CN114398400A - Serverless resource pool system based on active learning - Google Patents
Serverless resource pool system based on active learning Download PDFInfo
- Publication number
- CN114398400A CN114398400A CN202210295299.2A CN202210295299A CN114398400A CN 114398400 A CN114398400 A CN 114398400A CN 202210295299 A CN202210295299 A CN 202210295299A CN 114398400 A CN114398400 A CN 114398400A
- Authority
- CN
- China
- Prior art keywords
- module
- query
- configuration parameter
- service
- parameter values
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
- G06F16/24534—Query rewriting; Transformation
- G06F16/24549—Run-time optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/302—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a software system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/80—Database-specific techniques
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention provides a Serverless resource pool system based on active learning, which comprises a processing module, a system storage module, a transmission module, an input/output module, a network adaptation module, a query engine and a learning optimization module, wherein the processing module is used for executing resources, the system storage module is used for storing the resources, the transmission module is used for transmitting the resources and information in the system, the input/output module is used for communicating with external equipment, the network adaptation module is used for providing a network environment, the query engine is used for querying from the system storage module to obtain proper resources, and the learning optimization module is used for optimizing the query configuration of the query engine; the system can actively learn the query engine of the query resource pool, so that the query effect of the query engine is continuously improved, and the use experience of a user is improved.
Description
Technical Field
The present invention relates generally to cloud service systems, and more particularly to a server less resource pool system based on active learning.
Background
Serverless cloud services refer to any service provided to users on demand from a cloud computing provider's servers over the internet, rather than from the company's own local servers, and are intended to provide simple, scalable access to applications, resources and services, and are managed entirely by the cloud service provider, and cloud storage is a computer data storage model in which digital data is stored in logical pools, physical storage spans multiple servers, and the physical environment is typically owned and managed by a hosting company, these cloud storage providers are responsible for maintaining the availability and accessibility of data, and protecting and operating the physical environment, and as the number of users increases, improvements in the process of users querying service resources are needed.
Now, a plurality of resource pool systems have been developed, and through a lot of search and reference, it is found that the existing authorization systems are, for example, the systems disclosed by the publication numbers KR101640231B1, KR101765725B1, CN112256439B and KR1020160032881A, and the resource admission module is used for acquiring service parameters of each service in the cloud computing resource pool through an API interface; the service parameter convergence module is used for dividing the service parameters into service class parameters and resource class parameters; the parameter assignment module is used for assigning the resource type parameters according to the use specifications of each service in the cloud computing resource pool; and the service directory generation module is used for generating a service directory according to the service class parameters and the resource class parameters. When the system acquires resources, the process cannot be improved, so that the stability problem is easy to occur after the number of users is increased.
Disclosure of Invention
The invention aims to provide a Serverless resource pool system based on active learning, aiming at the existing defects.
The invention adopts the following technical scheme:
a Serverless resource pool system based on active learning comprises a processing module, a system storage module, a transmission module, an input/output module, a network adaptation module, a query engine and a learning optimization module, wherein the system storage module is used for storing service resources, the processing module is used for executing the service resources, the transmission module is used for transmitting the resources and information, the input/output module is used for interacting with external equipment, the network adaptation module is used for providing stable network service, the query engine is used for querying the service resources, and the learning optimization module is used for optimizing the query of the query engine;
the query engine queries to obtain service resources according to query types and configuration parameter values, generates performance data after the service resources are searched, the query types, the configuration parameter values and the performance data are called metadata, the learning optimization module processes to obtain the correlation of the metadata, optimizes the configuration parameter values according to the correlation, and the optimized configuration parameter values are used for formal service query;
the learning optimization module obtains through calculation
Expressing the correlation between the jth configuration parameter value and ith individual performance data in the B-type query service and establishing an optimized objective function
：
Wherein the content of the first and second substances,
is the (i) th weight coefficient,
as the number of the i-th standard,
for the ith optimization goal, n is the number of properties
The number of categories according to;
the ith optimization target is related to configuration parameter values and is obtained by calculation according to historical data;
the learning optimization module is based on
Calculating a proportional distribution function of the distribution of values
Satisfying the following equation:
wherein m is the number of types of configuration parameter values;
the learning optimization module is used
Representing distribution functions to class i performance data
And accumulating the distribution functions according to the following formula to obtain a comprehensive distribution function U (j):
the third determining unit adjusts the configuration parameter value according to the distribution of U (j), and selects to make
The smallest one is the most optimal one;
further, the learning optimization module constructs an input matrix X and an output matrix Y according to historical data, processes the matrix X and the matrix Y to obtain a matrix Z, and elements in the matrix Z
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
are the elements of the matrix Y and,
the number of times of query in the historical data is k;
further, correlation
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
an ith performance data value representing metadata in the current query,
a jth configuration parameter value representing metadata in a current query;
further, the configuration parameter values include memory size, buffer size, serialization options, compression parameter values, network parameter values, scheduling specific values, and execution option values;
furthermore, the query engine is provided with a monitoring unit for monitoring the use state of the query engine, when the query engine is not largely used for the service query of the user, the learning optimization module optimizes the configuration parameter value by using the query engine, and when the query engine is largely used for the service query of the user, the learning optimization module is in a dormant state.
The beneficial effects obtained by the invention are as follows:
the system actively learns through the learning optimization module when the query engine is idle, improves the query effect of the query engine, enables a user to quickly acquire more suitable service resources, takes the configuration parameter values as input data and the performance data as output data in the optimization process, establishes correlation relation between the input data and the output data in the historical query process, and continuously optimizes the configuration parameter values according to the correlation, thereby improving the use effect of the query engine.
For a better understanding of the features and technical content of the present invention, reference should be made to the following detailed description of the invention and accompanying drawings, which are provided for purposes of illustration and description only and are not intended to limit the invention.
Drawings
FIG. 1 is a schematic view of the overall structural framework of the present invention;
FIG. 2 is a schematic diagram of a process for optimizing by the learning optimization module according to the present invention;
FIG. 3 is a block diagram of a service query optimization framework according to the present invention;
FIG. 4 is a schematic diagram of the memory module of the system of the present invention;
FIG. 5 is a schematic diagram of the learning optimization module according to the present invention.
Detailed Description
The following is a description of embodiments of the present invention with reference to specific embodiments, and those skilled in the art will understand the advantages and effects of the present invention from the disclosure of the present specification. The invention is capable of other and different embodiments and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present invention. The drawings of the present invention are for illustrative purposes only and are not intended to be drawn to scale. The following embodiments will further explain the related art of the present invention in detail, but the disclosure is not intended to limit the scope of the present invention.
The first embodiment.
The embodiment provides a server less resource pool system based on active learning, which, with reference to fig. 1, includes a processing module, a system storage module, a transmission module, an input/output module, a network adaptation module, a query engine, and a learning optimization module, where the system storage module is used to store service resources, the processing module is used to execute the service resources, the transmission module is used to transmit resources and information, the input/output module is used to interact with external devices, the network adaptation module is used to provide stable network services, the query engine is used to query the service resources, and the learning optimization module is used to optimize the query of the query engine;
the query engine queries to obtain service resources according to query types and configuration parameter values, generates performance data after the service resources are searched, the query types, the configuration parameter values and the performance data are called metadata, the learning optimization module processes to obtain the correlation of the metadata, optimizes the configuration parameter values according to the correlation, and the optimized configuration parameter values are used for formal service query;
the learning optimization module obtains through calculation
Expressing the correlation between the jth configuration parameter value and ith individual performance data in the B-type query service and establishing an optimized objective function
：
Wherein the content of the first and second substances,
is the (i) th weight coefficient,
as the number of the i-th standard,
for the ith optimization goal, n isNumber of properties
The number of categories according to;
the ith optimization target is related to configuration parameter values and is obtained by calculation according to historical data;
the learning optimization module is based on
Calculating a proportional distribution function of the distribution of values
Satisfying the following equation:
wherein m is the number of types of configuration parameter values;
the learning optimization module is used
Representing distribution functions to class i performance data
And accumulating the distribution functions according to the following formula to obtain a comprehensive distribution function U (j):
the third determining unit adjusts the configuration parameter value according to the distribution of U (j), and selects to make
The smallest one is the most optimal one;
the learning optimization module constructs an input matrix X and an output matrix Y according to historical data, processes the matrix X and the matrix Y to obtain a matrix Z, and elements in the matrix Z
Is calculated byThe formula is as follows:
wherein the content of the first and second substances,
are the elements of the matrix Y and,
the number of times of query in the historical data is k;
correlation
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
an ith performance data value representing metadata in the current query,
a jth configuration parameter value representing metadata in a current query;
the configuration parameter values include memory size, buffer size, serialization options, compression parameter values, network parameter values, scheduling specific values, and execution option values;
the query engine is provided with a monitoring unit for monitoring the use state of the query engine, when the query engine is not used for the service query of a user in a large quantity, the learning optimization module optimizes the configuration parameter value by using the query engine, and when the query engine is used for the service query of the user in a large quantity, the learning optimization module is in a dormant state.
Example two.
The embodiment includes the whole content of the first embodiment, and provides a Serverless resource pool system based on active learning, which includes a processing module, a system storage module, a transmission module, an input/output module, a network adaptation module, a query engine and a learning optimization module, wherein the processing module is used for executing resources, the system storage module is used for storing resources, the transmission module is used for transmitting resources and information in a system, the input/output module is used for communicating with external equipment, the network adaptation module is used for providing a network environment, the query engine is used for querying from the system storage module to obtain appropriate resources, and the learning optimization module is used for optimizing query configuration of the query engine;
with reference to fig. 4, the system memory module includes a random access memory, a cache memory, a driver and a program unit, the random access memory and the cache memory are used for storing data in the system, the driver is used for driving data to be read from or written into the random access memory and the cache memory, the program unit includes a plurality of code packages configured to execute specific functions, and belongs to resources for user query and call;
with reference to fig. 5, the learning optimization module includes a first determination unit, a second determination unit, a third determination unit, a transmitter, a recorder and an application unit, wherein the first determination unit is configured to determine a service category to be queried, the transmitter is configured to transmit a query instruction to a query engine, the recorder is configured to record metadata of a service query executed on the query engine, the metadata includes performance data, the query category and a configuration parameter value, the second determination unit is configured to determine a correlation between the metadata, the third determination unit is configured to determine an optimal configuration parameter value according to the correlation, and the application unit is configured to apply the optimal configuration parameter value to the query engine;
the units of the learning optimization module carry out data communication through the transmission module;
with reference to fig. 2, the optimization process of the learning optimization module for service query includes the following steps:
s1, the first determining unit determines the category of the service needing to be inquired;
s2, the transmitter transmits the service query of the service category to the query engine;
s3, the query engine queries according to the received metadata of the service query;
s4, the second determining unit calculates and determines the correlation of the metadata;
s5, the third determining unit optimizes and adjusts the configuration parameter value according to the correlation;
s6, the application unit applies the optimized and adjusted configuration parameter values to the same type of extended service query set on the query engine to verify the configuration;
s7, when a positive result of the verification is obtained, executing the service inquiry of the same category by using the optimized configuration parameter value;
with reference to fig. 3, a group of service queries of the same category is sent to a query engine, the service queries are related to configuration parameter values, metadata and configuration parameter values for query execution are recorded together in a storage area and are associated, an activation unit activates the metadata to determine a query category, a selected configuration parameter value and result performance data, and obtains an optimized set of the configuration parameter values based on correlation analysis, and then associates the optimized set with another group of service queries of the same category and sends the optimized set to the query engine for circular optimization, and when the service queries of the same category are all optimized, the finally obtained optimized set is output as a formal configuration;
the query engine is provided with a monitoring unit for monitoring the use state of the query engine, when the query engine is not used for the service query of a user in a large quantity, the learning optimization module optimizes the configuration parameter value by using the query engine, and when the query engine is used for the service query of the user in a large quantity, the learning optimization module is in a dormant state;
in step S4, the process of performing correlation calculation on the metadata is as follows:
using the performance data of the metadata in the query
It is shown that,
n is the number of categories of performance, query class is represented by B, and configuration parameter values are represented by
It is shown that,
m is the number of types of configuration parameter values;
the second determination unit calculates the data according to the history data and the current inquiry data
The correlation degree of the jth configuration parameter value and the ith performance in the class B service query category is expressed;
the second determination unit constructs an input matrix X according to the historical data:
wherein the content of the first and second substances,
representing the data value of the ith configuration parameter value in the jth historical data of the class B query service, and k is the number of historical queries;
the second determining unit constructs an output matrix Y according to the historical data:
wherein the content of the first and second substances,
a data value representing ith performance in jth historical data of the B-type query service;
the second determining unit processes the input matrix and the output matrix to obtain a correlation matrix Z:
wherein the elements
Comprises the following steps:
the second determination unit calculates the following equation
：
In step S5, the third determination unit establishes an optimization objective function
：
Wherein the content of the first and second substances,
is the (i) th weight coefficient,
as the number of the i-th standard,
optimizing the objective for the ith;
the third determining unit configures parameter values according to the correlation data calculated by the second determining unit
Adjusting to minimize an expected optimization objective function value;
the third determination unit is based on
Calculating a proportional distribution function of the distribution of values
Satisfying the following equation:
a set of distribution functions can be derived for a class of performance data
And n sets of distribution functions can be obtained for n types of performance data
By using
Representing distribution functions to class i performance data
The third determining unit adds the distribution functions according to the following formula to obtain a comprehensive distribution function u (j):
the third determining unit configures parameter values according to the distribution pairs of U (j)
Making adjustment and selecting
The smallest one is the most optimal one.
The disclosure is only a preferred embodiment of the invention, and is not intended to limit the scope of the invention, so that all equivalent technical changes made by using the contents of the specification and the drawings are included in the scope of the invention, and further, the elements thereof can be updated as the technology develops.
Claims (5)
1. A Serverless resource pool system based on active learning is characterized by comprising a processing module, a system storage module, a transmission module, an input/output module, a network adaptation module, a query engine and a learning optimization module, wherein the system storage module is used for storing service resources, the processing module is used for executing the service resources, the transmission module is used for transmitting the service resources and information, the input/output module is used for interacting with external equipment, the network adaptation module is used for providing stable network service, the query engine is used for querying the service resources, and the learning optimization module is used for optimizing the query of the query engine;
the query engine queries to obtain service resources according to query types and configuration parameter values, generates performance data after the service resources are queried, wherein the query types, the configuration parameter values and the performance data are called metadata, the learning optimization module processes to obtain the correlation of the metadata, optimizes the configuration parameter values according to the correlation, and the optimized configuration parameter values are used for service query of users;
the learning optimization module obtains through calculation
Expressing the correlation between the jth configuration parameter value and ith individual performance data in the B-type query service and establishing an optimized objective function
：
Wherein the content of the first and second substances,
is the (i) th weight coefficient,
as the number of the i-th standard,
for the ith optimization goal, n is the number of properties
The number of categories according to;
the ith optimization target is related to configuration parameter values and is obtained by calculation according to historical data;
the learning optimization module is based on
Calculating a proportional distribution function of the distribution of values
Satisfying the following equation:
wherein m is the number of types of configuration parameter values;
the learning optimization module is used
Representing distribution functions to class i performance data
And accumulating the distribution functions according to the following formula to obtain a comprehensive distribution function U (j):
the learning optimization module adjusts the configuration parameter values according to the distribution of U (j), and selects to ensure that
The smallest one is the optimization scheme.
2. The Serverless resource pool system based on active learning of claim 1, wherein the learning optimization module constructs an input matrix X and an output matrix Y according to historical data, processes the matrix X and the matrix Y to obtain a matrix Z, and the elements in the matrix Z
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
are the elements of the matrix Y and,
k is the number of times the service resource is queried in the history data for the element in the matrix X.
3. The Serverless resource pool system based on active learning of claim 2, wherein correlation
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
an ith performance data value representing metadata in the current query,
the jth configuration parameter value representing the metadata in the current query.
4. The Serverless resource pool system based on active learning of claim 3, wherein the configuration parameter values comprise memory size, buffer size, serialization options, compression parameter values, network parameter values, scheduling specific values, and execution option values.
5. The Serverless resource pool system based on active learning of claim 4, wherein the query engine is provided with a monitoring unit for monitoring the usage status of the query engine, the learning optimization module optimizes the configuration parameter values using the query engine when the query engine is not heavily used for the service query of the user, and the learning optimization module is in a dormant state when the query engine is heavily used for the service query of the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210295299.2A CN114398400B (en) | 2022-03-24 | 2022-03-24 | Serverless resource pool system based on active learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210295299.2A CN114398400B (en) | 2022-03-24 | 2022-03-24 | Serverless resource pool system based on active learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114398400A true CN114398400A (en) | 2022-04-26 |
| CN114398400B CN114398400B (en) | 2022-06-03 |
Family
ID=81234445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210295299.2A Active CN114398400B (en) | 2022-03-24 | 2022-03-24 | Serverless resource pool system based on active learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114398400B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120089664A1 (en) * | 2010-10-12 | 2012-04-12 | Sap Portals Israel, Ltd. | Optimizing Distributed Computer Networks |
| CN106209975A (en) * | 2016-06-23 | 2016-12-07 | 中国人民解放军国防科学技术大学 | A kind of resource provision method across data center's cloud computing system |
| US20170286488A1 (en) * | 2016-03-30 | 2017-10-05 | Linkedin Corporation | Techniques for search optimization on mobile devices |
| CN110569060A (en) * | 2019-09-19 | 2019-12-13 | 山东浪潮通软信息科技有限公司 | High concurrency implementation method based on micro-service framework |
| CN111339066A (en) * | 2020-05-20 | 2020-06-26 | 腾讯科技（深圳）有限公司 | Database optimization method and device, electronic equipment and computer-readable storage medium |
| CN111913939A (en) * | 2020-08-12 | 2020-11-10 | 莫毓昌 | Database cluster optimization system and method based on reinforcement learning |
| CN113064879A (en) * | 2021-03-12 | 2021-07-02 | 腾讯科技（深圳）有限公司 | Database parameter adjusting method and device and computer readable storage medium |
| US20220043822A1 (en) * | 2020-08-04 | 2022-02-10 | International Business Machines Corporation | Shadow experiments for serverless multi-tenant cloud services |
-
2022
- 2022-03-24 CN CN202210295299.2A patent/CN114398400B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120089664A1 (en) * | 2010-10-12 | 2012-04-12 | Sap Portals Israel, Ltd. | Optimizing Distributed Computer Networks |
| US20170286488A1 (en) * | 2016-03-30 | 2017-10-05 | Linkedin Corporation | Techniques for search optimization on mobile devices |
| CN106209975A (en) * | 2016-06-23 | 2016-12-07 | 中国人民解放军国防科学技术大学 | A kind of resource provision method across data center's cloud computing system |
| CN110569060A (en) * | 2019-09-19 | 2019-12-13 | 山东浪潮通软信息科技有限公司 | High concurrency implementation method based on micro-service framework |
| CN111339066A (en) * | 2020-05-20 | 2020-06-26 | 腾讯科技（深圳）有限公司 | Database optimization method and device, electronic equipment and computer-readable storage medium |
| US20220043822A1 (en) * | 2020-08-04 | 2022-02-10 | International Business Machines Corporation | Shadow experiments for serverless multi-tenant cloud services |
| CN111913939A (en) * | 2020-08-12 | 2020-11-10 | 莫毓昌 | Database cluster optimization system and method based on reinforcement learning |
| CN113064879A (en) * | 2021-03-12 | 2021-07-02 | 腾讯科技（深圳）有限公司 | Database parameter adjusting method and device and computer readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 史英杰 等: "《云数据管理系统中查询技术研究综述》", 《计算机学报》, vol. 36, no. 2, 28 February 2013 (2013-02-28), pages 209 - 225 * |
| 孟令玺 等: "《云计算下的资源负载均衡性调度仿真》", 《计算机仿真》, vol. 35, no. 4, 30 April 2018 (2018-04-30), pages 386 - 389 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114398400B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111027736B (en) | Micro-service combined deployment and scheduling method under multi-objective optimization | |
| CN109120715A (en) | Dynamic load balancing method under a kind of cloud environment | |
| CN112511342B (en) | Network slicing method and device, electronic equipment and storage medium | |
| US20060036743A1 (en) | System for balance distribution of requests across multiple servers using dynamic metrics | |
| CN112822526A (en) | Video recommendation method, server and readable storage medium | |
| GB2594367A (en) | Low entropy browsing history for content quasi-personalization | |
| CN108885641A (en) | High Performance Data Query processing and data analysis | |
| CN104063501B (en) | copy balance method based on HDFS | |
| CN112822051B (en) | Service acceleration method based on service perception | |
| CN113094181A (en) | Multi-task federal learning method and device facing edge equipment | |
| CN109948016A (en) | Application message method for pushing, device, server and computer readable storage medium | |
| CN111881358A (en) | Object recommendation system, method and device, electronic equipment and storage medium | |
| CN113778683A (en) | Handle identification system analysis load balancing method based on neural network | |
| CN108829846B (en) | Service recommendation platform data clustering optimization system and method based on user characteristics | |
| CN114398400B (en) | Serverless resource pool system based on active learning | |
| CN111629216B (en) | VOD service cache replacement method based on random forest algorithm under edge network environment | |
| KR102605598B1 (en) | Content provider recommendations to improve targetting and other settings | |
| CN115016889A (en) | Virtual machine optimization scheduling method for cloud computing | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| Shalini Lakshmi et al. | A predictive context aware collaborative offloading framework for compute-intensive applications | |
| CN113298115A (en) | User grouping method, device, equipment and storage medium based on clustering | |
| Saravanan et al. | Cloud resource optimization based on Poisson linear deep gradient learning for mobile cloud computing | |
| Dong et al. | Accelerating skycube computation with partial and parallel processing for service selection | |
| CN111258743A (en) | Cloud task scheduling method, device, equipment and storage medium based on discrete coding | |
| Li et al. | Information gain based dynamic support set construction for cold-start recommendation |
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 |