CN112398663A - Elastic IP charging method and system based on deep neural network - Google Patents
Elastic IP charging method and system based on deep neural network Download PDFInfo
- Publication number
- CN112398663A CN112398663A CN202011227646.5A CN202011227646A CN112398663A CN 112398663 A CN112398663 A CN 112398663A CN 202011227646 A CN202011227646 A CN 202011227646A CN 112398663 A CN112398663 A CN 112398663A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- network model
- user
- elastic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 15
- 238000000034 method Methods 0.000 title claims abstract description 15
- 230000015654 memory Effects 0.000 claims abstract description 39
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 35
- 238000007781 pre-processing Methods 0.000 claims abstract description 23
- 238000000605 extraction Methods 0.000 claims description 19
- 230000004913 activation Effects 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 2
- 230000007787 long-term memory Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/02—Details
- H04L12/14—Charging, metering or billing arrangements for data wireline or wireless communications
- H04L12/1432—Metric aspects
- H04L12/1439—Metric aspects time-based
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an elastic IP charging method and system based on a deep neural network, belonging to the technical field of cloud computing; the method comprises the following specific steps: s1, preprocessing the data of the user during the internet surfing period; s2, utilizing a residual error network to automatically extract the characteristics of the preprocessed data; s3, utilizing the relevance of time before and after the characteristic data is trained by learning of a bidirectional long-time memory network model; s4, re-pricing the elastic IP unit price by using the bidirectional long-short time memory network model; the invention provides a dynamic adjustment elastic IP charging mode for a cloud computing server, which is characterized in that based on previous internet surfing period data of a user, a residual error network model is adopted to extract characteristics of the internet surfing period data of the user, the extracted characteristics are sent to a bidirectional long-time and short-time memory network model, and the internet surfing time possibly occurring by the user is predicted, so that the charging mode of the elastic IP is dynamically adjusted.
Description
Technical Field
The invention discloses an elastic IP charging method and system based on a deep neural network, and relates to the technical field of cloud computing.
Background
At present, cloud computing technology based on openstack is more mature, and scheduling and management of computing, storage and network virtualization resources are more convenient. The cloud computing service provider deploys the cloud platform of the cloud computing service provider to achieve resource virtualization, and therefore hardware resources are used more efficiently. The elastic IP belongs to a virtual resource for providing an out-of-network service for a user in a network service. At present, the charging modes of the flexible IP are mainly divided into two modes, namely a charging mode for covering a month in a year and a charging mode for charging according to needs. Here, the rationality of the pay-as-you-go approach is analyzed with emphasis here. The on-demand charging is charging the user based on the total amount of traffic used. However, this charging method is relatively fixed for the user, and the user cannot be charged flexibly according to the internet access habit of the user.
The cloud computing service provider can provide various required virtual resources for the user, after the user purchases the required service, the user needs to use the flexible IP to access the network, but the internet surfing time of the user may change along with different time periods, and in a charging-on-demand mode, a fixed charging mode cannot maximize the use rights and experience of the user. In order to maximize the use rights and interests of users and provide a more reasonable charging mode for cloud service providers, the invention provides the elastic IP charging method and the elastic IP charging system based on the deep neural network so as to solve the problems.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an elastic IP charging method and system based on a deep neural network, and the adopted technical scheme is as follows: an elastic IP charging method based on a deep neural network comprises the following specific steps:
s1, preprocessing the data of the user during the internet surfing period;
s2, utilizing a residual error network to automatically extract the characteristics of the preprocessed data;
s3, utilizing the relevance of time before and after the characteristic data is trained by learning of a bidirectional long-time memory network model;
s4 re-pricing the flexible IP unit price by using the bidirectional long-time memory network model.
The specific steps of the S1 for preprocessing the user internet surfing period data are as follows:
s101, counting user internet traffic data at intervals and drawing a line graph;
s102, converting the line graph into a power graph through Fourier transformation.
And S2, adopting a residual error network with 50 layers to automatically extract the characteristics of the preprocessed data.
The specific steps of the S4 for re-pricing the resilient IP unit price by using the bidirectional long-and-short term memory network model are as follows:
s401, removing an activation function in the bidirectional long-short time memory network model to obtain internet surfing time prediction data;
s402, the bidirectional long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data.
An elastic IP billing system based on a deep neural network specifically comprises a data preprocessing module, a feature extraction module, a weight training module and a model prediction module:
a data preprocessing module: preprocessing the data of the user online time period;
a feature extraction module: utilizing a residual error network to perform automatic feature extraction on the preprocessed data;
the weight training module: the relevance of time before and after the characteristic data is trained by learning by utilizing a bidirectional long-time memory network model;
a model prediction module: and re-pricing the elastic IP unit price by utilizing a bidirectional long-time memory network model.
The data preprocessing module specifically comprises a data statistics module and a statistics conversion module:
a data statistics module: counting user internet traffic data at intervals and drawing a line graph;
a statistic conversion module: and converting the line graph into a power graph through Fourier transformation.
And the feature extraction module adopts a residual error network with 50 layers to automatically extract features of the preprocessed data.
The model prediction module specifically comprises a data prediction module and a data prediction module:
a data prediction module: removing an activation function in the bidirectional long-short time memory network model to obtain internet surfing time prediction data;
a data prediction module: and the bidirectional long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data.
The invention has the beneficial effects that: the invention provides a dynamic adjustment elastic IP charging mode for a cloud computing server, which is characterized in that based on previous internet surfing period data of a user, a residual error network model is adopted to extract characteristics of the internet surfing period data of the user, the extracted characteristics are sent to a bidirectional long-time and short-time memory network model, and the internet surfing time possibly occurring by the user is predicted, so that the charging mode of the elastic IP is dynamically adjusted.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of the method of the present invention; FIG. 2 is a schematic diagram of the system of the present invention; fig. 3 is a schematic diagram of an embodiment of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
The first embodiment is as follows:
an elastic IP charging method based on a deep neural network comprises the following specific steps:
s1, preprocessing the data of the user during the internet surfing period;
s2, utilizing a residual error network to automatically extract the characteristics of the preprocessed data;
s3, utilizing the relevance of time before and after the characteristic data is trained by learning of a bidirectional long-time memory network model;
s4, re-pricing the elastic IP unit price by using the bidirectional long-short time memory network model;
firstly, preprocessing data of a user internet surfing time period according to S1, wherein a deep neural network is composed of a residual error network and a bidirectional long-short time memory network model, automatically extracting characteristics of the preprocessed data by using the residual error network according to S2, then, using the bidirectional long-short time memory network model to train the relevance of characteristic data before and after learning according to S3, and finally, re-pricing the unit price of the elastic IP by using the bidirectional long-short time memory network model according to S4, wherein the obtained unit price of the elastic IP is the unit price fusing the internet surfing habits of the user;
the invention carries out certain pretreatment on the past internet surfing time period data of the user and predicts the internet surfing time period which is possibly generated by the user based on the residual error network and the bidirectional long-time and short-time memory network model, thereby dynamically adjusting the charging unit price of the flexible IP. Compared with the original charging mode of fixed unit price for charging according to needs, the model for dynamically adjusting the charging mode of the flexible IP can provide more reasonable charging unit price for a user and can also provide a scheme for relieving bandwidth pressure for a cloud service provider;
further, the step of S1 preprocessing the internet surfing period data of the user includes:
s101, counting user internet traffic data at intervals and drawing a line graph;
s102, converting the line graph into a power graph through Fourier transform;
collecting the internet surfing data of the user, firstly counting the flow data used by the user every thirty minutes according to S101, drawing the flow data of the user into a line graph with the horizontal axis as time and the vertical axis as flow usage, and observing the internet surfing habit of the user; then converting the visual line graph into a power graph through fast Fourier transform according to S102, and conveniently extracting the user characteristics by adopting a residual error network model;
further, the S2 adopts a residual error network with 50 layers to perform automatic feature extraction on the preprocessed data;
the residual error network model can extract deeper user characteristics in the aspect of characteristic extraction compared with other neural network models, so that the internet surfing time of a user can be predicted more conveniently;
and for the power diagram obtained in the step S102, a residual error network model with 50 layers is adopted to automatically extract the characteristics of the preprocessed data according to the step S2, and users have different internet surfing durations at different time and are reflected in the residual error network model, namely different weights of all layers of the model. The user data is not simple discrete data, but data with time relevance before and after, so that the features extracted by the model are also related before and after;
further, the specific step of S4 re-pricing the unit price of the flexible IP by using the long and short term memory network model is as follows:
s401, removing an activation function in the long and short time memory network model to obtain internet surfing time prediction data;
s402, the bidirectional long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data.
Removing an activation function layer in the bidirectional long-and-short-term memory network model according to S401 to obtain the prediction of the internet surfing time; and re-pricing the unit price of the next flexible IP according to the internet surfing duration which is obtained by the S402 according to the bidirectional long-and-short-term memory network model and is possible to occur to the user, wherein the unit price of the obtained flexible IP is the unit price which integrates the internet surfing habits of the user.
For the model for dynamically adjusting the elastic IP charging mode, which is provided by the invention, the individual internet surfing time period data can be trained independently to extract the characteristics of a single user to obtain the special charging model of the user, and the internet surfing time period data of a plurality of people can be trained in a fusion manner to extract the characteristics to train a charging model with universality; for example, for normal office workers, most of the workers use the flexible IP for continuous network access in the daytime, while a small part of the workers may have longer working time at night, so that the demand for accessing the network in the daytime is high, the bandwidth pressure borne by the cloud service provider is naturally high, the demand for accessing the network at night is low, and the bandwidth pressure borne by the cloud service provider is naturally low; two models can be respectively trained on the internet access period data of the two types of people, so that the network duration which will occur is predicted; when the network access time of the user is long and the demand is large and the bandwidth pressure borne by the cloud service provider is large, the charging unit price of the flexible IP can be improved, and the charging unit price of the flexible IP is properly reduced when the network access demand is not large;
in addition, some crowds do not always carry out network access, and the network access time length of the user about to happen can be accurately predicted after the past network access time period data of the users are analyzed; the charging unit price of the flexible IP is relatively reduced for the users which are likely to have long time, and the charging unit price of the flexible IP is relatively improved for the users which are likely to have short internet surfing time.
Example two:
an elastic IP billing system based on a deep neural network specifically comprises a data preprocessing module, a feature extraction module, a weight training module and a model prediction module:
a data preprocessing module: preprocessing the data of the user online time period;
a feature extraction module: utilizing a residual error network to perform automatic feature extraction on the preprocessed data;
the weight training module: the relevance of time before and after the characteristic data is trained by learning by utilizing a bidirectional long-time memory network model;
a model prediction module: re-pricing the elastic IP unit price by using a bidirectional long-time memory network model;
firstly, preprocessing data of a user internet surfing time period through a data preprocessing module, then automatically extracting characteristics of the preprocessed data through a deep neural network by using a characteristic extraction module, using the relevance of time before and after learning and training characteristic data of a bidirectional long-time memory network model by using a weight training module, and finally re-pricing the unit price of the elastic IP through a model prediction module by using the long-time memory network model, wherein the unit price of the elastic IP obtained at the moment is the unit price fusing the internet surfing habits of the user;
the invention carries out certain pre-processing on the previous internet surfing time interval data of the user, and predicts the possible internet surfing of the user frequently based on the deep neural network and the bidirectional long-time and short-time memory network model, thereby dynamically adjusting the charging unit price of the elastic IP. Compared with the original charging mode of fixed unit price for charging according to needs, the model for dynamically adjusting the charging mode of the flexible IP can provide more reasonable charging unit price for a user and can also provide a scheme for relieving bandwidth pressure for a cloud service provider;
further, the data preprocessing module specifically includes a data statistics module and a statistics conversion module:
a data statistics module: counting user internet traffic data at intervals and drawing a line graph;
a statistic conversion module: converting the line graph into a power graph through Fourier transformation;
collecting the internet surfing data of the user, firstly, counting the flow data used by the user every thirty minutes by using a data counting module, drawing the flow data of the user into a line graph with the horizontal axis as time and the vertical axis as flow usage, and observing the internet surfing habit of the user; then, the intuitive line graph is converted into a power graph through fast Fourier transform through a statistical conversion module, and the user characteristics are conveniently extracted by adopting a deep neural network model;
further, the feature extraction module adopts a residual error network with 50 layers to automatically extract features of the preprocessed data;
the residual error network model can extract deeper user characteristics in the aspect of characteristic extraction compared with other neural network models, so that the internet surfing time of a user can be predicted more conveniently;
for the power diagram obtained by the statistic conversion module, the feature extraction module adopts a residual error network model with 50 layers to automatically extract features of the preprocessed data, and users have different internet surfing durations at different times and are reflected in the residual error network model, namely different weights of each layer of the model. The user data is not simple discrete data, but data with time relevance before and after, so that the features extracted by the model are also related before and after;
still further, the model prediction module specifically includes a data prediction module and a data prediction module:
a data prediction module: removing an activation function in the long and short time memory network model to obtain internet surfing time prediction data;
a data prediction module: the long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data;
removing an activation function layer in the long and short term memory network model through a data prediction module to obtain the prediction of the internet surfing time; then, a data prediction module is used for re-pricing the unit price of the next elastic IP according to the internet surfing duration which is obtained by memorizing a network model at long and short times and is possible for a user to surf the internet, and the unit price of the elastic IP obtained at the moment is the unit price which integrates the internet surfing habit of the user;
for the model for dynamically adjusting the elastic IP charging mode, which is provided by the invention, the individual internet surfing time period data can be trained independently to extract the characteristics of a single user to obtain the special charging model of the user, and the internet surfing time period data of a plurality of people can be trained in a fusion manner to extract the characteristics to train a charging model with universality; for example, for normal office workers, most of the workers use the flexible IP for continuous network access in the daytime, while a small part of the workers may have longer working time at night, so that the demand for accessing the network in the daytime is high, the bandwidth pressure borne by the cloud service provider is naturally high, the demand for accessing the network at night is low, and the bandwidth pressure borne by the cloud service provider is naturally low; two models can be respectively trained on the internet access period data of the two types of people, so that the network duration which will occur is predicted; when the network access time of the user is long and the demand is large and the bandwidth pressure borne by the cloud service provider is large, the charging unit price of the flexible IP can be improved, and the charging unit price of the flexible IP is properly reduced when the network access demand is not large;
in addition, some crowds do not always carry out network access, and after the past internet surfing time period data of the users are analyzed, the impending internet surfing frequent of the users can be accurately predicted; the charging unit price of the flexible IP is relatively reduced for the users which are likely to have long time, and the charging unit price of the flexible IP is relatively improved for the users which are likely to have short internet surfing time.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. An elastic IP charging method based on a deep neural network is characterized by comprising the following specific steps:
s1, preprocessing the data of the user during the internet surfing period;
s2, utilizing a residual error network to automatically extract the characteristics of the preprocessed data;
s3, utilizing the relevance of time before and after the characteristic data is trained by learning of a bidirectional long-time memory network model;
s4 re-pricing the flexible IP unit price by using the bidirectional long-time memory network model.
2. The method as claimed in claim 1, wherein the step of S1 preprocessing the internet session data of the user comprises:
s101, counting user internet traffic data at intervals and drawing a line graph;
s102, converting the line graph into a power graph through Fourier transformation.
3. The method of claim 2, wherein the S2 adopts a residual error network of 50 layers to perform automatic feature extraction on the preprocessed data.
4. The method as claimed in claim 3, wherein the step of S4 for re-pricing the resilient IP unit price by using the bidirectional long-and-short memory network model comprises the following steps:
s401, removing an activation function in the bidirectional long-short time memory network model to obtain internet surfing time prediction data;
s402, the bidirectional long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data.
5. An elastic IP billing system based on a deep neural network is characterized by specifically comprising a data preprocessing module, a feature extraction module, a weight training module and a model prediction module:
a data preprocessing module: preprocessing the data of the user online time period;
a feature extraction module: utilizing a residual error network to perform automatic feature extraction on the preprocessed data;
the weight training module: the relevance of time before and after the characteristic data is trained by learning by utilizing a bidirectional long-time memory network model;
a model prediction module: and re-pricing the elastic IP unit price by utilizing a bidirectional long-time memory network model.
6. The system of claim 5, wherein the data preprocessing module comprises a data statistics module and a statistics transformation module:
a data statistics module: counting user internet traffic data at intervals and drawing a line graph;
a statistic conversion module: and converting the line graph into a power graph through Fourier transformation.
7. The system of claim 6, wherein the feature extraction module performs automatic feature extraction on the preprocessed data using a 50-layer residual network.
8. The system of claim 7, wherein the model prediction module comprises in particular a data prediction module and a data prediction module:
a data prediction module: removing an activation function in the bidirectional long-short time memory network model to obtain internet surfing time prediction data;
a data prediction module: and the bidirectional long-time memory network model continuously re-prices the elastic IP unit price according to the prediction data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011227646.5A CN112398663A (en) | 2020-11-06 | 2020-11-06 | Elastic IP charging method and system based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011227646.5A CN112398663A (en) | 2020-11-06 | 2020-11-06 | Elastic IP charging method and system based on deep neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112398663A true CN112398663A (en) | 2021-02-23 |
Family
ID=74598921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011227646.5A Pending CN112398663A (en) | 2020-11-06 | 2020-11-06 | Elastic IP charging method and system based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112398663A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101707788A (en) * | 2009-10-27 | 2010-05-12 | 北京邮电大学 | Differential pricing strategy based dynamic programming method of multilayer network services |
| CN102045681A (en) * | 2009-10-12 | 2011-05-04 | 钟巨航 | Method and device for network charging |
| CN102883294A (en) * | 2012-09-12 | 2013-01-16 | 西南交通大学 | Segmental time interval billing method relevant to user behaviors |
| CN107995007A (en) * | 2017-12-01 | 2018-05-04 | 中国联合网络通信集团有限公司 | Customer charge collocation method and customer charge configuration system |
| TW201914324A (en) * | 2017-08-28 | 2019-04-01 | 中華電信股份有限公司 | Machine learning based time-dependent smart data pricing structure |
| CN110111139A (en) * | 2019-04-23 | 2019-08-09 | 上海淇玥信息技术有限公司 | Behavior prediction model generation method, device, electronic equipment and readable medium |
| CN111221896A (en) * | 2018-11-27 | 2020-06-02 | 北京京东尚科信息技术有限公司 | User behavior prediction method and device, electronic equipment and storage medium |
-
2020
- 2020-11-06 CN CN202011227646.5A patent/CN112398663A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102045681A (en) * | 2009-10-12 | 2011-05-04 | 钟巨航 | Method and device for network charging |
| CN101707788A (en) * | 2009-10-27 | 2010-05-12 | 北京邮电大学 | Differential pricing strategy based dynamic programming method of multilayer network services |
| CN102883294A (en) * | 2012-09-12 | 2013-01-16 | 西南交通大学 | Segmental time interval billing method relevant to user behaviors |
| TW201914324A (en) * | 2017-08-28 | 2019-04-01 | 中華電信股份有限公司 | Machine learning based time-dependent smart data pricing structure |
| CN107995007A (en) * | 2017-12-01 | 2018-05-04 | 中国联合网络通信集团有限公司 | Customer charge collocation method and customer charge configuration system |
| CN111221896A (en) * | 2018-11-27 | 2020-06-02 | 北京京东尚科信息技术有限公司 | User behavior prediction method and device, electronic equipment and storage medium |
| CN110111139A (en) * | 2019-04-23 | 2019-08-09 | 上海淇玥信息技术有限公司 | Behavior prediction model generation method, device, electronic equipment and readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20140301218A1 (en) | Statistical analysis and prompting method and system for mobile terminal internet traffic | |
| CN102801792B (en) | Statistical-prediction-based automatic cloud CDN (Content Delivery Network) resource automatic deployment method | |
| CN103200046B (en) | The method and system of monitoring network element device performance | |
| CN104038392A (en) | Method for evaluating service quality of cloud computing resources | |
| CN102208991A (en) | Blog processing method, device and system | |
| CN111935025B (en) | Control method, device, equipment and medium for TCP transmission performance | |
| CN104468257A (en) | Cloud application availability prediction method and system based on mobile user time-space behaviors | |
| CN102520785A (en) | Energy consumption management method and system for cloud data center | |
| CN110213097A (en) | A kind of edge service supply optimization method based on Resource dynamic allocation | |
| CN109492863A (en) | The automatic generation method and device of financial document | |
| US20220245495A1 (en) | Information processing method and apparatus, and computer-readable storage medium | |
| CN104135384A (en) | Internet surfing right management system and method thereof | |
| CN115513984A (en) | Method and device for determining day-ahead charging and discharging power of energy storage system and storage medium | |
| CN113821421B (en) | Method, system, device and storage medium for collecting server performance data | |
| CN112398663A (en) | Elastic IP charging method and system based on deep neural network | |
| CN113079045B (en) | Bandwidth allocation method, device, server and storage medium | |
| CN112884532A (en) | Low-power consumption thing networking collection system | |
| CN103379168A (en) | Data center resource distribution management method and system | |
| CN105407383A (en) | Multi-version video-on-demand streaming media server cluster resource prediction method | |
| CN116074256A (en) | Content distribution network node bandwidth optimization method and device, electronic equipment and medium | |
| CN114510639A (en) | Smart city personal customization recommendation platform and method based on big data | |
| CN107493328B (en) | A kind of Cooperative caching method based on Fusion Features | |
| CN115396955A (en) | Resource allocation method and device based on deep reinforcement learning algorithm | |
| CN117294652A (en) | Flow control method, server, electronic device, and computer-readable storage medium | |
| CN111314124B (en) | Network problem analysis method, device, equipment and storage medium for high-speed rail network |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210223 |