CN114510350A - Intelligent load processing system based on server - Google Patents

Intelligent load processing system based on server Download PDF

Info

Publication number
CN114510350A
CN114510350A CN202210104445.9A CN202210104445A CN114510350A CN 114510350 A CN114510350 A CN 114510350A CN 202210104445 A CN202210104445 A CN 202210104445A CN 114510350 A CN114510350 A CN 114510350A
Authority
CN
China
Prior art keywords
server
task
servers
tertiary
characteristic value
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
CN202210104445.9A
Other languages
Chinese (zh)
Inventor
徐振华
刘海燕
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.)
BEIJING INFORMATION TECHNOLOGY COLLEGE
Original Assignee
BEIJING INFORMATION TECHNOLOGY COLLEGE
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 BEIJING INFORMATION TECHNOLOGY COLLEGE filed Critical BEIJING INFORMATION TECHNOLOGY COLLEGE
Priority to CN202210104445.9A priority Critical patent/CN114510350A/en
Publication of CN114510350A publication Critical patent/CN114510350A/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
    • G06F9/505—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 load
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/5013—Request control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/503—Resource availability
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/50—Indexing scheme relating to G06F9/50
    • G06F2209/508—Monitor

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multi Processors (AREA)

Abstract

The invention relates to a server-based intelligent load processing system, which comprises a primary server, secondary servers and tertiary servers, wherein the primary server distributes a task request to a target secondary server according to the task characteristic value of the task request and the server characteristic value of each secondary server; the secondary server distributes the task request according to the load of the subordinate tertiary server, and dynamically updates the server characteristic value according to the task execution state. The system improves the overall operation efficiency of the system through intelligent load processing.

Description

Intelligent load processing system based on server
[ technical field ] A method for producing a semiconductor device
The invention belongs to the technical field of computers, and particularly relates to an intelligent load processing system based on a server.
[ background of the invention ]
With the rapid development of computer technology in various industries, a large amount of task processing work required to be completed by a computer appears, and in a common architecture of computer task processing, a server is generally used to receive a certain task sent by a client, then the server executes a corresponding task processing process, accesses data information related to the task (for example, reads task data stored in a database), and returns an execution result of the process, that is, a task processing result to the client or writes the task processing result into the database.
When a large number of tasks need to be processed simultaneously, a single server is hard to bear, and at this time, a system architecture capable of processing tasks in parallel, such as a server cluster, is usually adopted. At this time, a load processing system of one server needs to be introduced to perform load management on the server cluster, and distribute the received tasks to different servers for processing, so as to improve the task processing efficiency. The existing load processing system usually obtains the current load of each server and distributes a new task to the server with the smallest load, so that the load of each server is as close as possible to achieve load balance. However, the inventor finds that, in the actual working process, the simple load balancing method in the prior art still leaves room for improvement, for example, in many cases, two servers respectively execute the same task, the task needs to access the same task data in the database, so that the two servers respectively need to read the same data in the database, and the repeatedly executed work affects the overall processing efficiency of the system.
[ summary of the invention ]
In order to solve the above problems in the prior art, the present invention provides an intelligent load processing system based on a server.
The technical scheme adopted by the invention is as follows:
an intelligent load processing system based on server comprises a primary server, K secondary servers and a plurality of tertiary servers, wherein the primary servers, the K secondary servers and the plurality of tertiary servers are connected in series
The primary server is connected with a plurality of secondary servers and is communicated with the plurality of secondary servers, and the primary server is used for receiving task requests and distributing the secondary servers to process the task requests according to the task requests;
each secondary server is subordinate to a plurality of tertiary servers, and each tertiary server is subordinate to only one secondary server; the secondary server receives a task request sent by the primary server and distributes a subordinate tertiary server to process the task request according to the task request;
the system for intelligent load processing comprises the following specific steps:
(1) calculating a corresponding initial task characteristic value for each secondary server;
(2) respectively distributing the K initial task characteristic values to K secondary servers, and setting the distributed initial task characteristic values as server characteristic values by the secondary servers;
(3) the primary server receives the task request and calculates a corresponding task characteristic value FV according to the task request;
(4) the primary server determines a current available secondary server set;
(5) the primary server calculates the task feature values FV and server feature values SV for each available secondary serveriDistance D ofiI is more than or equal to 1 and less than or equal to M, and M is the number of available secondary servers;
(6) the primary server determines the distance DiDetermining the available secondary server corresponding to the minimum value as the target server;
(7) the primary server distributes the task request to the target server, and the target server distributes the task request to the tertiary server with the minimum load according to the load of the tertiary server under the target server;
(8) and each secondary server updates the corresponding server characteristic value according to the task execution state of the subordinate tertiary server.
Further, the step 1 includes:
(1.1) collecting a certain number of tasks to generate an initial task set;
(1.2) calculating a task characteristic value of each task in an initial task set to obtain a characteristic value set, clustering the characteristic value set, and dividing the characteristic value set into K characteristic value categories;
and (1.3) for each characteristic value category, calculating the clustering center of the category to obtain an initial task characteristic value.
Further, a certain number of historical tasks are collected from the operation process of other systems to form an initial task set.
Further, a certain number of tasks are randomly generated to form an initial task set.
Further, the cluster center of the feature value category is the average of all feature values in the category.
Further, in step 4, the primary server determines available secondary servers according to a current average task amount of each secondary server, where the current average task amount refers to an average value of currently executed task amounts of all tertiary servers under the secondary server.
Further, the primary server determines a secondary server with the current average task amount smaller than a preset threshold value as an available secondary server; and if the secondary servers with the current average task quantity smaller than the preset threshold value do not exist, the primary server determines a plurality of secondary servers with the minimum current average task quantity as available servers.
Further, the load of the tertiary server is the number of tasks that it is currently performing.
Further, the secondary server obtains the task execution state of each subordinate tertiary server in real time, and when the task execution state changes, the secondary server updates the server characteristic value according to the task execution state of the subordinate tertiary server.
Further, the updating of the server feature value by the secondary server specifically includes: setting the initial task characteristic value of the secondary server as IV, calculating the updated server characteristic value V of the secondary server according to the following method:
if all the tertiary servers under the secondary server do not execute any task, V is IV;
otherwise, the secondary server counts the central values of the task characteristic values of all currently executed tasks of all the secondary servers under the secondary server, and if the central value is CV, V is a multiplied by CV + (1-a) multiplied by IV; wherein a is a predefined weight value, and a is more than 1 and less than 1.
The invention has the beneficial effects that: in the multitask processing system, the load of the server is intelligently distributed, and the overall operation efficiency of the system is improved.
[ description of the drawings ]
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, and are not to be considered limiting of the invention, in which:
fig. 1 is a schematic diagram of the structure of the smart load handling system of the present invention.
[ detailed description ] embodiments
The present invention will now be described in detail with reference to the drawings and specific embodiments, wherein the exemplary embodiments and descriptions are provided only for the purpose of illustrating the present invention and are not to be construed as limiting the present invention.
Referring to fig. 1, there is shown a basic structure diagram of the intelligent load processing system of the present invention, wherein the basic structure of the system is a three-layer multi-way tree structure, and each node in the tree structure represents a server. The root node of the tree is a first-level server, each node of the next layer of the root node is a second-level server, and the leaf nodes are third-level servers. The primary server and each secondary server can be connected and communicated with each other through an internal network. As shown in fig. 1, each of the secondary servers is connected to a plurality of tertiary servers, that is, each of the tertiary servers is subordinate to one of the secondary servers, and the secondary servers and the tertiary servers subordinate to the secondary server are also connected via an internal network and communicate with each other.
The primary server is a central server of the intelligent load processing system, is an external interface server of the system, is used for being connected with an external network, receiving a task request sent by the outside, and distributing a secondary server to process the task request according to the task request and a preset rule. For example, according to an embodiment of the present invention, the intelligent load processing system is configured to operate a Web system, when a client browser needs to access a Web page, the client browser sends an access request to the system through the internet, and the primary server receives the Web access request through the internet and allocates a corresponding secondary server to perform processing according to the access request.
The second-level server is the next-level server of the first-level server, receives the task request sent by the first-level server, and distributes a subordinate third-level server to process the task request according to the task request and a preset rule. The third-level servers are specific servers for executing the task requests, and in order to facilitate calculation and processing of task loads, in the system, all the third-level servers are the same servers; the method receives the task request from the secondary server, executes a corresponding task according to the task request, and returns an execution result of the task. Still taking the Web system as an example, the third-level server may read the database according to the access request, obtain corresponding page data and a program, execute the program to generate a Web page, and return the generated Web page to the client.
It should be noted that the number of servers at each level shown in fig. 1 is exemplary, and different secondary servers may also manage different numbers of tertiary servers. In practical applications, a person skilled in the art may set the specific number of servers according to needs, and the present invention is not limited to this.
Based on the system architecture of the intelligent load processing system, the following describes the specific process of intelligent load processing in the system in detail.
(1) And calculating corresponding initial task characteristic values for each secondary server.
The task characteristic value is a characteristic vector calculated for one task and can represent relevant characteristics of the task. For example, for a Web system where each task is a request for access to a Web page, the task feature value may include a literal feature of the address of the access page, as well as a numeric feature of the associated access parameter. For another example, for an image processing system where a task is processing for an image, the task feature value may be an image feature calculated from the corresponding image. In short, a relevant feature representing the task may be extracted from the task, and a feature vector of the task may be calculated from the relevant feature as a feature value of the task. The specific task characteristic value calculation method is based on different task types, but the specific calculation method is not too complex for the efficiency of task allocation. Those skilled in the art can adopt a corresponding feature value calculation method in the prior art according to a specific task type, which is not the focus of the present invention, and the present invention is not described herein again.
In the initial operation of the system, an initial task characteristic value needs to be calculated for each secondary server so as to be used for subsequent task allocation of the primary server. The method specifically comprises the following steps:
and (1.1) collecting a certain number of tasks and generating an initial task set.
According to one embodiment of the invention, a certain number of historical tasks may be collected from the operation of other systems to form an initial task set. According to another embodiment of the present invention, a certain number of tasks may be randomly generated to form an initial set of tasks. Whatever means is used to generate the initial set of tasks is preferably a collection of multiple tasks with broad representativeness.
(1.2) calculating the task characteristic value of each task in the initial task set, obtaining a characteristic value set, clustering the characteristic value set, and dividing the characteristic value set into K characteristic value categories, wherein K is the number of the secondary servers.
Specifically, assuming that n tasks exist in the initial task set, and each task obtains one task feature value by calculation, the obtained n task feature values form a feature value set. The set of eigenvalues is then clustered, for example, the set may be clustered using a K-Means clustering algorithm to obtain K eigenvalue classes, where the number of classes is the same as the number of secondary servers, so that each secondary server corresponds to one of the classes.
Preferably, the obtained number of feature values in the K categories should be as average as possible, and to satisfy this, the value of K may be appropriately adjusted (and the number of secondary servers may be adjusted at the same time), and after clustering, categories with too large number of feature values may be split, or categories with too small number of feature values may be merged. In short, through clustering, K categories with more average feature value number should be obtained finally, which is beneficial to evenly distributing tasks.
And (1.3) for each characteristic value category, calculating the clustering center of the category to obtain an initial task characteristic value.
Specifically, if the number of feature values of K categories is relatively average, each feature value category may have about n/K feature values, and the cluster center of these feature values is calculated to obtain an initial task feature value. And respectively calculating clustering centers for the K categories to obtain K initial task characteristic values.
Any conventional clustering center algorithm may be used for calculating the clustering center, for example, an average value of all feature values in a class is used as the clustering center of the class.
(2) And respectively distributing the K initial task characteristic values to K secondary servers, and setting the distributed initial task characteristic values as server characteristic values by the secondary servers.
Through the calculation, K initial task characteristic values of the system can be obtained, and the K initial task characteristic values are respectively distributed to K secondary servers, namely the K secondary servers correspond to the K initial task characteristic values one by one. For example, the ith secondary server assigns an initial task feature value of ViThen server characteristic value SV of the secondary serveriIs initially Vi(1≤i≤K)。
After determining the server characteristic values of the secondary servers, the primary servers can perform intelligent load processing accordingly.
(3) The primary server receives the task request and calculates a corresponding task characteristic value FV according to the task request.
In particular, the primary server may receive a task request from an external client indicating some task that needs to be performed by the system. The task request comprises the relevant information of the task, and the primary server calculates a corresponding task characteristic value FV according to the relevant information of the task.
(4) The primary server determines a current set of available secondary servers.
According to one embodiment of the invention, the primary server may determine available secondary servers based on the current average task volume of each secondary server. The current average task amount refers to an average value of the current executed task amounts of all the tertiary servers under the secondary server.
For example, suppose that one secondary server belongs to three tertiary servers, and the number of tasks currently executed by the three tertiary servers is 3, 6 and 12 respectively; the current average task size of the secondary server is (3+6+ 12)/3-7.
And the primary server determines the secondary server with the current average task amount smaller than a preset threshold value as an available secondary server. And if the secondary servers with the current average task quantity smaller than the preset threshold value do not exist, the primary server determines a plurality of secondary servers with the minimum current average task quantity as available servers. All of the determined available secondary servers constitute the current set of available secondary servers.
The above method for determining available secondary servers is only one embodiment of the present invention, and those skilled in the art may also use other ways to determine the current available set of secondary servers, and the present invention is not limited thereto.
(5) The primary server calculates the task feature values FV and server feature values SV for each available secondary serveriDistance D ofi。
Specifically, assuming that there are M available secondary servers, the primary server calculates the task feature value FV and the server feature values SV of the M available secondary servers, respectivelyi(1 ≦ i ≦ M) to obtain M distance values D1，D2，……，DM. The distance is used to indicate the similarity between two task feature values.
Since the task feature value is usually a feature vector, any distance calculation algorithm in the prior art, such as euclidean distance, cosine distance, etc., may be used, which is not limited by the present invention.
(6) The primary server determines the distance DiSo as to determine that the available secondary server corresponding to the minimum value is the target server.
In particular, assume that the server eigenvalue of the ith available secondary server is SViAt a distance D from FVi(1 ≦ i ≦ M), the primary server determines a minimum value D from the M distance valuesm(M is more than or equal to 1 and less than or equal to M), the minimum value corresponds to the mth available secondary server, namely the mth available secondary serverThe level server is used as a target server for processing the task.
(7) And the primary server distributes the task request to the target server, and the target server distributes the task request to the tertiary server with the minimum load according to the load of the tertiary server under the target server.
Specifically, the primary server sends the task request to the target server to instruct the target server to process the task request. And after receiving the task request, the target server determines the load of each tertiary server subordinate to the target server, thereby determining the tertiary server with the minimum load. According to an embodiment of the present invention, the load of the tertiary server can be directly determined according to the number of tasks currently executed by the tertiary server, that is, the tertiary server with the least number of tasks currently executed is the least loaded tertiary server. And finally, the target server forwards the task request to the tertiary server with the minimum load, so that the task allocation is completed. The tertiary server receiving the task request may execute a corresponding task according to the task request.
(8) And each secondary server updates the corresponding server characteristic value according to the task execution state of the subordinate tertiary server.
Specifically, the task of each tertiary server is assigned by its corresponding secondary server, and after the task is completed, the tertiary server also notifies its corresponding secondary server of the completion of the task. Therefore, each secondary server can grasp the task execution state of each tertiary server subordinate to the secondary server in real time, namely, which tasks are executed by each tertiary server currently.
And when the task execution state changes, the secondary server updates the server characteristic value according to the task execution state of the subordinate tertiary server. Specifically, the secondary server first counts the center values of the task feature values of all currently executed tasks of all the secondary servers under the secondary server. The center value may be the same cluster center as step 1.3, e.g. the average of all task feature values. And setting the central value as CV and the initial task characteristic value of the secondary server as IV, and calculating the updated server characteristic value V of the secondary server according to the following method:
if all the tertiary servers under the secondary server do not execute any task, V is IV;
otherwise, V ═ a × CV + (1-a) × IV; wherein a is a predefined weight value, and a is more than 1 and less than 1.
Therefore, the secondary server can update the server characteristic value in real time after the task execution state changes, namely, the server characteristic value of the secondary server is dynamically changed, so that the primary server performs task load distribution according to the dynamically changed server characteristic value of the secondary server. By the method, similar tasks can be distributed to the same or similar servers as much as possible for processing, and the overall operation efficiency of the system is improved.
The above description is only a preferred embodiment of the present invention, and all equivalent changes or modifications of the structure, characteristics and principles described in the present invention are included in the scope of the present invention.

Claims (10)

1. An intelligent load processing system based on server is characterized in that the system comprises a primary server, K secondary servers and a plurality of tertiary servers, wherein the primary servers, the K secondary servers and the plurality of tertiary servers are connected in series
The primary server is connected with a plurality of secondary servers and is communicated with the plurality of secondary servers, and the primary server is used for receiving task requests and distributing the secondary servers to process the task requests according to the task requests;
each secondary server is subordinate to a plurality of tertiary servers, and each tertiary server is subordinate to only one secondary server; the secondary server receives a task request sent by the primary server and distributes a subordinate tertiary server to process the task request according to the task request;
the system for intelligent load processing comprises the following specific steps:
(1) calculating a corresponding initial task characteristic value for each secondary server;
(2) respectively distributing the K initial task characteristic values to K secondary servers, and setting the distributed initial task characteristic values as server characteristic values by the secondary servers;
(3) the primary server receives the task request and calculates a corresponding task characteristic value FV according to the task request;
(4) the primary server determines a current available secondary server set;
(5) the primary server calculates the task feature values FV and server feature values SV for each available secondary serveriDistance D ofiI is more than or equal to 1 and less than or equal to M, and M is the number of available secondary servers;
(6) the primary server determines the distance DiDetermining the available secondary server corresponding to the minimum value as the target server;
(7) the primary server distributes the task request to the target server, and the target server distributes the task request to the tertiary server with the minimum load according to the load of the tertiary server under the target server;
(8) and each secondary server updates the corresponding server characteristic value according to the task execution state of the subordinate tertiary server.
2. The smart load handling system of claim 1 wherein step 1 comprises:
(1.1) collecting a certain number of tasks to generate an initial task set;
(1.2) calculating a task characteristic value of each task in an initial task set to obtain a characteristic value set, clustering the characteristic value set, and dividing the characteristic value set into K characteristic value categories;
and (1.3) for each characteristic value category, calculating the clustering center of the category to obtain an initial task characteristic value.
3. The intelligent load handling system of claim 2, wherein a number of historical tasks are collected from the operation of other systems to form an initial set of tasks.
4. The intelligent load handling system of claim 2, wherein a number of tasks are randomly generated to form an initial set of tasks.
5. The smart load processing system according to any one of claims 2 to 4, wherein the cluster center of the class of eigenvalues is the average of all eigenvalues in that class.
6. The intelligent load processing system according to any of claims 1-5, wherein in step 4, the primary server determines available secondary servers according to the current average task amount of each secondary server, wherein the current average task amount refers to the average of the current task execution amount of all tertiary servers under the secondary server.
7. The intelligent load handling system of claim 6, wherein the primary server determines, as available secondary servers, secondary servers for which the current average task size is less than a predetermined threshold; and if the secondary servers with the current average task quantity smaller than the preset threshold value do not exist, the primary server determines a plurality of secondary servers with the minimum current average task quantity as available servers.
8. The intelligent load handling system of claim 1, wherein the load of the tertiary server is the number of tasks it is currently performing.
9. The intelligent load processing system according to claim 1, wherein the secondary server obtains the task execution status of each subordinate tertiary server in real time, and when the task execution status changes, the secondary server updates the server characteristic value according to the task execution status of the subordinate tertiary server.
10. The intelligent load processing system according to claim 9, wherein the updating of the server eigenvalues by the secondary servers specifically comprises: setting the initial task characteristic value of the secondary server as IV, calculating the server characteristic value V of the secondary server according to the following method:
if all the tertiary servers under the secondary server do not execute any task, V is IV;
otherwise, the secondary server counts the central values of the task characteristic values of all currently executed tasks of all the secondary servers under the secondary server, and if the central value is CV, V is a multiplied by CV + (1-a) multiplied by IV; wherein a is a predefined weight value, and a is more than 1 and less than 1.
CN202210104445.9A 2022-01-28 2022-01-28 Intelligent load processing system based on server Pending CN114510350A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210104445.9A CN114510350A (en) 2022-01-28 2022-01-28 Intelligent load processing system based on server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210104445.9A CN114510350A (en) 2022-01-28 2022-01-28 Intelligent load processing system based on server

Publications (1)

Publication Number Publication Date
CN114510350A true CN114510350A (en) 2022-05-17

Family

ID=81550680

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210104445.9A Pending CN114510350A (en) 2022-01-28 2022-01-28 Intelligent load processing system based on server

Country Status (1)

Country Link
CN (1) CN114510350A (en)

Similar Documents

Publication Publication Date Title
CN111722806B (en) Cloud disk allocation method and device, electronic equipment and storage medium
CN1956457B (en) Method and apparatus for arranging mesh work in mesh computing system
US11423085B2 (en) Graph database super vertex partitioning
CN106446959B (en) A kind of cloud computing resources dynamic matching method and device
US10223437B2 (en) Adaptive data repartitioning and adaptive data replication
CN110597719B (en) Image clustering method, device and medium for adaptation test
CN118069380B (en) Computing power resource processing method
CN116501711A (en) Computing power network task scheduling method based on 'memory computing separation' architecture
US10698955B1 (en) Weighted abstract path graph database partitioning
CN116339932A (en) Resource scheduling method, device and server
CN117931454A (en) Computing power resource scheduling method, computing power resource scheduling device, computing power resource scheduling equipment, storage medium and program product
CN111415261B (en) Control method, system and device for dynamically updating flow control threshold of bank system
CN112328332B (en) Database configuration optimization method for cloud computing environment
CN116866440B (en) Cluster node selection scheduling method and device, electronic equipment and storage medium
US7647592B2 (en) Methods and systems for assigning objects to processing units
CN113377866A (en) Load balancing method and device for virtualized database proxy service
CN114510350A (en) Intelligent load processing system based on server
CN115412401B (en) Method and device for training virtual network embedding model and virtual network embedding
US20230137673A1 (en) Systems and methods for dynamically scaling remote resources
CN116187675A (en) Task allocation method, device, equipment and storage medium
CN115168509A (en) Processing method and device of wind control data, storage medium and computer equipment
CN111901410A (en) Server grouping method, device, equipment and computer readable storage medium
KR100795593B1 (en) Grid computing system using dynamic grading and grouping, and data processing method thereof
CN116192858B (en) Load balancing method and device based on weighted traffic
CN116405500B (en) System resource management method based on data analysis and cloud computing data analysis

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