CN112685151A - Resource scheduling system based on optimized weight model - Google Patents
Resource scheduling system based on optimized weight model Download PDFInfo
- Publication number
- CN112685151A CN112685151A CN202011551287.9A CN202011551287A CN112685151A CN 112685151 A CN112685151 A CN 112685151A CN 202011551287 A CN202011551287 A CN 202011551287A CN 112685151 A CN112685151 A CN 112685151A
- Authority
- CN
- China
- Prior art keywords
- configuration
- weight
- resource
- threshold
- calling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Debugging And Monitoring (AREA)
- Stored Programmes (AREA)
Abstract
The invention discloses a resource scheduling system based on an optimized weight model, which comprises: a base configuration module configured to implement configuration management of base data; a weighted algorithm module configured to manage invocation of resources and management of a full lifecycle of state nodes based on a weight configuration; a call trace module configured to record a call log trace for each resource call; a threshold control module configured to implement processing when a resource invocation request exceeds a maximum threshold number of times frequency; an asynchronous delay processing module configured to support asynchronous delay processing for service calls based on the queue middleware; a plug-in support module configured to custom develop plug-in processing when a threshold is exceeded. The invention adds global resource threshold control, a calling threshold formula of the exceeding weight, error processing when the exceeding weight and supporting the flexible realization of the processing when the exceeding weight through the plug-in, and meets the scheduling adaptation under the condition of limited resources.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a resource scheduling system based on an optimized weight model.
Background
Most of the current micro-service invocation is focused on the realization of authority safety verification and load scheduling, and Sentinel in the industry realizes the control of a certain gateway current-limiting rule, but the control is still simpler, and the complex and flexible scheduling configuration cannot be matched. There are problems that: 1) the method can only simply adapt to the calling frequency of each channel, and cannot meet the scheduling adaptation of complex scenes; 2) the flexible configuration of a developer for establishing related weight, an influence factor, a factor formula and a code mode (mainly supporting the code configuration mode of drool) cannot be supported; 3) the existing industry framework does not abstract and define the access limit of limited resources to manage the whole life cycle, so that when a developer needs to realize the scene, the developer can only realize the scene through a large amount of hard codes, and the development and realization efficiency is very low.
Disclosure of Invention
The invention aims to provide a resource scheduling system based on an optimized weight model, which is used for solving the problem that micro-service calling in the prior art cannot meet scheduling adaptation under the condition of limited resources.
The invention solves the problems through the following technical scheme:
a resource scheduling system based on an optimized weight model comprises a basic configuration module, a weighting algorithm module, a calling track module, a threshold control module, an asynchronous delay processing module and a plug-in supporting module, wherein:
the basic configuration module is configured to realize configuration management on basic data, and comprises global resource threshold configuration, weight configuration of each channel calling resource, delay processing formula configuration for calling exceeding the weight, triggering threshold global idle judgment after exceeding the weight, calling frequency and maximum threshold frequency configuration of each channel, configuration management on channel related plug-ins and error processing mode configuration exceeding the weight;
the weighting algorithm module is configured to be configured as a basis according to the weight, when the resource does not exceed the threshold value of the global configuration, the resource is free, a threshold value calling formula exceeding the weight is supported, and the frequency configuration is based on the maximum threshold value times of each channel in the basic configuration module; if the frequency configuration parameter does not exceed the threshold frequency, the resource can be called by exceeding the weight configuration; and performing full lifecycle management for each state node processed in the framework;
the calling track module is configured to record and store a calling log track of each resource calling, and provide a data acquisition inlet for providing historical calling data for subsequent analysis;
a threshold control module configured to select denial of service or delay of service and how long to delay processing according to a configuration when a channel reaches a threshold, to implement processing when a resource invocation request exceeds a maximum threshold frequency;
an asynchronous delay processing module configured to support asynchronous delay processing for service calls based on the queue middleware;
a plug-in support module configured to custom develop plug-in processing when a threshold is exceeded; the plug-in is used for realizing flexible realization and processing based on the current global resource threshold condition and the actual calling frequency condition of the current channel.
The configuration of the delay processing formula for exceeding the weight call includes the number of delays and the duration of each delay. Such as a 1 minute delay for the first time and a 3 minute delay for the second time.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the invention configures the weight as a core, on the basis, global resource threshold control, a calling threshold formula of the exceeding weight, error processing (rejection or delayed calling) when the exceeding weight and processing when the exceeding weight is flexibly realized through a plug-in are added, and the scheduling adaptation under the condition of limited resources is met; the calling tracks of the resources of all channels in the system are recorded in detail, and related data acquisition entries are exposed according to the historical tracks, so that more flexible threshold processing can be realized during plug-in development.
Drawings
FIG. 1 is a block diagram of the system of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples, but the embodiments of the present invention are not limited thereto.
Example (b):
referring to fig. 1, in order to ensure the maximum flexibility, a resource scheduling system based on an optimized weight model is abstracted as the following modules: the device comprises a basic configuration module, a weighting algorithm module, a calling track module, a threshold control module, an asynchronous delay processing module and a plug-in supporting module. The following is a detailed description of the various module layers:
1. a basic configuration module: the method realizes configuration management of basic data, including resource limited request number, weight configuration called by each channel, delayed processing formula configuration called by exceeding weight (such as delaying for 1 minute for the first time and delaying for 3 minutes for the second time), triggering threshold global idle judgment after exceeding weight, maximum threshold frequency configuration of each channel, configuration management of plug-ins related to channels (mainly aiming at a mechanism that a system can carry out more flexible processing on each channel according to plug-in callback configured at the moment when exceeding weight), and configuration of error processing modes (rejecting, delaying processing with the highest frequency, and the like) for exceeding weight.
2. A weighting algorithm module: the module is the core of the framework, based on a weighting model, and in order to ensure the maximum flexibility of the system, the model is only based on the weight, when the resource does not exceed the threshold value of the global configuration, the resource is free, the calling threshold value formula of the exceeding weight can be supported, based on the maximum threshold value frequency configuration of each channel in the basic configuration, if the resource does not exceed the threshold value frequency configuration parameter, the exceeding weight configuration can be carried out to call the resource, in addition, the management of the full life cycle is carried out aiming at each state node processed in the framework, the module mainly provides that each channel is used as the exceeding weight, and under the condition that the whole resource is allowed, besides the threshold value frequency configuration, the module can be flexibly realized based on the plug-in to meet the actual complex processing of each channel.
3. Calling a track module: the method comprises the steps of carrying out detailed call log track recording aiming at each resource call, storing detailed call tracks, aiming at providing basic historical call data for subsequent statistical analysis, threshold control and an ai learning model, providing corresponding api query interfaces, flexibly calling and querying the historical data by each channel when a plug-in is realized, and supporting the ai model to carry out deep analysis on the tracks so as to meet more flexible threshold frequency configuration control, for example, flexibly adjusting weight configuration of difference in each time period according to analysis conditions.
4. A threshold control module: when the channel reaches the threshold, the processing of refusing service or delaying service and delaying for a long time can be carried out according to the configuration, and the module realizes how to carry out the processing of exceeding part of the resource calling request when the calling threshold frequency is exceeded.
5. An asynchronous delay processing module: the module mainly supports asynchronous delay processing aiming at service calling on the basis of queue middleware, and the actual processing of resources is carried out in a queue middleware mode.
6. A plug-in supporting module: the method mainly aims at custom development plug-in processing when the threshold value is exceeded in the system, and plug-in implementation can be flexibly implemented and processed based on the current global resource threshold value condition and the actual calling frequency condition of the current channel.
The calling process is shown in FIG. 1:
1) initiating a resource calling request by a third-party channel;
2) triggering the system framework to call and process, and triggering a weighting algorithm module based on the weight to start processing;
3) inquiring related configurations such as a total resource threshold value, weight information of current channel configuration, delayed processing formula configuration of exceeding weight calling, plug-in of threshold value processing, condition of refusing processing and the like from a basic configuration module;
4) and performing subsequent flow processing according to the acquired basic configuration information in the step 3):
firstly, the actual requested times of the current resource is compared with the global resource limit number, and the following conditions are processed according to the comparison result:
the requested times are more than or equal to the global resource limit number, and the processing is carried out according to the refusing processing configuration of the channel configuration;
if the requested times of the channel are less than the weight proportion ratio, directly carrying out resource calling processing and making calling track recording data according to the weight configured by the channel and the requested times of the channel are less than the weight proportion ratio;
and if the requested times of the channel are greater than or equal to the weight proportion ratio, determining whether to perform normal calling processing or reject processing according to the threshold value configuration of the channel. Specifically, at this time, if the channel is configured with the relevant threshold processing plug-in, the configured plug-in is called to perform the determination processing.
Although the present invention has been described herein with reference to the illustrated embodiments thereof, which are intended to be preferred embodiments of the present invention, it is to be understood that the invention is not limited thereto, and that numerous other modifications and embodiments can be devised by those skilled in the art that will fall within the spirit and scope of the principles of this disclosure.
Claims (2)
1. A resource scheduling system based on an optimized weight model is characterized by comprising a basic configuration module, a weighting algorithm module, a calling track module, a threshold control module, an asynchronous delay processing module and a plug-in supporting module, wherein:
the basic configuration module is configured to realize configuration management on basic data, and comprises global resource threshold configuration, weight configuration of each channel calling resource, delay processing formula configuration for calling exceeding the weight, triggering threshold global idle judgment after exceeding the weight, calling frequency and maximum threshold frequency configuration of each channel, configuration management on channel related plug-ins and error processing mode configuration exceeding the weight;
the weighting algorithm module is configured to be configured as a basis according to the weight, when the resource does not exceed the threshold value of the global configuration, the resource is free, a threshold value calling formula exceeding the weight is supported, and the frequency configuration is based on the maximum threshold value times of each channel in the basic configuration module; if the frequency configuration parameter does not exceed the threshold frequency, the resource can be called by exceeding the weight configuration; and performing full lifecycle management for each state node processed in the framework;
the calling track module is configured to record and store a calling log track of each resource calling, and provide a data acquisition inlet for providing historical calling data for subsequent analysis;
a threshold control module configured to select denial of service or delay of service and how long to delay processing according to a configuration when a channel reaches a threshold, to implement processing when a resource invocation request exceeds a maximum threshold frequency;
an asynchronous delay processing module configured to support asynchronous delay processing for service calls based on the queue middleware;
a plug-in support module configured to custom develop plug-in processing when a threshold is exceeded; the plug-in is used for realizing flexible realization and processing based on the current global resource threshold condition and the actual calling frequency condition of the current channel.
2. The resource scheduling system based on optimized weight model of claim 1, wherein the configuration of the delay processing formula for exceeding the weight call comprises the number of delays and the duration of each delay.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011551287.9A CN112685151B (en) | 2020-12-24 | 2020-12-24 | Resource scheduling system based on optimized weight model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011551287.9A CN112685151B (en) | 2020-12-24 | 2020-12-24 | Resource scheduling system based on optimized weight model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112685151A true CN112685151A (en) | 2021-04-20 |
| CN112685151B CN112685151B (en) | 2022-02-01 |
Family
ID=75452697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011551287.9A Active CN112685151B (en) | 2020-12-24 | 2020-12-24 | Resource scheduling system based on optimized weight model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112685151B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106899518A (en) * | 2017-02-27 | 2017-06-27 | 腾讯科技（深圳）有限公司 | A kind of method for processing resource and device based on Internet data center |
| CN107291547A (en) * | 2016-03-31 | 2017-10-24 | 阿里巴巴集团控股有限公司 | A kind of task scheduling processing method, apparatus and system |
| US20190068269A1 (en) * | 2016-04-29 | 2019-02-28 | Huawei Technologies Co., Ltd. | Multi-user multiple-input multiple-output u-mimo data transmission method and base station |
| CN111475282A (en) * | 2020-03-08 | 2020-07-31 | 苏州浪潮智能科技有限公司 | Distributed storage load balancing method and device based on client |
| CN112101674A (en) * | 2020-09-22 | 2020-12-18 | 广东睿盟计算机科技有限公司 | Resource allocation matching method, device, equipment and medium based on group intelligent algorithm |
-
2020
- 2020-12-24 CN CN202011551287.9A patent/CN112685151B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107291547A (en) * | 2016-03-31 | 2017-10-24 | 阿里巴巴集团控股有限公司 | A kind of task scheduling processing method, apparatus and system |
| US20190068269A1 (en) * | 2016-04-29 | 2019-02-28 | Huawei Technologies Co., Ltd. | Multi-user multiple-input multiple-output u-mimo data transmission method and base station |
| CN106899518A (en) * | 2017-02-27 | 2017-06-27 | 腾讯科技（深圳）有限公司 | A kind of method for processing resource and device based on Internet data center |
| CN111475282A (en) * | 2020-03-08 | 2020-07-31 | 苏州浪潮智能科技有限公司 | Distributed storage load balancing method and device based on client |
| CN112101674A (en) * | 2020-09-22 | 2020-12-18 | 广东睿盟计算机科技有限公司 | Resource allocation matching method, device, equipment and medium based on group intelligent algorithm |
Non-Patent Citations (2)
| Title |
|---|
| JIANXIANG LI等: "Availability Analysis of Web-Server Clusters with QoS-Aware Load Balancing", 《2010 INTERNATIONAL SYMPOSIUM ON COMPUTATIONAL INTELLIGENCE AND DESIGN》 * |
| 何倩等: "基于软件定义网络的反饱和分组云负载均衡", 《计算及应用》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112685151B (en) | 2022-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20070174839A1 (en) | Method and system for managing programs within systems | |
| CN109617829B (en) | Method, device and system for processing service request data | |
| CN108770017B (en) | Dynamic equalization method and system for wireless resources | |
| CN114356557B (en) | Cluster capacity expansion method and device | |
| CN108650334A (en) | A kind of setting method and device of session failed | |
| JP2021523429A (en) | Automatic database query load assessment and adaptive processing | |
| RU2641250C2 (en) | Device and method of queue management | |
| CN116610701A (en) | Time slicing method and device and electronic equipment | |
| CN112685151B (en) | Resource scheduling system based on optimized weight model | |
| CN111078651A (en) | Method and device for counting usage amount of object storage | |
| CN113079062B (en) | Resource adjusting method and device, computer equipment and storage medium | |
| CN111309442B (en) | Method, device, system, medium and equipment for adjusting number of micro-service containers | |
| CN114143263B (en) | Method, equipment and medium for limiting current of user request | |
| CN115174487A (en) | High-concurrency current limiting method and device and computer storage medium | |
| CN115904785A (en) | Abnormity positioning method, device, equipment and readable storage medium | |
| CN114500544A (en) | Method, system, equipment and medium for load balancing among nodes | |
| CN108304252B (en) | Task scheduling method and device | |
| CN113657635A (en) | Method for predicting communication user loss and electronic equipment | |
| CN112583742A (en) | Message processing method and device | |
| CN117591301B (en) | Monitoring cloud resource allocation optimization method and system | |
| CN103379629A (en) | Resource scheduling method and resource scheduling apparatus | |
| CN116149959B (en) | Data processing device, method, monitoring equipment and computer program product | |
| CN103297926A (en) | Cluster paging method in private network and base station | |
| CN111124310B (en) | Storage system scheduling optimization method and related components | |
| CN113741810B (en) | Data migration method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |