.
Then, for the first resource needed by the second current simulation example, the predicted quantity is F times of the historical quantity of the first resource needed by the corresponding historical simulation example in the past, wherein F is a non-zero natural number, and a user can set the specific numerical value of F according to the actual situation.
It should be noted that: the F times of the historical quantity of the first resource required by the past corresponding historical simulation example can be set as F times of the historical quantity of the first resource required by the past corresponding historical simulation embodiment, or can be set as F times of the average historical quantity of the first resource required by the past corresponding historical simulation embodiments for multiple times.
For the second resource required by the second current simulation example, the predicted quantity is G times of the historical quantity of the second resource required by the corresponding historical simulation example in the past, wherein G is a non-zero natural number, and a user can set the specific numerical value of G according to the actual situation.
It should be noted that: the G times of the historical quantity of the second resource required by the past corresponding historical simulation example can be set to be G times of the historical quantity of the second resource required by the past corresponding historical simulation embodiment for one time, and can also be set to be G times of the average historical quantity of the second resource required by the past corresponding historical simulation embodiments for multiple times.
By analogy, traversing each resource required by the second current simulation instance until the last resource required by the second current simulation instance (namely the Nth resource required by the second current simulation instance) is predicted to be H times of the historical quantity of the Nth resource required by the corresponding historical simulation instance in the past, wherein H and N are both non-zero natural numbers, and a user can set the specific numerical value of H according to the actual situation.
It should be noted that: the H times of the historical quantity of the nth resource required by the historical simulation example corresponding to the past can be set as H times of the historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past, or can be set as H times of the average historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past for multiple times.
By analogy, traversing the last current simulation example (namely the Mth current simulation example) required by the current large-scale simulation, and for the first resource required by the Mth current simulation example, the predicted quantity is I times of the historical quantity of the first resource required by the past corresponding historical simulation example, wherein I is a non-zero natural number, and a user can set a specific numerical value of I according to actual needs.
It should be noted that: the I times of the historical quantity of the first resource required by the past corresponding historical simulation example can be set as the I times of the historical quantity of the first resource required by the past corresponding historical simulation embodiment, or can be set as the I times of the average historical quantity of the first resource required by the past corresponding historical simulation embodiments for multiple times.
For the second resource required by the Mth current simulation example, the predicted quantity is J times of the historical quantity of the second resource required by the corresponding historical simulation example in the past, wherein J is a non-zero natural number, and a user can set the specific numerical value of J according to the actual situation.
It should be noted that: the J times of the historical quantity of the second resource required by the past corresponding historical simulation example can be set to be J times of the historical quantity of the second resource required by the past corresponding historical simulation embodiment for one time, and can also be set to be J times of the average historical quantity of the second resource required by the past corresponding historical simulation embodiments for multiple times.
By analogy, the resources required by the Mth current simulation instance are traversed until the last resource required by the Mth current simulation instance (namely the Nth resource required by the Mth current simulation instance) is predicted, the predicted quantity of the resources is K times of the historical quantity of the Nth resource required by the past corresponding historical simulation instance, K and N are non-zero natural numbers, and a user can set the specific numerical value of K according to the actual situation.
It should be noted that: the K times of the historical quantity of the nth resource required by the historical simulation example corresponding to the past may be set to be K times of the historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past, or may be set to be K times of the average historical quantity of the nth resource required by the historical simulation embodiment corresponding to the past for a plurality of times.
In summary, for the first resource required by the current large-scale simulation, the calculation formula of the prediction summary is as follows: the first resourcePrediction quantitySum (first resource)L) Wherein L is the Lth current simulation example required by the current large-scale simulation; l is a non-zero natural number, and the value range of L is [1, M]。
For the second resource required by the current large-scale simulation, the calculation formula of the prediction summary is as follows: a second resourcePrediction quantitySum (second resource)L) Wherein L is the Lth current simulation example required by the current large-scale simulation; l is a non-zero natural number, and the value range of L is [1, M]。
By analogy, for the last resource (i.e., nth resource) required by the current large-scale simulation, the calculation formula of the prediction summary is as follows: the Nth resourcePrediction quantitySum (nth resource)L) Wherein L is the Lth current simulation example required by the current large-scale simulation; l is a non-zero natural number, and the value range of L is [1, M]。
The large-scale simulation resource adjustment method provided by the seventh exemplary embodiment of the present application is further improved on the basis of the third exemplary embodiment of the present application, and the specific improvements are as follows:
the resource demand real-time information comprises real-time simulation time required by current large-scale simulation and real-time summarization of each resource required by the current large-scale simulation; the S3 includes:
s31, collecting the real-time simulation time needed by the current large-scale simulation;
s32, collecting the real-time summary of each resource needed by the current large-scale simulation;
and S33, storing the real-time simulation time and the real-time summary of each resource.
The method and the device can realize the collection of real-time information of resource requirements required by the current large-scale simulation through the seventh exemplary embodiment of the application.
It should be noted that the collecting of the real-time summaries of the resources required by the current large-scale simulation means that the real-time resource quantities required by each current simulation instance related to the resource are respectively summed in the current large-scale simulation according to the type of the resource to obtain the real-time summaries corresponding to the resources.
It should be noted that the specific step of S3 is implemented by PCpuometheus monitoring software.
Fig. 3 is a large-scale simulation resource adjustment method provided by an eighth exemplary embodiment of the present application, which is further improved on the basis of all the above embodiments of the present application, and the specific improvements are as follows:
the large-scale simulation resource adjustment method further includes S5, where S5 includes:
and optimizing the machine requirement of next large-scale simulation.
The eighth exemplary embodiment of the present application sets a step of optimizing the machine requirements of next large-scale simulation, so as to further optimize the use condition of the machine, improve the use efficiency of the machine, and reduce the iteration cost of the automatic driving software stack.
The ninth exemplary embodiment of the present application provides a large-scale simulation resource adjustment method, which is further improved on the basis of the eighth exemplary embodiment of the present application, and the specific improvements are as follows:
the optimizing of the machine requirements for the next large-scale simulation includes:
and S51, generating at least two different machine combination requests according to the machine requirements.
The ninth exemplary embodiment of the present application optimizes the machine requirements of the next large-scale simulation by setting a step of generating at least two different machine combination requests according to the machine requirements, thereby enabling the selection of the machine requirements to be more diversified and intelligent.
The large-scale simulation resource adjustment method provided by the tenth exemplary embodiment of the present application is further improved on the basis of the ninth exemplary embodiment of the present application, and the specific improvements are as follows:
the optimizing the machine requirements of the next large-scale simulation further comprises:
and S52, selecting a proper machine combination request according to the running condition of the simulation system.
The tenth exemplary embodiment of the present application improves the flexibility of use of the simulation system by providing a step of selecting an appropriate machine combination request according to the operation condition of the simulation system.
The large-scale simulation resource adjustment method provided by the eleventh exemplary embodiment of the present application is further improved on the basis of the tenth exemplary embodiment of the present application, and the specific improvements are as follows:
the large-scale simulation resource adjustment method further includes S6, where S6 includes:
a machine demand is issued to the public cloud system.
The eleventh exemplary embodiment of the present application enables the simulation system to accurately obtain the appropriate machine combination request in real time.
It should be noted that: in all of the above embodiments, the types of resources include, but are not limited to: CPU, memory, GPU, and flash memory. It will be apparent to those skilled in the art that the types of resources may also be any other hardware resources used to implement the simulation process.
The following describes in detail the flow steps of the large-scale simulation resource adjustment method in the above embodiment with reference to the specific resource type of the CPU, specifically as follows:
s1, acquiring historical CPU requirement historical information of historical large-scale simulation, specifically comprising:
during the running period of past large-scale simulation (namely, large-scale simulation), continuously collecting all the CPU quantity required by all simulation examples related to the simulation process by utilizing PCpuometheus monitoring software, and storing the CPU quantity in a storage system as CPU requirement historical information;
s2, generating the CPU demand forecasting information of the current large-scale simulation according to the CPU demand historical information, which specifically comprises the following steps:
for each current simulation instance of the current large-scale simulation, predicting that the CPU resource used by each current simulation instance is a certain multiple of the average CPU number required by the corresponding historical simulation instance in the past 7 times, namely, the CPU resource used by each current simulation instance is represented by the following formula (1):
CPUs＝Q×AVG7(CPUs) (1)；
wherein in the formula (1), s represents the s-th simulation example; q is a constant and represents a multiple;
then, the CPU prediction resources required by all current simulation instances of the current large-scale simulation, i.e., the CPU demand prediction information of the current large-scale simulation, are represented by the following formula (2):
CPUcuCpuCpuent_Cpueq＝Sum(CPUs) (2)；
s3, acquiring real-time information of CPU requirements of current large-scale simulation, specifically: during the current large-scale simulation operation, continuously collecting all the CPU quantities required to be used by all simulation examples related to the simulation process by utilizing PCpuometheus monitoring software, namely CPUcuCpuCpuentAnd the CPU is connected tocuCpuCpuentStoring the real-time information as CPU requirement in a storage system;
s4, generating the machine requirement of the next large-scale simulation according to the CPU requirement prediction information and the CPU requirement real-time information, which specifically comprises the following steps:
s41, according to CPU demand forecast information, it is CPUcuCpuCpuent_Cpueq＝Sum(CPUs) And the CPU requires real-time information as CPUcuCpuCpuent，Obtaining the number of CPUs required by the next large-scale simulation, namely, the number is represented by the following formula (3):
CPU＝max(CPUcuCpuCpuent，CPUcuCpuCpuent_Cpueq)+CPUCpueq (3)；
s42, generating corresponding machine requirements according to the quantity of the CPUs obtained in S41, wherein for example, the next large-scale simulation needs 100 CPUs in total, and in the public cloud operated by the large-scale simulation, 2 types of machines in total are respectively: providing a machine R of 16 CPUs and a machine T of 32 CPUs, the machine requirement of 3 × machine T +1 × machine R can be generated according to the above conditions;
s5 includes:
optimizing the machine requirement of next large-scale simulation;
since the machine types that can be applied for in the public cloud system where the large-scale simulation runs are heterogeneous, and the number of each type of machine is also limited and dynamically variable, optimizing the machine requirements of the next large-scale simulation mainly includes:
s51, generating at least two different machine combination requests according to the machine requirements, or taking the CPU as the case, the next large-scale simulation needs 100 CPUs in total, and in the public cloud system operated by the large-scale simulation, 2 types of machines are shared in total, wherein the types of machines are respectively as follows: providing a machine R of 16 CPUs and providing a machine T of 32 CPUs, four different machine combination requests of 3 × machine T +1 × machine R, 2 × machine T +3 × machine T, 1 × machine T +5 × machine R, and 0 × machine T +7 × machine R can be generated according to the above conditions.
S52, selecting a proper machine combination request according to the running condition of the simulation system, for example, in actual use, the larger the scale of the used machine is, namely the more the resources owned by the machine are, the higher the overall performance utilization rate of the simulation system is, so in the above different machine combination requests, the simulation system will preferentially apply for 3 machines T from the public cloud system, and if the application for 3 machines T can be successful, then apply for 1 machine R; however, if only 1 machine T can be applied, 5 machines R are continuously applied.
S6, sending machine requirements to the public cloud system, specifically: issuing machine requirements to the public cloud system according to the appropriate machine combination request.
By analogy, the types of other resources required by the large-scale simulation process, such as the memory, the GPU, and the flash memory, can be obtained according to the specific steps from S1 to S5, so as to obtain the types of machines and the number of machines required to be used.
In general, the algorithm for generating the machine requirements in the embodiments of the present application is based on the following general principles:
(1) because the GPU is a limited resource, the GPU resource requirement is met preferentially;
(2) because the simulation system cannot compress the memory during operation and only generates an exception, the embodiment of the application meets the memory requirement in the resource allocation;
(3) and finally, the requirements of other resources are met.
Fig. 4 is a large-scale simulation resource adjustment apparatus provided in a twelfth exemplary embodiment of the present application, where the adjustment apparatus corresponds to the adjustment methods in the foregoing embodiments one to one, and the adjustment apparatus includes:
the first acquisition module is used for acquiring historical resource demand history information of historical large-scale simulation;
the first processing module is used for generating resource demand prediction information of current large-scale simulation according to the resource demand historical information;
the second acquisition module is used for acquiring the real-time information of the resource demand of the current large-scale simulation;
and the second processing module is used for generating the machine requirement of the next large-scale simulation according to the resource requirement prediction information and the resource requirement real-time information.
The modules of the adjusting device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 5 is a device, which may be a server, provided in a thirteenth exemplary embodiment of the present application. The device includes a processor, a memory, and a communication interface connected by a system bus. Wherein the processor of the device is configured to provide computing and control capabilities. The memory of the device may be implemented by any type or combination of volatile or non-volatile storage devices, including but not limited to: magnetic disk, optical disk, EEPCPUOM, EPCPUOM, SCPUAM, CPUOM, magnetic memory, flash memory, and PCPUOM. The memory of the device provides an environment for the running of an operating system and computer programs stored within it. The communication interface of the device is a network interface, and the network interface is used for connecting and communicating with an external terminal through a network. The computer program, when executed by a processor, implements the steps of the adaptation method described in the above embodiments.
In a further embodiment of the present application, a storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the adaptation method described in the above embodiments. Such storage media include, but are not limited to: CPUOM, CPUAM, CD-CPUOM, diskette, and floppy disk.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of each functional unit or module is illustrated, and in practical applications, the above-mentioned function may be distributed as different functional units or modules as required, that is, the internal structure of the apparatus described in this application may be divided into different functional units or modules to implement all or part of the above-mentioned functions.
While the present embodiments have been described with reference to the accompanying drawings, it is to be understood that the invention is not limited to the precise embodiments described above, which are meant to be illustrative and not restrictive, and that various changes may be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (11)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111232829.0A CN114154678A (en) | 2021-10-22 | 2021-10-22 | A large-scale simulation resource adjustment method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111232829.0A CN114154678A (en) | 2021-10-22 | 2021-10-22 | A large-scale simulation resource adjustment method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114154678A true CN114154678A (en) | 2022-03-08 |
Family
ID=80458543
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111232829.0A Pending CN114154678A (en) | 2021-10-22 | 2021-10-22 | A large-scale simulation resource adjustment method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114154678A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105991667A (en) * | 2015-01-27 | 2016-10-05 | 华为软件技术有限公司 | Method and device for correcting resource prediction error |
| CN110007929A (en) * | 2018-01-02 | 2019-07-12 | 中国移动通信有限公司研究院 | The method and device of resource is obtained under a kind of mixed deployment |
| CN113158435A (en) * | 2021-03-26 | 2021-07-23 | 中国人民解放军国防科技大学 | Complex system simulation running time prediction method and device based on ensemble learning |
| CN113378498A (en) * | 2021-08-12 | 2021-09-10 | 新华三半导体技术有限公司 | Task allocation method and device |
-
2021
- 2021-10-22 CN CN202111232829.0A patent/CN114154678A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105991667A (en) * | 2015-01-27 | 2016-10-05 | 华为软件技术有限公司 | Method and device for correcting resource prediction error |
| CN110007929A (en) * | 2018-01-02 | 2019-07-12 | 中国移动通信有限公司研究院 | The method and device of resource is obtained under a kind of mixed deployment |
| CN113158435A (en) * | 2021-03-26 | 2021-07-23 | 中国人民解放军国防科技大学 | Complex system simulation running time prediction method and device based on ensemble learning |
| CN113378498A (en) * | 2021-08-12 | 2021-09-10 | 新华三半导体技术有限公司 | Task allocation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alarifi et al. | Energy-efficient hybrid framework for green cloud computing | |
| Toosi et al. | Revenue maximization with optimal capacity control in infrastructure as a service cloud markets | |
| US7752239B2 (en) | Risk-modulated proactive data migration for maximizing utility in storage systems | |
| US8214327B2 (en) | Optimization and staging method and system | |
| Huang et al. | Cap3: A cloud auto-provisioning framework for parallel processing using on-demand and spot instances | |
| US9374314B2 (en) | QoS aware balancing in data centers | |
| US20130007244A1 (en) | Optimizing provisioning workflows in cloud computing | |
| Bi et al. | SLA-based optimisation of virtualised resource for multi-tier web applications in cloud data centres | |
| CN113515382B (en) | Cloud resource allocation method and device, electronic equipment and storage medium | |
| CN114090239B (en) | Method and device for dispatching edge resources based on model reinforcement learning | |
| CN104035747B (en) | Method and apparatus for parallel computation | |
| CN108459905A (en) | Resource pool method for planning capacity and server | |
| CN103473115A (en) | Virtual machine placing method and device | |
| CN114741955A (en) | Multi-objective optimization task scheduling method based on security cloud | |
| Venugopal et al. | A deadline and budget constrained scheduling algorithm for eScience applications on data grids | |
| CN109634714B (en) | An intelligent scheduling method and device | |
| CN119323336A (en) | Virtual power plant demand response potential evaluation method, system, terminal and storage medium | |
| CN116360921A (en) | Cloud platform resource optimal scheduling method and system for electric power Internet of things | |
| CN113535346A (en) | Method, device, device and computer storage medium for adjusting the number of threads | |
| CN119883640A (en) | Dynamic resource scheduling system and method based on decoupling of CPU and memory | |
| Mehta | Efficient resource scheduling in cloud computing | |
| Singh et al. | Value and energy optimizing dynamic resource allocation in many-core HPC systems | |
| CN114154678A (en) | A large-scale simulation resource adjustment method, device, equipment and storage medium | |
| CN108536525A (en) | A kind of host dispatching method and device | |
| Luo et al. | An enhanced workflow scheduling strategy for deadline guarantee on hybrid grid/cloud infrastructure |
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 |