CN114416361A - Adjusting method for reducing load in high-concurrency data scene of fusion terminal - Google Patents

Adjusting method for reducing load in high-concurrency data scene of fusion terminal Download PDF

Info

Publication number
CN114416361A
CN114416361A CN202210027801.1A CN202210027801A CN114416361A CN 114416361 A CN114416361 A CN 114416361A CN 202210027801 A CN202210027801 A CN 202210027801A CN 114416361 A CN114416361 A CN 114416361A
Authority
CN
China
Prior art keywords
data
delay interval
data type
app
request
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
Application number
CN202210027801.1A
Other languages
Chinese (zh)
Inventor
不公告发明人
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Leaguer Microelectronics Co ltd
Original Assignee
Leaguer Microelectronics Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Leaguer Microelectronics Co ltd filed Critical Leaguer Microelectronics Co ltd
Priority to CN202210027801.1A priority Critical patent/CN114416361A/en
Publication of CN114416361A publication Critical patent/CN114416361A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083—Techniques for rebalancing the load in a distributed system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/546—Message passing systems or structures, e.g. queues

Abstract

The invention discloses a method for adjusting the reduction of load in a high concurrent data scene of a fusion terminal, which comprises the following steps: s1, initializing; s2, writing the data type A into the request WAPut request queue QA(ii) a S3, starting the first timer according to the estimated maximum time delay interval MT1 to monitor whether there is a similar request, if so, inputting MT1 to MAEstimating the maximum time delay interval of the next operation and returning to S2; if not, the first timer finishes timing, the monitoring is continued without finishing, and S5 is finished; s5, get QAAll the request tasks of A are written into the data center, and the update notification of A is sent to each subscription APP after the writing is finished; s6, after receiving the update notification, the subscription APP sets the effective mark of the A to be overdue, starts a second timer, monitors whether the APP has read operation on the A, if yes, the subscription APP proceeds to S9, and if not, the subscription APP proceeds to S7; s7, judging whether the second timer finishes timing or not, continuing monitoring if the second timer does not finish timing, and entering S8; s8, judging whether the current data center is idle or not, and continuing monitoring without idleIdles at S9; and S9, reloading the data of the A from the data center by the subscription APP.

Description

Adjusting method for reducing load in high-concurrency data scene of fusion terminal
Technical Field
The invention relates to the technical field of intelligent electric meter communication, in particular to a method for adjusting low load in a high-concurrency data scene of a convergence terminal and a related computer-readable storage medium.
Background
The intelligent integration terminal (SCU) of the transformer area continues the design concept of 'hardware modularization and software APP' of an energy controller, is researched and designed by combining physical hardware with high main frequency and large storage, and has a plurality of advanced characteristics of distributed architecture, data and service separation, matrix type application expansion, cross-platform service interaction and the like. The hardware modularized design can realize flexible access of metering and sensing equipment at a client side and a power distribution side, and has the service functions of data acquisition, intelligent cost control, clock synchronization, accurate metering, ordered charging, energy utilization management, loop state inspection, household variable relation identification, power failure event reporting and the like; the software APP design supports the exchange of the same APP of different manufacturers and supports the expansion installation of the APP of a new service. The product can be widely applied to the fields of power utilization information acquisition, electric energy quality analysis, comprehensive energy metering acquisition, distributed energy management, orderly charging of electric automobiles and the like.
For traditional concentrator terminal equipment, a plurality of processes are usually adopted for working, data interaction among the processes is realized in a memory sharing or file sharing mode, and when data is updated, a process for writing data sends a notice to a process for reading data in an inter-process communication mode. For example: the method comprises the steps that a user continuously adds 100 pieces of table equipment to a terminal through a master station, according to the original scheme, on a new intelligent fusion terminal of a transformer area, every time the user operates the added equipment, a terminal APP operates data center write operation once, and the data center sends notification to other subscribed APPs, so that the data center is operated 100 times in the current operation, 100 notifications are sent to the APPs, and the APPs load data of the data center 100 times; however, each APP usually does not need to use the updated data immediately at this time, and if all APPs are loaded at the same time, a large amount of IO resources of the system will be occupied, resulting in a tidal effect and a performance bottleneck, so that the APP which really needs to update the data cannot acquire the data in time.
Disclosure of Invention
In view of this, the invention provides a method for adjusting the reduction of load in a high-concurrency data scene of an intelligent platform area convergence terminal, so as to solve the technical problems in the prior art.
In order to solve the technical problems, the invention adopts the following technical scheme:
a method for adjusting the reduction of load in a high concurrent data scene of a convergence terminal comprises the following steps: s1, the front-end data processing APP initializes a time delay interval statistical module and a request queue corresponding to each data type; the APP is an application program running on the terminal; s2, for any data type a in the data types, the front-end data processing APP receives a write request W from the front end to the data type aAPut into the corresponding request queue QAPerforming the following steps; s3, time delay interval statistical module M of front end data processing APP according to data type AAAnalyzing the estimated maximum time delay interval of the time, starting a first timer, and monitoring whether a continuous writing request of the data type A exists or not; if yes, executing step S4.1; if not, executing the step S4.2; s4.1, inputting the time delay interval sample of the operation to a corresponding time delay interval statistical module MATime delay interval statistic module MAAnalyzing and estimating the maximum time delay interval of the next operation, and then returning to the step S2; s4.2, judging whether the first timer finishes timing; if yes, go to step S5; if not, continuously monitoring whether continuous write requests of the data type A exist or not; s5, the front end data processing APP takes out the request queue QAAll the request tasks of the data type A are uniformly written into a data center in a transaction batch processing mode, and after the writing is finished, a data type A updating notice is sent to all APPs subscribing the data type A in a message queue telemetering transmission mode; s6, after receiving the data type A updating notification, the APP subscribing the data type A sets the effective mark of the data type A to be overdue, starts a second timer, and monitors whether the APP has a reading operation on the data type A next time; if the read operation on the data type a is monitored, executing step S9; if the read operation on the data type a is not monitored, executing step S7; s7, judging whether the second timer finishes timing; if not, continuing monitoring; if yes, go to step S8; s8, judgmentJudging whether the current data center APP is idle or not; if not, continuing monitoring; if yes, go to step S9; and S9, reloading the data of the data type A from the data center by the APP which receives the update notification of the data type A.
Further, in step S1, initialization is performed when the terminal leaves the factory, where the initialization includes setting an initial default delay interval, and initializing the initial statistical sample number of each module of the front-end data processing APP to 0.
Further, in step S2, the write requests of each data type enter the corresponding request queue using a first-in first-out strategy to ensure that each task is executed in order.
Further, in step S2, the write request of each data type is not limited to the follow-up copy request of the master station front end, but also includes LCD interface operation, a copy-ordering task, or event reporting of the terminal.
Further, in step S3, the front-end data processing APP uses an asynchronous mode to monitor the front end, suspend the current thread, and wake up when monitoring the write request, so as to improve the utilization rate of the CPU.
Further, in step S3, one request queue corresponds to only one timer; for a request queue, when a new request of the same type arrives, the timer of the queue does not end the timing, but restarts the timing according to a new delay value.
Further, the delay interval statistic module M in step S41AAnalyzing and estimating the maximum delay interval of the next operation, comprising:
time delay interval statistic module MAThe operating delay interval samples of each input obey a standard normal distribution, i.e.
Figure BDA0003464980830000031
Figure BDA0003464980830000032
Statistic module for time delay intervalBlock MAWith an expectation of μ and a standard deviation of σ, the maximum delay interval MT2 for the next operation is calculated as follows:
MT2＝μ+3σ+MT1*k
wherein, MT1 is the input delay interval sample of this operation, and k is the correlation factor with the last operation.
Further, the delay interval statistic module M in step S41AWhen analyzing and estimating the maximum delay interval of the next operation, if it is the first input sample, since there are no expected μ, standard deviation σ and MT1, the calculation formula of MT2 is:
MT2＝MT
wherein, the MT is a default delay interval when the terminal leaves the factory.
Further, in step S5, the data center is used to perform write once submission, and if the attempt fails multiple times, the request queue Q continues to be reservedAAnd restarts the counting of the first timer, and the process returns to the step S3.
The present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, can implement the steps of the foregoing method for adjusting a low load in a high-concurrency data scene of a convergence terminal.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, the time delay interval of each operation at the front end is intelligently analyzed and estimated through the terminal, a large amount of repeated operations of the same type are filtered, and the network IO load of the terminal in a high concurrent data scene is greatly reduced. The maximum time delay interval of the next front-end operation can be accurately estimated through analysis of historical statistics values and actual values of the last time, and then the purpose of duplicate removal is achieved through detecting whether the current operation is the previous operation of the same type, so that data updating notification interaction between terminal APPs is greatly reduced, the utilization rate of network IO is improved, and the guarantee is provided for other services of the terminal, such as meter reading, event reporting and the like.
In addition, the invention fully utilizes the idle time of each APP for reading data and the idle time of the data center request through the strategy of delaying the loading of data by the terminal, thereby avoiding the high concurrency behavior that a large number of APPs simultaneously operate the same data type of the data center at the same time, and greatly reducing the load of the data center.
Drawings
Fig. 1 is a flowchart of a method for adjusting a low load in a high-concurrency data scene of a convergence terminal according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and detailed description of embodiments.
As shown in fig. 1, an embodiment of the present invention provides a method for adjusting a low load in a high-concurrency data scene of a convergence terminal, including the following steps S1 to S9:
s1, the front-end data processing APP initializes a time delay interval statistical module and a request queue corresponding to each data type; the APP is an application program running on a terminal, the terminal is a distribution area intelligent fusion terminal, and three types of APPs of other services, namely a front-end data processing APP, a data center APP and an APP for subscribing data update notification of the data center APP, run on the terminal. The front-end data processing APP is responsible for receiving and processing a request of the front end.
The data type of the platform intelligent convergence terminal may include multiple types according to actual service conditions, such as a daily curve collection scheme, a minute curve collection task, or electric meter archive information, and the like, but is not limited thereto. The subsequent steps apply to either data type.
The step S1 is initialized to be specifically as follows: when the terminal leaves a factory, the number of initial statistical samples of each module of the front-end data processing APP is 0, the initial default delay interval is recorded as MT, the initial default delay interval is generally set to 2 seconds, and the terminal can also be modified by custom loading of an existing configuration file. In addition, the statistical correlation values of the modules can be serialized into a file in real time, and the terminal can be continuously used when being powered on and started again even if events such as power failure occur in the terminal, so that the method has strong recoverability.
S2, for any data type A in the data types, the front end numberAccording to the processing APP, the received write request W of the front end to the data type AAPut into the corresponding request queue QAIn (1). The concrete expression is as follows: when the terminal leaves a factory, each request queue has no task, each type of request can be continuously stored in the corresponding queue along with the time, a first-in first-out strategy is used to ensure that each task is orderly executed, and it should be noted that the request is not limited to a copy-along request at the front end of the main station, and may also be an LCD interface operation, a copy-determining task, an event report and the like of the terminal.
S3, time delay interval statistical module M of front end data processing APP according to data type AAAnalyzing the estimated maximum time delay interval of the time, starting a first timer, and monitoring whether a continuous writing request of the data type A exists or not; if yes, executing step S4.1; if not, step S4.2 is executed. The concrete expression is as follows: the front-end data processing APP monitors the front end in an asynchronous mode, suspends the current thread, and immediately wakes up when a request comes, so that the utilization rate of a CPU is greatly improved; in addition, it should be noted that a request queue only corresponds to a Timer, and when a new request of the same type arrives, the Timer of the queue does not end the timing, but restarts the timing according to a new delay value.
S4.1, inputting the time delay interval sample of the operation to a corresponding time delay interval statistical module MATime delay interval statistic module MAThe analysis estimates the maximum delay interval for the next operation and then returns to step S2. The calculation method for analyzing and estimating the maximum time delay interval MT2 of the next operation comprises the following steps:
time delay interval statistic module MAThe operation delay interval samples of each input can be considered to follow a standard normal distribution, namely:
Figure BDA0003464980830000051
P(ak|yi)＝ɡ(ak,ηyi,σyi)
statistic module M for statistic delay intervalAWith an expectation of μ and a standard deviation of σ, the maximum delay interval MT2 for the next operation is calculated as follows:
MT2＝μ+3σ+MT1*k
the MT1 is an input delay interval sample of the current operation, k is a correlation factor with the previous operation, and a general default value is 0.1, which can also be configured by user according to the data type.
It should be noted that if the sample is input for the first time, since μ, standard deviation σ and MT1 are not expected, the calculation formula of MT2 is:
MT2＝MT
the MT is a default delay interval of the terminal when leaving the factory, and is generally 2 seconds.
S4.2, judging whether the first timer finishes timing; if yes, go to step S5; if not, continuously monitoring whether continuous write requests of the data type A exist or not;
s5, the front end data processing APP takes out the request queue QAAnd uniformly writing all the taken request tasks into a data center in a transaction batch processing mode, and sending a data type A updating notice to each APP subscribing the data type A in a message queue telemetry transmission mode after the writing is finished. The concrete expression is as follows: write-once commit using a batch interface of a data center, and if multiple attempts fail, continuing to reserve queue QAAnd restarts the counting of the first timer, and the process returns to the step S3. At this time, it should be noted that the maximum delay interval value in step S3 will not change.
S6, after receiving the data type A updating notification, the APP subscribing the data type A sets the effective mark of the data type A to be overdue, starts a second timer, and monitors whether the APP has a reading operation on the data type A next time; if the read operation on the data type A is monitored, the step S9 is executed, so that the APP can obtain correct latest data in time while the data load is balanced, and the generation of dirty data is avoided; if the read operation on the data type a is not monitored, executing step S7;
s7, judging whether the second timer finishes timing; if not, continuing monitoring; if yes, go to step S8;
s8, judging whether the current data center APP is idle or not; the concrete expression is as follows: the terminal service APP can check whether an uncompleted task which is being executed exists in a current data center operation queue, if yes, the data center operation is considered to be busy, namely, the data center operation queue is not idle, and monitoring is continued; if the current data center APP is detected to be idle, executing step S9;
and S9, reloading the data of the data type A from the data center by the APP receiving the update notification of the data type A, and updating the memory.
The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for adjusting a low load in a high-concurrent data scene of a convergence terminal according to the foregoing embodiment may be implemented.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several equivalent substitutions or obvious modifications can be made without departing from the spirit of the invention, and all the properties or uses are considered to be within the scope of the invention.

Claims (10)

1. A method for adjusting the reduction of load in a high concurrent data scene of a convergence terminal is characterized by comprising the following steps:
s1, the front-end data processing APP initializes a time delay interval statistical module and a request queue corresponding to each data type; the APP is an application program running on the terminal;
s2, for any data type a in the data types, the front-end data processing APP receives a write request W from the front end to the data type aAPut into the corresponding request queue QAPerforming the following steps;
s3, the front end data processing APP rootTime delay interval statistical module M according to data type AAAnalyzing the estimated maximum time delay interval of the time, starting a first timer, and monitoring whether a continuous writing request of the data type A exists or not; if yes, executing step S4.1; if not, executing the step S4.2;
s4.1, inputting the time delay interval sample of the operation to a corresponding time delay interval statistical module MATime delay interval statistic module MAAnalyzing and estimating the maximum time delay interval of the next operation, and then returning to the step S2;
s4.2, judging whether the first timer finishes timing; if yes, go to step S5; if not, continuously monitoring whether continuous write requests of the data type A exist or not;
s5, the front end data processing APP takes out the request queue QAAll the request tasks of the data type A are uniformly written into a data center in a transaction batch processing mode, and after the writing is finished, a data type A updating notice is sent to all APPs subscribing the data type A in a message queue telemetering transmission mode;
s6, after receiving the data type A updating notification, the APP subscribing the data type A sets the effective mark of the data type A to be overdue, starts a second timer, and monitors whether the APP has a reading operation on the data type A next time; if the read operation on the data type a is monitored, executing step S9; if the read operation on the data type a is not monitored, executing step S7;
s7, judging whether the second timer finishes timing; if not, continuing monitoring; if yes, go to step S8;
s8, judging whether the current data center APP is idle or not; if not, continuing monitoring; if yes, go to step S9;
and S9, reloading the data of the data type A from the data center by the APP which receives the update notification of the data type A.
2. The method for adjusting low load in a converged terminal high concurrent data scenario according to claim 1, wherein in step S1, initialization is performed when the terminal leaves a factory, which includes setting an initial default delay interval, and initializing the initial statistical sample number of each module of the front-end data processing APP to 0.
3. The method for adjusting load reduction in a high-concurrency data scenario of a convergence terminal according to claim 1, wherein in step S2, the write requests of each data type enter the corresponding request queue by using a first-in first-out strategy to ensure that each task is executed in order.
4. The method for adjusting the load reduction in the high-concurrency data scene of the convergence terminal as claimed in claim 1, wherein in step S2, the write request of each data type is not limited to the follow-up copy request of the front end of the master station, and further comprises LCD interface operation, a copy-ordering task or event report of the terminal.
5. The method for adjusting the load reduction in the high-concurrency data scenario of the convergence terminal as claimed in claim 1, wherein in step S3, the front-end data processing APP uses an asynchronous mode to monitor the front end, suspend the current thread, and wake up when monitoring the write request, so as to improve the utilization rate of the CPU.
6. The method for adjusting load reduction in a converged terminal high concurrent data scenario according to claim 5, wherein in step S3, one request queue corresponds to only one timer; for a request queue, when a new request of the same type arrives, the timer of the queue does not end the timing, but restarts the timing according to a new delay value.
7. The method for adjusting the load reduction in the high-concurrency data scenario of the convergence terminal according to claim 1, wherein the time delay interval statistics module M in step S41AAnalyzing and estimating the maximum delay interval of the next operation, comprising:
time delay interval statistic module MAThe operating delay interval samples of each input obey a standard normal distribution, i.e.
Figure FDA0003464980820000021
Figure FDA0003464980820000022
Statistic module M for statistic delay intervalAWith an expectation of μ and a standard deviation of σ, the maximum delay interval MT2 for the next operation is calculated as follows:
MT2＝μ+3σ+MT1*k
wherein, MT1 is the input delay interval sample of this operation, and k is the correlation factor with the last operation.
8. The method for adjusting load reduction in the high-concurrency data scenario of the converged terminal according to claim 7, wherein the time delay interval statistics module M in step S41AWhen analyzing and estimating the maximum delay interval of the next operation, if it is the first input sample, since there are no expected μ, standard deviation σ and MT1, the calculation formula of MT2 is:
MT2＝MT
wherein, the MT is a default delay interval when the terminal leaves the factory.
9. The method for adjusting load reduction in the high-concurrency data scenario of the converged terminal according to claim 1, wherein in step S5, a batch processing interface of the data center is used for write-once submission, and if multiple attempts fail, the request queue Q continues to be reservedAAnd restarts the counting of the first timer, and the process returns to the step S3.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, is capable of implementing the steps of the method for throttling for reducing load in a converged terminal high concurrency data scenario according to any one of claims 1 to 9.
CN202210027801.1A 2022-01-11 2022-01-11 Adjusting method for reducing load in high-concurrency data scene of fusion terminal Pending CN114416361A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210027801.1A CN114416361A (en) 2022-01-11 2022-01-11 Adjusting method for reducing load in high-concurrency data scene of fusion terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210027801.1A CN114416361A (en) 2022-01-11 2022-01-11 Adjusting method for reducing load in high-concurrency data scene of fusion terminal

Publications (1)

Publication Number Publication Date
CN114416361A true CN114416361A (en) 2022-04-29

Family

ID=81271377

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210027801.1A Pending CN114416361A (en) 2022-01-11 2022-01-11 Adjusting method for reducing load in high-concurrency data scene of fusion terminal

Country Status (1)

Country Link
CN (1) CN114416361A (en)

Similar Documents

Publication Publication Date Title
Ferscha et al. Estimating rollback overhead for optimism control in Time Warp
CN101290668B (en) Time sharing operation dynamic dispatching method and device
CN101981531A (en) Aggregating recurrent schedules to optimize resource consumption
CN109117280A (en) The method that is communicated between electronic device and its limiting process, storage medium
Wang et al. A cluster autoscaler based on multiple node types in kubernetes
CN115994053A (en) Parallel playback method and device of database backup machine, electronic equipment and medium
CN111475213A (en) Power consumption reduction method and device for multi-core structure solid state disk and computer equipment
CN109117279A (en) The method that is communicated between electronic device and its limiting process, storage medium
CN110096339B (en) System load-based capacity expansion and contraction configuration recommendation system and method
US20220129745A1 (en) Prediction and Management of System Loading
CN102681650B (en) The storage system of a kind of Energy control power-economizing method and correspondence thereof
US8352398B2 (en) Time-based conflict resolution
CN114416361A (en) Adjusting method for reducing load in high-concurrency data scene of fusion terminal
CN111858656A (en) Static data query method and device based on distributed architecture
CN112035255A (en) Thread pool resource management task processing method, device, equipment and storage medium
CN110990227A (en) Numerical pool application characteristic performance acquisition and monitoring system and operation method thereof
CN104427595A (en) Communication-terminal standby electricity-saving control method and control device
CN110009477B (en) Interest meter lifting method, system and equipment
CN112711606A (en) Database access method and device, computer equipment and storage medium
CN110674214A (en) Big data synchronization method and device, computer equipment and storage medium
CN113835896B (en) Iterative computation-oriented parallelism dynamic adjustment method in Gaia system
CN115543450B (en) Method for dynamically dormancy of server-free calculation of resource optimization in edge scene and application
CN102221875A (en) Microprocessor, method of operating the microprocessor and computer program product
CN116526678B (en) Intelligent computing center power supply elastic scheduling system and control method thereof
CN113225228B (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